But if you've heard someone use that Vegas line to describe their Vegas time, you know there is more to the story.

And so it is with equity trading. Whether you are the venue analyzing competitive intelligence, the sell-side optimizing trade strategies, the market-maker analyzing routing, or the buy-side seeking liquidity, knowing what happens off-exchange matters.

Like knowing you might wake up between Mike Tyson and a tiger in a Vegas hotel.

Off-exchange activity starts with Alternative Trading Systems (ATSs). In recent years new startups have entered the market (OneChronos, Purestream, and Blue Ocean). ATSs have merged (Luminex and LeveL). Big banks have launched ATSs (Citi-ONE). And brokers have consolidated ATS platforms (Weeden, Piper Jaffrey, and Sandler O'Neil into the concise, stealth ATS name “XE”).

You might wonder what allures startups, banks, and brokers to enter this incredibly competitive market to play a zero-sum game where the price of entry is SEC and FINRA regulation and compliance. Like high rollers to the lights of Vegas, venues of all shapes and sizes are attracted to the thirty to forty percent of U.S. equity volumes that trade off-exchange. We love analyzing the space because its intense competition and ever-changing market dynamics can spur trading innovation.

When one ATS wins market share another ATS loses market share. That’s if they’re lucky. It is more cut-throat than that because ATS off-exchange market share itself is not static. ATSs compete with over a hundred brokers with off-exchange single-dealer platforms (SDPs) and internal crossing networks. Some of them are large: Citadel, Jane Street, Robinhood. Some have both ATSs and SDPs: Goldman Sachs, Morgan Stanley. And together, the ATSs and brokers compete with a growing number of listed market exchanges almost wholly incentivized to reduce the amount of volume trading off-exchange.

The off-exchange game has a simple objective and complex methods. The objective is to match buyers and sellers. The methods include negotiating market price, volume, volatility, and market-impact. Integrating with other market participants and trading systems, often ‘frienemies’, adds more complexity. All of which must be navigated to optimize a matching objective measured in time slices of milliseconds1.

Want to launch an ATS?

We don’t. Though a few of our Focus Signal subscribers did, so rest assured you are in mad company here.

Subscribe now

Venue shape.

Point Focal’s off-exchange liquidity module helps market participants understand the space. And our new analytics helps one see the space differently.

A venue’s value is measured by how efficiently it connects participants. And this can be observed through volume and average trade size rankings, both proxies for market share. Depending on the design of a venue's matching logic, venues may rank higher on total volume or average trade size. Or their rankings may be a blended balance between the measures.

We're showing the shape of the venue landscape across these rankings. And we are shaping portfolios within these rankings. Shapes change across ATS and broker venues, and across market cap and sector. They morph further across price, volume, and volatility characteristics.

This is what it means to see a portfolio differently. Let's look at some shapes.

The following is a series of off-exchange venue and portfolio shapes. A new analytic lens. And for a certain kind of market practitioner, information value.

First a note on interpretation:

Scatterplots

• X & Y axes are average trade-size rank and total volume rank respectively

  • Both axes are inverted (wait, what?)

    • Up (volume) and to the right (trade-size) are highest ranks

• Markers are symbols, colored by market-cap, and sized by number of trades

  • Except the last chart in each series where markers are venues

• Markers are sized by trade count

• Vertical and horizontal lines are drawn at top-10 rank cutoffs

  • Markers above and right of the lines are top-10 in volume and trade-size

Vertical Bar Charts

• X-axis: volume and trade-size rank position

• Y-axis: symbol count in selected portfolio or index

The Startups

OneChronos and Purestream recent activity in the KRE Regional Banking ETF2.

OneChronos matches orders using periodic auctions with randomized start times. The auctions use mathematical optimization to seek optimal matches across all orders. Expressive bidding allows users to input specific execution instructions at the matching engine level precisely tailored to execution or investment strategy objectives3.

PureStream order types enable differentiation by separating the price and liquidity discovery process. Percentage rate based order types are used to match orders. The matched orders percentage rate is applied to each of the stocks subsequent trades reported to the “tape” to create a stream of trades4.

OneChronos and Purestream average volume and trade-size ranks across all ATS venues in recent KRE activity.

Both startup platforms are using unique, innovative matching strategies to capture market share. Going from zero volume to meaningful volume is a hard thing to do. This achievement alone means they are worthy of attention. Transforming standard off-exchange data structures to capture venue-specific strategies may improve our analytic lens. We consider this opportunity for further exploration.

Subscribe now

Two ATSs. One company.

LeveL ATS and Luminex recent activity in the S&P 500.

LeveL ATS is a continuous crossing US Equity Alternative Trading System (ATS), designed to transform dark pool executions with crossing innovation5.

Luminex is a completely anonymous, non-quoted trading venue that satisfies the liquidity needs of buy-side firms and prevents the main challenges of dark pools, such as intermediation, fragmentation, fall downs, information leakage, and concerns around ownership and control of trade data6.

LeveL and Luminex average volume and trade-size ranks across all ATS venues in recent S&P 500 activity.

The answer to why these platforms merged is explained by their shapes. Look again and note how their distribution shapes are mirror images of each other. In fact, if we were to fold the venues chart in half on its straight-line fit7, LeveL and Luminex might actually touch each other. Their shapes are complementary liquidity profiles. They each have what the other lacks. Symbiotic venues.

An Adverse Selection Protection Engine and the 3rd largest U.S. bank.

IntelligentCross and Citi-ONE recent activity in the Russell 3,000.

IntelligentCross is an ATS that uses AI to optimize price discovery. It matches orders near-continuously, within microseconds of arrival, to achieve maximum price stability after trades8.

Citi-ONE is a Citigroup Global Markets, Inc. (“CGMI”) ATS launched in 2022.

IntelligentCross and Citi-ONE average volume and trade-size ranks across all ATS venues in recent Russell 3,000 activity.

Citi has been through the ATS life-cycle before, from launch to exit. Their return suggests they have lessons learned and new strategic purpose. IntelligentCross is backed by Point 72 Ventures. Their Russell 3,000 shape, like balloons to a ceiling, reflects their continued growth.

Top non-ATS broker average volume and trade-size ranks in recent S&P 500 activity.

While this post is focused on the ATS segment of off-exchange activity, we can also see Citadel, Virtu, Jane Street, and Robinhood among a different venue segment, with different shapes.

What happens off-exchange can be observed. The players allocating resources to off-exchange activity is signal that there is signal off-exchange.

When market activity goes from opaque to transparent, information value and opportunity arise.

Viva las venues.

Pie charts usually suck. Humans aren’t designed to interpret relative sizes of angular shapes. 80% of pie charts should be bar charts. Most of the rest shouldn’t be charts at all.

It is a rare scenario when a pie chart is the best way to interpret data. And even then, the pie chart can usually be constructed in one’s mind, without financial technology.

Consider a pie chart representing the probability of a coin toss result. A circle split equally between heads and tails. Symmetry.

Now consider a pie chart representing what we know, “knowns” and what we do not know, “unknowns”. The unknown pie slice dwarfs the known pie slice. Asymmetry.

The giant unknown slice can itself be split into known-unknowns and unknown- unknowns. Things we know we don’t know. And things we don’t know we don’t know. Finally, a useful 3-slice pie chart. I won’t present it here because you can see it in your mind’s eye.

Known-unknowns and unknown-unknowns span a spectrum of blind spots from information disparity to information asymmetry. In asset management and in life, the greater the asymmetry, the greater the blind spot.

Blind spots are risk. And opportunity.

Subscribe now

Market Behavior

Knowns: volumes, prices, and related liquidity metrics.

Known-unknowns: market participant behavior and activity. Demand, supply, behavior, and divergence.

Unknown-unknown: amplitude scores.

Market behavior equilibrium is a measure of stock strength and weakness derived from market structure physics: regulations, incentives, participant behaviors, and activity. Amplitude scores measure a stock’s distance from equilibrium. Market behavior equilibrium information can be used in pre-trade preparation, through-trade execution, and post-trade performance reporting. Amplitude scores provide behavioral strength and weakness information that enhance program trading outcomes.

Monitoring equilibrium creates information value about price-setting behavior that affects volumes, prices, and volatility. Equilibrium can be observed as demand and supply rising, falling, diverging, and converging across market, index, portfolio, industry, and sector classifications, as well as by stock. It is a critically important dimension of trade flows.

In the visual below, amplitude scores identify the S&P 500’s extreme positive and negative demand supply divergences. Knowing market behavior characteristics across portfolios, sectors and securities highlights risky positions and enables managers to improve trading strategies to produce better economic outcomes.

Demand and supply trends signal equilibrium danger zones that further inform trade implementation. The likelihood of gains falls as demand rises above 6 on its 10-point scale. Similarly, the likelihood of losses falls as demand falls below 4.

High levels of supply over 50% signal bearish market participant behavior - an important input to trade strategy decisions.

A stock’s demand-supply distance to equilibrium can be observed and classified into market structure quadrants of strength, weakness, muted upside, and muted downside. The following analytic captures market structure equilibrium information value.

Market Behavior Demand Supply Quadrants

• Each marker is a symbol

  • Symbols are sized by intraday volatility

  • Symbols are colored by demand state

• Equilibrium: is where current demand and supply meet 30-day demand and supply

  • Y-axis: current demand minus 30-day demand

  • X-axis: current supply minus 30-day supply (inverted)

• Amplitude is distance from equilibrium

• Quadrants

  • Strength: high demand, low supply

  • Muted Upside: high demand, high supply

  • Muted Downside: low demand, low supply

  • Weakness: low demand, high supply

Market behavior information asymmetry.

Earnings

Knowns: sell-side earnings estimates, earnings reports, post-earnings performance.

Known-unknowns: alternative earnings estimates, revisions, and trends, statistical measures of earnings beat and miss materiality, and post-earnings drift performance attribution.

Unknown-unknown: post-earnings drift classification scores.

Our post-earnings drift quantitative model is built with earnings unknowns: alternative earnings estimates quantitatively superior to traditional estimates. Each earnings release is classified into one of ten deciles based on the statistical meaningfulness of the earnings report beat or miss.

In February, seven S&P 500 earnings reports were classified into the most extreme earnings surprise deciles, 1 and 10. The 2-day post-earnings drift performance difference between the deciles was 270 basis points (10th decile: +2.7%, 1st decile: 0.0%). The spread was 290 basis points over ten days (10th decile -1.3%, 1st decile: -4.2%).

More broadly, across nearly 1,500 companies for which we have alternative estimate data, our post-earnings drift model produced the following results.

• 439 earnings reports (10.5%) were classified in decile 10, with an average 20-day post-earnings-drift performance of 3.39%.

• 402 earnings releases (9.5%) were classified in decile 1, with an average 20-day post-earnings-drift performance of -0.27%.

• A difference of 366 basis points.

Post-earnings drift quantitative deciles and performance should not be unknown.

Earnings information asymmetry.

News Sentiment

Knowns: news stories and activity.

Known-unknowns: natural language processed (NLP) news sentiment scores and activity alerts.

Unknown-unknown: Portfolio performance of NLP classified sentiment scores.

Our news sentiment analytic combines NLP scored headlines and activity with traditional pricing and technical indicators. Combining AI level information with standard financial information is like adding the unknown to the known to create an unknown-unknown.

TDY is the top scored 30-day news sentiment stock from the S&P 500.

The 30-day average news sentiment score across the S&P 500 is 0.13 (symbol level sentiment oscillates between -1 and +1). The S&P 500 30-day performance is down 2.5%.

The S&P 500 top-10 news sentiment stocks have 30-day scores ranging from 0.29 to 0.37. Their 30-day average performance is 4.9%.

The S&P 500 bottom-10 news sentiment stocks have 30-day scores ranging from -0.11 to -0.24. Their 30-day average performance is -7.1%.

News sentiment information asymmetry.

For as long as markets have existed there has been information asymmetry. And as markets evolve so does its information. Our knowledge also evolves. Gradually, then suddenly. Small changes to information knowledge can have a large impact on risk management and performance.

A mouse in a maze seeks its known-unknown, the location of cheese. But a greater blind spot exists. Its unknown-unknown is the human measuring the mouse’s ability to find the cheese.

If the mouse saw its blind spot, it would probably be shocked.

Wouldn’t we be shocked to see ours?

It’s a bit like measuring stock performance from all time highs and lows. Possibly useful measures, but arbitrary nonetheless. Consider the duration of time a security spends at all time highs and lows. By their very nature, all time highs and lows are ephemeral. Yet we fixate on them.

Despite my ambivalence towards arbitrary standards, herein lies a year in review of my Substack, Focus Signal. It was actually launched in October, 2021. You’ll have to live with the fact that the year in review spans 15 months, which gives me an arbitrary sense of pleasure.

The point of Focus Signal is to document and share the building of Point Focal, a portfolio analytics business. The goal of portfolio analytics and fintech businesses is to reduce noise and increase signal. To that end 2022 was a year of progress.

Startup stopwatches count time to existential demise. So things never tend to move as quickly as you would like them to move. And the noise you suffer through is always more clear in hindsight. But if you attract enough signal you can keep going. And if you can keep going you can, well, keep going.

Signals, the Focus Signal micro-pod, reached twenty episodes in 2022. Looking back reminded me that nine episodes were recorded in 2021 and eleven were recorded in 2022. Already, not bounding this review to the calendar is helpful!

This spoken word micro-pod is an attempt to share the inner workings of building Point Focal. It has become a useful way to connect with you, the audience, and the feedback to the recordings has been wonderfully surprising.

As an introvert, there is something unnatural about communicating in this fashion. But I think this is also why I’m enjoying the medium. The fact that I can record the pod in isolated silence knowing it will reach you if, when, and where you want to be reached seems to bridge the space between introversion and extroversion.

Here are my favorite episodes:

11 Crossing the Chasm: Pod Sentiment, FinTwit Signal, and Trial Silence

Mostly because the part about Trial Silence epitomizes the defeatism inherent in startup rejection that must be overcome.

14 Zero to One to 1.1: Podcasting and How Music Production Emulates Product Development

Mostly because as an amateur musician I am fascinated by how Rick Rubin’s approach to music production can translate to business, financial technology, and analytics.

15 Walking Towards Uncertainty: This Pod, Explosive Growth, and Portfolio Manager Happiness

Mostly because Ma said it made her proud of me. What else can one say, really?

18 Work Life Imbalance: Appreciating the Stories of Others

Mostly because of an exchange with my wife, which went like this:

“Do you know Lex listens to your podcast?”

“Not really, no.”

“Well he does. And he asked me if I listen to it. Which of course I don’t. And then he said, ‘Well you should listen to it. And Mom, one episode was about ME.’”

The most popular episode:

12 Split Personality: Highs, Lows, and Nothing in Between

Maybe because the highs and lows of daily life are a universal experience.

And the longest episode (for market structure nerds):

20 Twenty!: 20, GTM, and STA Market Structure

Because it was the longest episode, 34 minutes.

My goal in 2023 with the Signals pod is twofold. Make the experience more comfortable. And publish about two episodes per month.

Making it more comfortable means reducing production friction. It’s the answer to the Tim Ferris question, “What would this look like if it were easy?” When I figure it out, I’ll let you know. If it’s more comfortable, it will become a better experience for both of us.

Twice a month seems challenging and possible. It’s a cadence that should be useful to you and achieve what one audience member, a music producer, advises in the studio: leave them wanting more.

What twisted minds would want more of this, I don’t know. But I embrace you. This approach also lets me continue to publish semi-spontaneously, which I believe is important. It’s true that Substacks grow faster when content is published on a routine schedule. But I prefer to publish when I have something worth publishing, rather than when an arbitrary period of time has elapsed.

It feels truer to me and better for you.

I’ve published six long-form Focus Signal posts. Mostly about the strategy and experience of building a portfolio analytics business. If you want to understand if you understand something, write about it. In a sense, these posts are a self-imposed challenge, shared publicly, to understand. And to error correct.

Reflecting on these posts, it’s the most recent three that I like most. This is encouraging because I’m hopeful the content is improving. I’m also fortunate to have some incredible data partners that make it fun to write about portfolio analytics.

Here are the last three posts:

ESG and Impact Analytics: The Skeptical and the Soluble

Util is our answer to the client and regulatory demands forcing asset managers to understand portfolio impact. One reason we love Util is because whether one loves or hates ESG, Util is the answer. This is becoming more clear as Tesla, born from one of ESG’s greatest critics, Elon Musk, is starting to talk about impact. From Tesla’s latest Impact Report:

"As the world needs to strive for a substantial positive impact, we won’t be referring to ESG. Instead, we’ll talk about Impact.”

“Many ESG ratings evaluate: ‘Does this ESG issue impact the profitability of the company?’ We need a system that evaluates: ‘Does the growth of this company have a positive impact on the world?’”

An Alt-Data Story Explainer: Market Structure Concept, Characters, and Plot

Stories are noise. Story explanations are signal. This was fun to write because it chronicles a fun partnership. One that has been an education and holds great promise. Because market structure sentiment implications on portfolio management are profound.

When 10% is Bigger than One: Leverage in Music Production and Product Development

An expansion on Signals episode 14: Zero to One to 1.1 where I tried to articulate why Rick Rubin’s approach to the creative process can be as useful in fintech production as it is in music production.

And the most popular:

The Secret of Free Data: How Little Insights Lead to Business Growth

Maybe because people like secrets. Especially when they’re hiding in plain sight. Or maybe because it’s a candid story of how Point Focal went from zero to one.

Our team often use these posts to answer questions from prospects and clients. How can ESG analytics be useful? Why should we care about market structure sentiment? How did you build off-exchange analytics?

Focus Signal has the answers.

These posts will come in the future as they have in the past. Semi-spontaneously, when they’re ready. But a general target is to produce more, higher quality long-form posts.

Subscribe now

Signal, launched in September, is the newest section of Focus Signal. These are shorter, analytic posts. They are the portfolio analytics being produced by Point Focal. The output of Focus Signal.

These pieces are meant to connect macro strategy and experience to micro insight. It’s likely to resonate with the practitioners in the audience - traders, portfolio managers, and research analysts. But also with individuals who have some aspect of these institutional roles within themselves.

I published five Signal posts, mostly related to new market structure sentiment analytics - how to interpret them (Sentiment Ebb and Short Volume Flow), what they can tell us (Short Volume and Market Behaviors), and how they can be applied to asset management (Informing Portfolio Turnover and Stacking Portfolio Analytics).

I am particularly excited about the Signal section of Focus Signal. The expanding scope and quality of our data and our improving technology stack offers new opportunity to create uniquely valuable content.

The Signal section can evolve in many ways. And I don’t want to try to define it’s end state. Part of the larger Focus Signal experiment is to write the story from inside the story. This is how Signal will be written.

On the look and feel of Focus Signal. Point Focal is fortunate to be working with a world class go-to-market team. We have an entirely new brand and visual identity coming soon. I’m looking forward to making the new design consistent across Point Focal and Focus Signal.

You will see new icons, branding, and colors in Focus Signal soon. It should be an upgrade given the current look and feel was produced by me. I hope you like it and come to recognize it.

Lastly, I thought it might be interesting for you to know alongside whom you are subscribed. Here is an anonymously classified summary of the Focus Signal audience:

• Family

• Friends

• Teammates

• Clients

• Trial Users

• Portfolio Managers

• Research Analysts

• Traders

• Partners

• Advisors

• Investors

• Stock Exchange Data Providers

• Alternative Trading System Operators

• SaaS Technology Providers

• Universities

• Competitors (Welcome!)

• Music Producers

• Future Point Focal Executives

• The Truly Anonymous

Thank you all for following along. I am grateful you care about this content and I appreciate you being part of the story. Wishing you the best in 2023.

- Thomas

Laws of physics determine planetary ebb and flow in the universe. And the wonder that unfolds is immune to our wishes. This whole spectacle in which we exist is unmoved by our desires.

Yet we have studied the fundamental mechanics of the universe deeply enough to understand some of its order. It was not a surprise when the winter solstice reversed the durations of days and nights on December 21st.

We do not need stories when we know the laws of physics.

Market behavior, like the universe, can be explained by stories or laws. In the case of the market, stories attract the most attention. They are born from clickbait, fueled by crowds, and validated by the great bias of hindsight. Our tendency to look back at events that were not predictable at the time and believe the outcome was easily predictable.

Market behaviors are governed by laws far less grand than universal physics. Yet they share an important trait: mathematical determinism. Meaningful explanations in the form of probabilistic analyses.

Observable for those who wish to study them.

Let’s turn our attention to the morning of December 21st as daily light began to outlast darkness. Across the over the prior month, short volume rose from 46% to 52%, a significant 13% increase across this broad market proxy. Meanwhile demand fell from 8 to under 5 on its 10-point scale.

Further, since 12/16, the percent of stocks with rising demand halved from 28% to 14%. While the percent of stocks with falling demand more than tripled from 9% to 29%. All the while, fast money behavioral flow grew from about a quarter to a half of all market liquidity. Algorithms amidst expirations.

And our supply breadth ratio reached 1.6, a level last observed in September. The ratio is calculated as S&P 500 constituents with rising supply divided by those with falling supply. The higher the ratio, the more willing market makers are to sell the broad market. On 12/21 over 300 S&P 500 stocks had rising supply. Less than 200 had falling supply. Growing weakness over declining strength.

Finally, over 50% of the broad market was positioned in the weakness quadrant of our market structure map. The weakness quadrant is down-and-to-the-left, where current demand is below 30-day demand and current supply is above 30-day supply.

Since integrating market structure behavior into Point Focal in Q1-2021, I cannot recall seeing this quadrant so crowded. The average S&P constituent was 20% below its 30-day demand trend and 7% above its 30-day supply trend.

Low demand and high supply is a bearish divergence. Crowded weakness.

This view of market behavior is not driven by human biased headlines. It is constructed with rules-based explanations derived from market laws.

Specifically, short volume activity made possible by the SEC’s REG SHO short sale exemption for market makers. Plus 605 execution quality, 606 order routing regulations, and related reports.

The REG SHO exemption exists to facilitate orderly markets. It exempts market makers from having to borrow stock to short stock. The law removes friction from the process of market makers becoming short.

605 and 606 regulations exist to facilitate fair markets. They introduce transparency into venue decision making behavior - market participant interactions.

For a more thorough examination of market structure laws, spurious headlines, and why all market participants are not the same, see An Alt-Data Story Explainer.

With the market up about 1.5% on 12/21 CPI news stories, a client informed of the above, wrote, “Pretty big sell off coming based on that analysis?”

I responded, “I prefer Tim's language, which is to say that given this weak market structure profile, stocks are statistically less likely to achieve or sustain gains relative to stronger profiles. But yes, until these deteriorating trends halt and reverse, we should consider this a cautious-to-bearish input to trade strategies. In practice, thinking probabilistically, today's strength is a selling opportunity until the setup changes.”

12/22 saw the market give back the entirety of 12/21 gains. But one shouldn’t judge decisions on outcomes when thinking probabilistically1. Decisions should be made and judged based on the quality of information available at the time of the decision.

Understanding the relationship between market participant laws and behaviors enables us to derive and observe new market information. Short activity across a class of market participants. And a market participant interaction network.

Additive, high-quality information helpful to market entry and exit decision making.

Market structure demand and supply.

Strength and weakness.

From individuals to markets to the universe, stories are told and laws are observed. In all their chaos and beauty, through processes as measurable as the length of light in our days.

“The atoms of our body, as well, flow in and away from us. We, like waves and like all objects, are a flux of events; we are processes, for a brief time monotonous”

Carlo Rovelli, Reality Is Not What It Seems

Stacking portfolio analytics is the art and science of performance.

Here is a stack of explanations.

Earnings results are a baseline layer of market intelligence. Deere & Co. (DE), beat earnings expectations Wednesday, 11/23. Knowing that DE beat Wall Street estimates is common knowledge. Layer 1.

Knowing that DE beat Estimize estimates, a quantitatively superior reflection of future earnings relative to Wall Street, is less common knowledge. Layer 2.

Knowing the historical 3-day performance spread between DE earnings beats and misses equals 385 basis points, and the 20-day performance spread equals 807 basis points is uncommon knowledge. Layer 3.

Knowing that DE EPS of $7.44 falls into a post-earnings drift quantitative model’s 10th decile is rare knowledge. The model classifies earnings releases by the statistical significance of upside beat and downside miss levels, relative to a universe of trailing 1-year earnings releases. When an S&P 500 constituent’s earnings release lands in the 10th decile of upside earnings surprises, the 20-day return is 4.25%. This is layer 4.

Layers 1 and 2 are interesting, descriptive analytics. Layers 3 and 4 are useful, prescriptive analytics. Because n the 20-day post-earnings period, buyers should tilt towards aggressive behavior and sellers should tilt towards passive behavior.

Portfolio analytic layers are additive quantitative evidence strengthening the probability that reweighting a portfolio towards DE will improve performance. And because post-earnings drift performance is relative, it is probabilistically prudent to likewise reweight portfolios away from companies landing in the 1st decile of downside earnings surprise misses.

Stacking portfolio analytics is stacking probabilities together to gain an edge.

Now, let’s stack market structure analytics on earnings analytics. Market structure sentiment is an entire category of uncommon knowledge.

Preceding the 11/23 earnings release, market structure demand for DE bottomed twice. Uncommon knowledge, but layer 1.

And short volume fell from 56% to 36%, a 35% drop in market-maker willingness to sell short DE. Layer 2.

In the post-earnings drift period, DE had rising demand and short volume 30% below its 30-day trend. Its rising demand and falling supply diverged sufficiently to land in the strength quadrant of our market structure distribution for the S&P 500. Layer 3.

Market structure sentiment is the best measure of price setting behavior. Market participants reinforced the statistical likelihood that DE would make gains through its earnings release.

When DE reports we stack 4 layers of market structure analytics onto 4 layers of earnings analytics. After earnings, the post-earnings drift model and market structure profile developments further stack probabilities in one’s favor.

Without stacking, we only know that DE beat estimates. By stacking, we know that DE beat estimates in a statistically meaningful way while possessing a strong market structure behavior profile.

Stacking portfolio analytics improves decisions that affect positions, weights, and aggressiveness. Entries and exits. Performance and risk management.

The final product of which is AUM. A non-trivial performance measure.

Portfolio analytic stacks move you from common knowledge to rare signal.

Investment managers (IMs) are fired and hired regularly. It’s why performance, marketing, and relationships matter. And not necessarily in that order. When large institutions like pension funds fire one IM and hire another IM, they need to transition their assets from one to the other.

From a legacy portfolio to a target portfolio.

To do so, they use some of the world’s largest asset service providers. Because raising cash by liquidating one portfolio while spending cash to invest in another portfolio is more challenging than it sounds. Consider transitioning several hundred million dollars across global markets spanning four asset classes. Now do it with the VIX over 30.

Not as simple as selling META to buy XOM in a Schwab account.

Enter transition managers (TMs): State Street, Macquarie, Northern Trust, Russell. They will transition assets from legacy to target portfolios in a market neutral fashion while minimizing market impact. And they’ll charge commissions or flat fees financial institutions are happy to pay.

Demand, the firing and hiring of IMs, is high. Supply, transition managers, is low. And the assets involved are large. Revenue is recurring because today’s hired IM is tomorrow’s fired IM.

Transition management is a small market, adjacent to asset management, with large network effects.

We’re building market structure sentiment analytics for transition managers. When we’re done, we will have a TM solution, and an analytic workflow that can be applied to any event-driven, episodic investment process, including program trading.

And we’ll have a market structure sentiment quantitative model.

We are modeling a baseline of selling a legacy portfolio and buying a target portfolio in a simple, market-neutral process. And we’re comparing the baseline to a model of selling the same legacy portfolio and buying the same target portfolio using market structure sentiment to inform position entries and exits and buying and selling aggressiveness and passiveness.

If market structure sentiment can meaningfully explain price setting behavior, our model will improve transition manager performance.

We envision a pre-transition report with the following portfolio information:

• Average market structure sentiment (demand) and short-volume (supply)

• Demand and supply distributions

• Average dominant behavior activity (Fast Trading, Active, Passive, Risk Management)

• Distribution of positive and negative demand / supply divergences

• Demand direction breadth percentages (topped, rising, declining, bottomed)

• Supply direction breadth percentages (rising, declining)

• Market behavior profiles percentages (big and stable, active divergence, momentum, short trades, liquid volatile divergence)

• Aggregations across market-cap, sector, and market behavior

Pre-transition reports will quantitatively demonstrate how the legacy and target portfolios differ across these market structure sentiment metrics. And they will provide a basic model of how using market structure sentiment to inform the liquidation and purchase of legacy and target portfolios can reduce market impact and improve performance.

A critical market structure sentiment metric is divergence. Whether securities have demand and supply (sentiment and short volume) moving together or apart. The pace at which divergence and convergence are occurring. And the extremeness of divergence.

For the sake of this Signals piece, we’ll use divergence to explain an initial market structure sentiment model2.

An Alt-Data Story Explainer, well, explains, why market structure sentiment matters3:

Beneath market activity is market structure. And beneath market structure is market structure data. 605 best execution, 606 order routing, and FINRA Reg SHO short volume. Market structure sentiment. Demand and supply. The data that explains the stories.

Subscribe now

Because we have 2 portfolios (legacy and target) and 2 profiles (positive demand / supply divergence and negative demand / supply divergence), there are 4 resulting market structure sentiment strategy profiles:

Let’s use a simple example of demand and supply divergences. Consider a legacy portfolio with average sentiment of 6 (on a 1-10 scale) and average short volume of 40%. And consider a target portfolio with average sentiment of 3 and average short volume of 60%.

• Legacy portfolio: demand = 6, supply = 40%

• Target portfolio: demand = 3, supply = 60%

Holding other factors constant4, relative to each other, the legacy portfolio has a more favorable market structure sentiment profile (high-demand, low-supply) compared to the target portfolio (low-demand, high-supply).

Therefore, on a relative basis, it is statistically more likely for the legacy portfolio to make gains during the transition event and more likely for the target portfolio to suffer losses. Since the legacy portfolio is being sold and the target portfolio is being bought, transition managers can reduce market impact and improve performance by behaving more passively than aggressively with respect to both portfolios during the event.

In this scenario, market structure sentiment can be used to model more passive liquidation behavior to capture gains by selling the legacy portfolio into market structure strength. Likewise, passive behavior will be modeled for purchases to reduce acquisition cost by buying the target portfolio into market structure weakness.

Note that if the target portfolio exhibited a positive market structure sentiment divergence (like the legacy portfolio), buying behavior would be modeled more aggressively, to reduce acquisition costs by avoiding paying higher prices (buying the target portfolio into market structure strength).

Every transition has a duration: the number of days during which the transition will occur. Let’s make this transition’s duration 5 days. The baseline model will sell the legacy portfolio and buy the target portfolio evenly over the 5-day period, with market-neutral sales and purchases occurring throughout the event. The market sentiment strategy (type 2 from the profile table above) will sell the legacy portfolio and buy the target portfolio more passively than the baseline model throughout the event.

The two models are presented below:

Models will be evaluated by analyzing the net dollar value amount of raising cash by selling the legacy portfolio and spending cash by buying the target portfolio. If the baseline model raises $297M over 5 days selling the legacy portfolio and spends $303M over 5 days buying the target portfolio, the cost of transitioning from one IM to another (before TM fees) is $6M.

If the market structure sentiment model raises $299M over 5 days selling the legacy portfolio and spends $301M over 5 days buying the target portfolio, the cost of the transition is $2M (plus TM fees).

This is our objective: a better economic outcome using market structure sentiment analytics.

 A post-transition report will include the following portfolio information:

• Market structure sentiment profiles and metrics throughout the event

• Liquidation and purchase amounts (cash raised and spent) throughout the event

• Baseline and market structure sentiment model comparative performance metrics

This is the beginning of a portfolio analytics model on which we will expand. We will add other metrics beyond divergence to the analysis. We will map the degree of divergence to the levels of liquidation and purchasing aggressiveness and passiveness. And we will assess market structure sentiment more frequently throughout the event.

We will also build a stand-alone market structure sentiment quantitative model that can be used by any asset manager. The model will compare benchmark performance of buying and holding a portfolio of securities to a strategy of using market structure sentiment analytics to reweight the portfolio.

We will share the analytics described herein as we produce them. I suspect they will create utility for investment processes we have yet to consider.

For now, we will pursue a small market.

ESG and impact are not the same.

Environment. Social. Governance.

How these three words relate to each other to form an investment framework is not obvious. That they have come to dominate critical areas of asset management is undeniable. And whether their impact on corporations and investors is positive or negative is debatable.

What is certain is there is value in understanding, measuring, and analyzing the ESG impact of a portfolio. This is true whether you are a schoolgirl, a corporate executive, or a skeptic of current things. It is especially true if you are a portfolio manager overseeing the assets of schoolgirls, corporate executives and skeptics.

Like most of life, ESG has become politicized. Its politicization is accelerating which introduces uncertainty and opportunity. One challenge gaining attention is the fiduciary conundrum for investment managers who must choose between optimizing asset allocation for ROI or for ESG impact. Since investing involves tradeoffs and opportunity costs one can also imagine a strategy optimizing for balance between ROI and ESG impact.

Each of these investment strategies demand meaningful ESG explanations that will not be found in corporate disclosures or multi-factor models scanning social media, internet-of-things (IOT), and military movement patterns. Meaningful explanations come from objective processes based on an understandable methodology.

Regardless of ESG philosophy and motivation, one needs unbiased, accurate measures of impact. These measures are not in bundled scores produced by aggregating the myriad combinatorial outcomes of every (E) x (S) x (G) factor into an inexplicable number. Rather, an appreciation for tradeoffs recognizing most investments are not absolutely ESG positive or negative demands a clearer methodology.

One benefit of Point Focal’s inclusion in the Sand Hill East portfolio of companies has been engaging with a company whose presence in the portfolio is born from its vision to deliver a novel ESG solution. Util analyzes 120 million peer reviewed texts with natural-language processing to extract relationships and map revenue between company products and the 17 United Nations Sustainable Development Goals (UN SDGs).

In doing so, Util addresses the shortcomings of legacy ESG scores: bundled aggregations, noisy source data, and unexplainable methods. And by flipping the ESG model to one that measures impact in the direction of companies to outcomes rather than externalities to holdings, Util’s framework can be used to produce the ESG analytics schoolchildren, corporate executives, and skeptics have long sought.

Greta, Larry, and Elon, take note.

UN Sustainable Development Goals are the nearest representation of a universal ESG benchmark. Born in 2015, the closest approximation to normalized, outcome-based objectives are becoming a must-have input into ESG investment processes. UN SDGs have a target due date of 2030, itself an explicit acknowledgment that ESG impact can be tracked and measured.

This is an inflection point for ESG analytic utility. Finally, and for the first time, client demand and regulatory requirements necessitating a new ESG understanding can be satisfied.

Tony Fadell, the product genius behind iPods, iPhones, and Nest Thermostats, explains in Build that the best products are painkillers, not vitamins1. You can skip a vitamin and not notice an effect. Skip a painkiller and you experience suffering.

Util ESG analytics are painkillers helping their most important stakeholders: clients and regulators.

As much fun as it is to write about ESG analytics, it is more fun to design ESG analytics. Whether in our product or in this Substack, it is the combination of visual data design and narratives that create insight.

Nike is a useful case study. Before assessing impact, let’ s define Nike:

Nike, Inc. is an American multinational corporation that is engaged in the design, development, manufacturing, and worldwide marketing and sales of footwear, apparel, equipment, accessories, and services.

There is a lot to unravel from this one sentence description. A U.S. based international conglomerate operating across an important set of functions, geographies, and impact areas. More than a swoosh.

Let’s begin with the overall Util Sustainable Development Goals score. The number 3. By itself, uninstructive. But by understanding the underlying metrics that produce the 3 within the context of a broader portfolio or benchmark, the number 3 becomes insightful.

Start with the premise that impact is not universal or absolute. It is about context, incentives, and goals. All of which are moving targets set by shareholders, asset managers, and investors.

What companies do is worth more of our attention than what companies say they do. And yet it is easy to become obsessed with company disclosures, press releases, and analyst opinions. It has been difficult to focus on company actions. Until we could see product revenue mapped to UN SDGs.

Until Util2.

Subscribe now

Here is the Nike swoosh laid across the United Nations Sustainable Development goals. The 17 UN SDGs are ordered along the Y axis. Percent of Nike revenue positively and negatively aligned to each SDG is presented on the X axis.

First impressions include realizing that impact is not absolute. Aggregated ESG scores require unbundling and Util’s Nike score is no exception. What is unusual about Util is its framework is designed to enable an objective unbundling. Nike’s revenue is positively aligned with 8 SDGs and negatively aligned with 6 SDGs. This moves us toward understanding and optimizing investment impact.

Nike’s most positive impact is on SDG 16, “Peace, Justice, and Strong Institutions” and SDG 8, “Decent Work and Economic Growth” where 28% and 25% of product revenue is positively aligned, respectively.

Nike’s negative impact is clustered around SDGs 12 through 15: “Responsible Consumption and Production”, “Climate Action”, “Life Below Water”, and “Life on Land”. One does not need to be a famous schoolgirl or an executive to appreciate that these SDGs reflect environmental impact.

Util provides a useful classification of UN SDGs that sits between overall scores and individual SDGs. Nike’s strongest positive impact SDGs fall into Social Progress and Economy & Infrastructure categories. Each of Nike’s most negative impact SDGs are in the Environment category.

Util UN SDG Categories

• Basic Needs
  SDGs 1, 2, 3, 6, 7 (No Poverty, Zero Hunger, Good Health and Wellbeing, Clean Water and Sanitation, and Affordable and Clean Energy

• Economy and Infrastructure
  SDGs 8, 9, 11, 17 (Decent Work and Economic Growth, Industry, Innovation and Infrastructure, Sustainable Cities and Communities, and Partnerships for the Goals)

• Social Progress
  SDGs 4, 5, 10, 16 (Quality Education, Gender Equality, Reduced Inequalities, and Peace and Strong Institutions)

• Environment
  SDGs 12, 13, 14, 15 (Responsible Consumption and Production, Climate Action, Life Below Water, Life on Land)

Analyzing company percent of revenue aligned to UN SDGs is a helpful way to compare relative impact. But context matters. Twenty-eight percent of Nike revenue is $12.6B. Comparatively, thirty-seven percent of Beyond Meat revenue is positively aligned to UN SDG 13, “Climate Action”. But thirty-seven percent of Beyond Meat revenue is $173M, or 1% of the comparable Nike impact.

Actions matter, but so too does scale.

Accurately assessing ESG investment impact requires understanding corporate revenue allocations to goals and the scale of that revenue. This is true for asset managers with portfolios spanning market caps from nano to mega and for those “simply” tracking the S&P 500.

Context matters from stock selection to rebalancing, and all the way through to quantifying impact for clients and regulators (not necessarily in that order).

Beyond Meat’s stated mission is aligned with UN SDG 13, “Climate Action”:

Producing a Beyond Burger uses significantly less water, land and energy – and it generates fewer Greenhouse Gas Emissions (GHGE) than a beef burger.

However, their negative product revenue aligned with SDG 12 “Responsible Consumption and Production” is more than 50% of their positive revenue aligned with “Climate Action”.

Corporate actions mean more than corporate communications.

Attribution can also cut both ways on a single SDG. 19% of Nike revenue positively impacts SDG 10 “Reduced Inequalities” while 10% of its product revenue negatively impacts the same goal.

Lack of impact is also worth considering when measuring ESGness. Nike’s impact spans thirteen of the seventeen UN SDGs. Meanwhile, Beyond Meat only impacts three SDGs without touching Economy and Infrastructure or Social Progress categories. Like fighter jets that never return from war, sometimes we need to analyze what we cannot see.

All this is to reiterate that impact is not absolute. Even in a 2-stock portfolio, each company and goal requires analysis that identifies tradeoffs and explains outcomes. A universal approach with clear attribution can measure investment, portfolio, and fund impact.

Corporate revenue allocations to goals are the foundation of impact knowledge. Util’s methodology accounts for corporate product lines enabling a deeper impact analysis. Companies and investors can isolate product contributions to impact. This is further unbundling.

Nike’s overall score of 3 is less informative than its scores across Basic Needs, Social Progress, Environment, and Economy & Infrastructure categories.

Likewise, the four categorical scores are less informative than the seventeen impact measures, positive and negative, across Sustainable Development Goals. By adding the product mix to the equation ESGness becomes more granular.

Nike’s seven product lines do not have equal impact. Two of them, Sporting and Athletic Goods, and Sporting Equipment, have entirely positive impact. The other five product lines have mixed impact with Apparel and Apparel & Accessories products accounting for the most negative impact.

This can be viewed in our product heat map where Nike’s negative impact on UN Environment SDGs 12 through 15 can be observed across its products.

Building analytics is often about navigating between macro and micro themes. Themes that should be integrated through data & analytics such that one helps explain the other. One of the most important analytic considerations is the three-word query, “compared to what?”

We have gone end-to-end in our assessment of company impact. But capital allocators must navigate both macro and micro themes to answer the “compared to what?” question. Here macro to micro means market to industry to sector to company.

Nike, as compared to the rest of the Consumer Goods sector, is more informative than Nike in isolation. In each of the eight UN SDG impact measures where Nike has positive impact, Nike outperforms the broad Consumer Goods sector. And with respect to negative impact, Nike beats its sector on three of six metrics.

Our company vs. sector bullet chart condenses comparative information into a single view. If you’re not familiar with the format, a brief explanation will help with interpretation.

Each horizontal “bullet” represents a single UN SDG. Let’s focus on SDG 8, Decent Work and Economic Growth. The dark-blue, horizontal bar informs us that Nike has positive (rightward) impact on SDG 8. Because the bar exceeds (is rightward of) the green vertical line reflecting the sector average, we also know Nike outperforms its sector on this metric.

From a prior view (and from hovering over the bar in our application), we know Nike’s percent of positive revenue impact for SDG 8 is 25%. And the positive revenue delta score of 6% for SDG 8 informs us that Nike is six percentage points better than the sector score of 19%.

The background color bands are red for negative revenue and blue for positive revenue. Shading cutoffs appear at 25% and 50%.

Nike impact is superior to its sector impact across 11 of 14 total metrics. And while in isolation Nike’s impact across the Environment category appears poor, its impact is better than the Consumer Goods sector impact on 3 of the 4 related SDGs.

Enabled with narrow and specific ESG analytics, investment managers can consider impact during ideation and portfolio construction. Companies can be assessed to determine if inclusion will strengthen or weaken portfolio impact.

Considering Nike’s weakest environmental impact on SDG 12, Responsible Consumption and Production, one could specifically filter for companies with positive impact, and scale, on the same SDG.

ESG criticism is escalating of late. And to be fair, much of the criticism is fair. Endless greenwashing and blindly sacrificing corporate goals for the next ESG current thing begs for skepticism.

But ESG is not going away. It is benefiting from its critics and being forced to mature. Separating impact from ESGness is like pulling signal from noise. It addresses the skepticism by moving the space closer to the truth.

These market forces are creating opportunity to thoughtfully apply impact awareness to investing. As asset managers and service providers embrace this development new investment processes will arise.

Here’s one: impact rebalancing.

To that end, let’s look at Nike’s impact profile within the S&P 500. By plotting overall score on the Y axis and then assessing various SDG category scores on the X axis we can understand where Nike sits within the benchmark.

Recall Nike’s overall score is 3 and since that metric remains static on the Y axis, in the below visualizations Nike remains in the same vertical space. However, as we assess S&P 500 impact on Social Progress and the Environment, we see Nike moves across horizontal space.

By highlighting Nike within the benchmark, we can observe absolute and relative impact. Because Nike’s position has moved from right to left, we know its Environment impact is worse than its Social Progress impact. But we can also observe relative impact. There are only a handful of companies that perform better than Nike on Social Progress SDGs (Quality Education, Gender Equality, Reduced Inequalities, and Peace and Strong Institutions). However, Nike has average impact on Environment SDGs (Responsible Consumption and Production, Climate Action, Life Below Water, Life on Land).

The distributions inform us that the entire S&P 500 impacts the environment more negatively than social progress. Both in number of companies and severity of impact.

Our next set of Util views extend this portfolio analysis. We’re making it simple to compare impact of any set of portfolios to themselves or to their benchmarks. And we are modeling how changes in composition will affect portfolio impact.

Util analytics are the painkiller for the world’s ESG challenges. They are a leveling mechanism that does more than align companies to goals. They align diverse stakeholders representing ESGs strongest supporters and fiercest critics.

On Util’s website, the acronym ESG appears only in the context of an award won: “ESG Research of the Year, North America3.” From the report:

“ESG funds have no significant additional positive impact compared to non-ESG funds.”

This is what it looks like to embrace ESG skepticism. But unlike most skeptics, Util has a solution. Objective, measurable data capable of answering the question, “compared to what?”

ESG redefined and spelled out: impact analytics.

Market structure sentiment is derived from 605 best execution and 606 order routing exchange flows coupled with FINRA short volume1 activity. Flows and activity are modeled in a demand supply framework. Active, Passive, Fast Trading, and Risk Management market behaviors and average price are inputs into demand sentiment scaled between 1 and 10. Short volume is supply measured as a percentage of total volume.

Demand Supply Quadrants

Our market structure sentiment demand supply quadrants plot the distribution of each stock’s current distance from demand and supply trends. Demand distance is current demand less 30-day demand. Inversely, supply distance is 30-day supply less current supply. These formulas reflect that high demand signals strength and high supply signals weakness.

In the visualization, note that Supply from Trend on the X axis is inverted which enables a simple heuristic when interpreting the scatter plot: up-and-to-the-right signals strength (high-demand, low-supply) and down-and-to-the-left signals weakness (low-demand, high-supply).

The strongest demand supply divergence for Banks is BK. Nevertheless, its current demand is still more than 20% under its 30-day demand. And all Banks are averaging demand sentiment 64% under their 30-day demand. Quadrants can be used to identify divergences.

The market structure sentiment demand supply profile of Mining, Oil, and Gas stocks has a very different distribution than Banks. Most Mining, Oil, and Gas stocks have demand above trend (Y axis) with their cycles topping and declining.

While BK has the strongest demand supply divergence profile (up-and-to-the-right) among Banks, its profile is weaker than the average demand of Mining, Oil, and Gas stocks. Understanding context strengthens market structure sentiment analysis. Context is relating demand and supply strength or weakness across stocks, sectors, portfolios, and benchmarks.

Following the 9/1/22 Bank bottoming cycle, the Bank ETF KRE rallied over 6% from 9/6 to 9/12. As observed below, this is consistent with the bottoming cycle and subsequent rally in the broad market. The Energy ETF XLE (representing Mining, Oil, and Gas stocks) rally over the same period was relatively weaker: 3.5%.

Distributions

Our market structure sentiment distributions display demand and supply histograms. In the left Demand Distribution histogram, most S&P 500 Banks have bottomed demand profiles (red) and all banks have demand under 5. In the right Supply Distribution histogram, most Banks have rising (green) rather than falling (yellow) short volume with peak supply concentration between 50 and 60% Furthermore, nearly 80% of banks have short volume supply over 50%. Thus, more than half of prior day (8/31) volume was manufactured stock produced by market makers.

The Mining, Oil, and Gas stocks have a very different profile. Most components have demand sentiment between 7 and 10 and short volume supply under 50%. As compared to Mining Oil and Gas stocks, Banks have relatively low demand and high supply. We also observe the demand cycle of Banks bottoming, while most Mining, Oil, and Gas demand is declining. Both sectors exhibit rising supply profiles.

Broad Market Structure Sentiment

To complete our understanding of market structure sentiment context we look at the broad market. The average demand and supply of the S&P 500 is a critical element of market structure sentiment. Let’s widen our lens from Banks and Mining, Oil, and Gas stocks to the entire S&P 500.

Now we can observe the strongest Banks and Mining, Oil, and Gas companies across the entire S&P 500. The Banks distribution observed above is more consistent with the broad market than the Mining, Oil, and Gas distribution. Like Banks, most of the S&P 500 has a bottoming demand cycle (red).

Bottoming cycles vary in duration which is why we also monitor demand direction breadth across the broad market. Here the bottom cycle peak preceded a 5% rally from 9/6 to 9/12.

This leads us to a macro view of broad market structure sentiment trends. Price is overlaid on demand sentiment above short volume. Note the short volume supply bottom coinciding with the August market top 8/16. This activity is followed by extreme levels of high supply in September which has muted rising demand. Persistent, elevated supply and the demand cycle moving from rising to declining (green to yellow) would be a negative market structure sentiment setup.

Summary

Market structure sentient is an underappreciated representation of demand and supply market dynamics that set prices. An understanding of how key investor behaviors and market maker short volume activity combine to influence prices can help asset managers and service providers optimize alpha generation strategies and risk management practices.

This analysis reviewed a non-exhaustive set of market structure sentiment elements.

Broad Market Structure Sentiment
Demand supply context of market structure health.

Divergence and Convergence
Understanding the relationship between demand and supply and how divergences can identify relative market structure strength and weakness.

Distributions
Interpreting the shape of sentiment demand and short volume supply.

Demand Cycles
Tracking demand breadth across topping, rising, declining, and bottoming cycles.

Relative Strength and Weakness
Applying context to analysis across stocks, sectors, portfolios, and benchmarks.

For the full story of market structure sentiment, see “An Alt-Data Story Explainer”.

It’s a story that explains other stories. And because nothing can be so deceptive as a good story, you should pay close attention. Indeed, truth can be stranger than headlines. Story elements allure, but the plot that matters lies beneath.

Never let facts get in the way of a good story, so the saying goes. But to avoid deception, never let stories omit explanation.

News network anchors and newspaper headlines continuously report stories about why markets are moving. The stories change as markets rise and fall. As markets revert, stories recycle. Finally, at the height of deception, stories that once explained a rising market, explain a falling market.

Here’s a sampling of headlines explaining different market moves:

“Markets rise on COVID hopes.” - Barron’s, 8/28/20
“Global markets fall on COVID-19 worries.” - ABC / Reuters, 10/28/20

“World stock markets rise on inflation data” - Reuters, 5/28/22
“Inflation angst drags S&P 500 into bear market” - Reuters 6/13/22

“Global stocks in steep weekly retreat after central bank rate rises” - FT, 6/10/22
“Stocks rise after the Fed takes historic action on inflation” - CNN, 6/15/22

Before the recent confluence of pandemic, inflation, and interest rate news cycles, media and newspapers substituted other topics to explain the same market movements. Russia and Ukraine, unemployment, and of course Elon Musk.

Pick a topic and match it to a market move.

“Twitter falls, wipes out all gains since Elon Musk disclosed his stake.” Mint - 5/16/22
“Twitter jumps after Musk increases commitment in takeover bid.” CNBC, 5/25/22

There is little debate about whether inflation, interest rates and Elon affect markets. But using headlines to explain markets is suspect when they are indifferent to market direction.

When every topic can be used to explain every market movement, they do not explain any market movement. So too with models, from The Color of Greed and Fear:

“If a sample of Apple tweets could be converted into emotions that produced meaningful information through an off-the-R-Studio-shelf sentiment model, it could explain anything. And a model that explains anything explains nothing.”

When themes and characters change while stories remain the same, apply scrutiny. Maybe there is a better explanation for the direction of $185 billion in daily transactions between buyers and sellers that make up the $49 trillion U.S. equity market. Maybe there is a more credible story.

A story where themes are continuous auction markets and best execution regulatory requirements. Characters are greed and fear, demand and supply. And climaxes and resolutions are a dance between performance and risk management.

This story is about seeking an objective explanation for market movements independent of news cycles. As with stories, what matters with analytics lies beneath.

Insight and signals rest atop explanatory data. Selecting data is like choosing theme and plot. The wrong data creates boring stories at best and untrue stories at worst. Useless analytics. The right data creates characters, tension, climax, and resolution that form meaningful stories. Useful analytics.

Equities are transferred between market participants operating in a complex system of networks. Exchanges, alternative trading systems, electronic communication networks, crossing engines, and voice brokers are network nodes connecting buyers and sellers. They use primitive and cutting edge technology in an intensely regulated environment where users are willing to tunnel cable across the country to reduce communication latency measured in milliseconds1.

Lying beneath this system are two data sets known by 3-digit codes, 605 and 6062. The codes and their underlying data are the product of the Securities and Exchange Commission (SEC) Disclosure of Order Execution and Routing Practices rule. From the SEC:

The Securities and Exchange Commission is adopting two rules to improve public disclosure of order execution and routing practices. Under Rule 11Ac1-5, market centers that trade national market system securities will be required to make available to the public monthly electronic reports that include uniform statistical measures of execution quality. Under Rule 11Ac1-6, broker-dealers that route customer orders in equity and option securities will be required to make publicly available quarterly reports that, among other things, identify the venues to which customer orders are routed for execution.

Also beneath the system is FINRA Reg SHO short sale volume data. From FINRA:

Pursuant to a U.S. Securities and Exchange Commission request, FINRA makes short sale trade data publicly available for off-exchange (i.e., OTC) trades in exchange-listed securities reported to a FINRA Trade Reporting Facility (TRF) or the Alternative Display Facility (ADF), as well as for trades in securities traded over-the-counter and reported to FINRA’s Over the Counter Reporting Facility (ORF).

Beneath market activity is market structure. And beneath market structure is market structure data. 605 best execution, 606 order routing, and FINRA Reg SHO short volume. Market structure sentiment. Demand and supply.

The data that explains the stories.

Analogies can be imperfect and helpful. I will attempt to use one here. Readers familiar with Michael Lewis’ Moneyball3 will have an interpretation edge. Moneyball is the baseball economics story of how analyzing fundamental player statistics changed the game. And for better or worse, the Venn diagram crossover of equities and baseball is likely material.

Ok, onto the analogy, here goes:

The old school baseball scouts in Moneyball were observing baseball players. Watching their swing, watching their bat-speed, watching them run, watching them respond to interview questions after the game. These observations make for great stories that can be used to buy and sell players.

But these stories are very different from the story told by on-base-percentage4 (OBP). OBP is universal, it is ingrained in the fundamental mechanics of baseball and statistically meaningful in determining player productivity.

Scouts watching bat-speed is not statistically meaningful. It is a bit like watching commentators reading headlines.

Billy Beane, General Manager of the Oakland A’s, combined Moneyball statistical principles with a budget one third the size of competitors to field a winning team. The A’s reached the playoffs in four consecutive years (2000 - 2003) and became the first American League team in over 100 years to win 20 consecutive games.

Data signal replaced scouting noise.

Trying to explain market behavior without market structure sentiment is like trying to evaluate a baseball team without OBP information. Likewise, buying and selling securities without understanding market structure sentiment is like buying and selling baseball players without understanding OBP.

To complete the imperfect analogy, baseball GMs using OBP to optimize team selection is like asset managers using market structure sentiment to optimize portfolio management. OBP informing rosters is like market structure sentiment informing stock selection, position size, and entry and exit decisions.

Market structure sentiment signal can replace headline noise.

If you want to tell useful stories with data, you need the right data. And if your goal is to tell a meaningful, explanatory story about market movement, that data must include the underpinnings of what makes our markets function. Data representing market participants where actors behave according to regulatory requirements.

This fundamental data may sound boring because it lacks the intrigue of alarmist headlines. No one on CNBC is pining on about divergences in 605, 606, or Reg SHO data that might explain, in terms of central tendencies, the probabilistic likelihood of market strength or weakness.

Well, actually, one person is. But they air him at 5am when most of CNBC’s audience is asleep. He’s the eccentric bloke we invited to the BSTA conference last year, Tim Quast. Here’s Tim explaining market structure on CNBC:

At the 3:40 mark, Brian Sullivan from CNBC makes a candid statement about the forces that drive current-thing headline stories:

“You kind of just described what is often the frustrating part of our job, which is trying to find a headline. We can’t come on the air and say, ‘I don’t know.’”

Brian, we have the solution.

I explained our partnership with Tim in the afterword of The Secret of Free Data:

Together we’re enriching market structure sentiment analytics. And we’re designing a more complete investor relations dashboard combining market structure with earnings, ESG, and news information.

Like our BSTA panel, our goal is to educate and help those who can benefit from a truer understanding of how market structure explains market behavior

Transforming best execution and short volume data into a demand supply framework of market behavior is as exciting as market structure gets. And its implications on investment processes is an unbelievable portfolio analytics story.

Subscribe now

Stories require balance. Good and evil. Love and hate. War and peace.

Our market structure story is balanced by demand and supply. Proxies for best execution and short volume routing information. It’s market rules and regulations all the way down.

Stories flow from characters to theme within a conceptual arc. Plot ties it all together. Our plot is market structure sentiment. And our story flows from SEC filings to market participant activity to portfolio optimization. The conceptual arc is made up of filings that create rules that govern behavior that produce orders that explain activity.

These are the fundamental mechanics of the equity market and statistically meaningful in explaining activity. This is the market structure equivalent of on-base-percentage.

So how does one use market structure sentiment?

Instead of watching headlines and talking heads, analyze market structure demand and supply trends. Understand whether demand and supply are rising or falling. Interpret their relationship to one another and their level of divergence. Is demand rising while supply falls? Is the inverse true? Or are demand and supply moving in sync?

Analyze stocks, sectors, and broad market sentiment. Compare the relative strength of portfolios to each other and to the broad market. Use demand and supply flow indicators to understand what percent of portfolio and market sentiment is rising or falling.

Use market structure information to optimize portfolio turnover. Use it to inform position sizing, portfolio tilting, and trading strategy aggressiveness. Use it to deliver new and meaningful insight to clients.

You won’t find these characters in the headlines. And you won’t be able to benefit from understanding their influence if you are watching the wrong stories. Do not underestimate the value of allocating attention to the right story.

Let’s look at the tension in our story. From our plot told through regulatory data (the anti-headline), we come to market structure behavior classifications. Our story’s real heroes and villains.

• Fast trading

• Active

• Passive

• Risk management

These characters lie beneath media hyperbole, spurious headlines, and bloviating pontification. Like scouts listening to a ball player in a post-game interview, listening to headlines misses what actually explains the market.

These characters are in our story too. Their behavior across our data-prose can be identified through analysis. They can be grouped into market structure profiles that rise and fall in power. And they produce demand and supply flows that connect our plot and theme to the concept of market structure sentiment.

Now one can act based on an understanding of how these characters explain stocks, portfolios, and markets.

This is Moneyball for market structure.

Let’s explore some scenes. Portfolios are typically groups of stocks classified by industry, sector, or investment style. These groups are merely supporting characters, background extras, and in video game parlance, NPCs (Non-Playable Characters).

Static classifications make for static analysis.

Classifying portfolios by market structure sentiment groups stocks by market behavior characteristics. These groups are alive and change over time.

Dynamic classifications enable dynamic analysis.

Transforming 605 best execution, 606 order routing, and Reg SHO short volume information into demand and supply is like transforming Brad Pitt into Billy Beane to explain Moneyball.

Demand sentiment is modeled on a scale of 1 to 10, giving our character an emotional range to explore. Short volume supply is measured in percentage terms, exposing how short, or anguished, is our character.

We’ve unveiled the axes around which the market revolves: demand, supply, activity, and price. Headlines and current-thing stories have been stripped away. If markets care about interest rates, or unemployment numbers, or fed minute commentary, demand and supply will tell us. It will be in the lines of our characters.

Likewise, in times when the market does not care about interest rates, or unemployment, or the fed, our characters will tell us with different lines that diverge from current-thing headlines.

Once we understand demand and supply are driving the story, we must only pay attention to what they are telling us.

Jim Cramer, Bob Pisani, Maria Bartiromo, and your favorite Bloomberg analysts are talking about headlines that distract from market explanations.

If you take nothing else away from this story about stories, let this be the point that resonates. Almost universally, on a scale so grand it is taken for granted, markets are being explained by distractions.

Character misdirection in a great story creates tension and suspense before being resolved. Like when the audience learns the secret to The 6th Sense. But market headline misdirections snowball from one media outlet to another without resolution.

They move the narrative from one current-thing to the next current-thing until a real story like the Big Short explains what happened. By which time everyone forgets all the current-thing headlines that were misused to explain the market.

Demand and supply characters remain unexamined while Michael Lewis, who authored Moneyball and The Big Short, pieces together the story that sat atop the data of prior markets.

Meanwhile, our characters remain active, underlying current market conditions that will lead to the next market events. And because they are misunderstood, or worse, ignored to focus on more tantalizing current-thing headlines, the cycle repeats. Not always with the magnitude of the Big Short, but over and over again, across market moves big and small.

Such is the allure of current-thing headlines. Because they are current and time is ephemeral, they are always here. Always now. And always fighting for our attention. But they are a distraction from the behaviors that drive the market.

These are the scenes we live in. And as we create buy and sell orders, we become part of the story. We may as well study the script.

Let’s return to our characters.

Active and passive money, fast trading, and risk management. These are the players who leave traces on 605, 606, and Reg SHO data.

Volume is a single number representing shares traded. And activity alone is meaningful. A great amount of time and analysis is focused on volume statistics. Up-volume, down-volume, increasing or decreasing volume, on-balance volume, accumulation distribution, money-flow, and of course, the great VWAP, volume-weighted average price.

All these volume measures fail to account for the players. They disregard the characters driving the plot. It’s like analyzing a story’s prose without knowing who is speaking. Surface level market activity is an abstraction of market participant activity. It’s the type of analysis that can lead to spurious conclusions.

Look at Coinbase. Market structure sentiment demand was on the floor for weeks amid passive and fast trading activity. Then there was a surge of volume driven by risk management - money tied to options, futures, and other derivatives that warp prices.

After the market closed on 5/11 Sam Bankman Fried (SBF), former Jane Street trader and current billionaire owner of Cryptocurrency Exchange FTX, announced a stake in Robinhood. Guess what the headlines reported? Robinhood - and its peers like Coinbase - traded higher on news that SBF purchased 5% of HOOD.

But something more interesting happened prior to and underneath the headlines. COINs market structure sentiment profile changed. On 5/11 short volume supply peaked at a lofty 68%. And total volume capitulated with a surge in risk management activity. Over the next several weeks demand sentiment went from 1 to 10 and supply fell 15%.

From 5/11 to 5/31 COIN appreciated 45%, trading from 53 to 78.

By transforming market structure activity into basic economics5, we can understand the characters driving market tension and resolution. This is what lies beneath the headlines. Current things sit atop real things.

Maybe you don’t believe it. I said you wouldn’t believe it. How could an entire industry of market experts overlook the real story of market activity? Maybe the same way experts overlook basic fundamentals underlying health, politic, and violence.

One must be careful trusting experts.

Our market structure behavioral groups produce a social experiment. They align in regimes, large and small, and their movement over time influence capital markets.

These regimes are the war armies, political movements, and religious classes of the market. And just as one must understand war, politic, and religion to explain the world, one must understand market regimes to explain the market.

Breadth and flow are regime characteristics. Breadth measures the number of components rising and falling. Flow captures breadth movement across regimes.

Our regime market structure classifications include demand and supply direction, topping and bottoming action, market behaviors, and demand supply quadrants. Think of the quadrants as typical 2-by-2 consulting diagrams, except useful. Here are the quadrants:

• Up and to the right: high-demand, low-supply (bullish)

• Up and to the left: high-demand, high-supply (neutral)

• Down and to the right: low-demand, low-supply (neutral)

• Down and to the left: low-demand, high-supply (bearish)

Individual stocks can be tracked across quadrants over time and compared to portfolio distributions. Two quadrants expose market structure behavior where demand and supply diverge. And two quadrants capture demand and supply moving together - what Tim Quast refers to as muted upside and muted downside.

Great stories conjure great images in our imaginations. Artists turn the images into paintings, film, video games, and more. We turn this story into images too. Demand and supply. Greed and fear. Explanations.

Nobody at CNBC, the Wall Street Journal, or in the Twitterverse are looking at these pictures to help explain markets. Because the images don’t match current-thing headline stories. The experts are looking at other pictures.

If every picture tells a story, and if stories are explanations, one should be discerning about to which pictures they allocate attention. Market structure pictures are images of significance orthogonal6 to current thing headlines.

The position that market structure sentiment matters more than headlines is a relative claim. The corollary is not that headlines are insignificant.

Rather, the position is a claim that there is value in understanding headlines sit atop first order market structure principles. Fundamental characters beneath the stories.

Think of this story I said you wouldn’t believe. Imagine if demand and supply characters modeled from the very rules that govern market behavior were real and if understanding them could help explain performance and risk. Consider the implications on assessing portfolio health, position size, and entry and exit points.

Awareness is alpha.

Despite our analogy’s imperfection, Moneyball’s subtitle could also be our story’s subtitle. “The Art of Winning an Unfair Game.”

What lies beneath matters.

“Make your favorite things… you be the audience. Make the thing for you, the audience.” – Rick Rubin on the Lex Fridman Podcast, 4/10/221

I listened to Rick Rubin listen to Lex Fridman. Rick is a music producer. Lex is a podcaster. Both men are considered by many to be among the greatest at doing what they do. As an amateur musician and an amateur podcaster, I listened with interest.

Two things struck me.

1. Rick did most of the listening despite being the interviewee. Notable in isolation, but exceptional given the interviewer, Lex, is a master of patience who once waited 21 seconds in silence for Elon Musk to answer a question2. Twenty-one interview seconds is a long time to wait.

2. When he spoke, Rick’s descriptions of producing music could have been describing producing financial technology.

Rick starts with something he did not know, “I didn’t know this was a job.”

Maybe the greatest work-life balance statement ever. Building Point Focal is not what I set out to do with my life. It is where my path of learnings and experiences took me. It is not a job I could have envisioned earlier in life. And it is not something that feels like a job.

If you do something that does not feel like a job, you are among the lucky. And if what you used to do felt like a job, you likely possess great appreciation for your luck. Rick and Lex share a profound appreciation for being able to do what they want to do. It is a state worth understanding, identifying, and pursuing.

Creativity takes many forms. Rick transforms sonic space and time to enrich songs. Lex transforms language to enrich understanding. Point Focal transforms words and numbers to enrich decisions.

In the end it’s all art.

A music artist creates something from nothing. A song from silence. Rick produces the song, turning something into something better.

The band is zero to one. Rick is 1 to 1.1.

This is the transition I have been reflecting on at Point Focal. Our path from zero through one to 1.1.

Going from zero to one is the greatest thing a company does. Even if going from 1 to 1.1 advances the company further, zero to one is the production of something from nothing. It is the genesis of a business in two senses. First, a product goes from not existing to existing. And second, the value generated by the product is shared between a customer and the business.

Zero to one is a business. Business creation is written in letters, zero to one, as taught by Peter Thiel3. Product updates are written numerically, 1 to 1.1. Zero to one wins one customer. 1 to 1.1 can win many customers.

A music artist cannot do better than reaching Rick Rubin as its one listener. When Rick helps the artist move from 1 to 1.1, they reach many listeners. The same is true of financial technology. The first user matters and can help a business reach many users.

Going from 1 to 1.1 can have greater impact while also being less difficult than going from zero to one. There is less distance to travel when something already exists.

Rick, on getting to a song’s end state: “I don’t think we need to know. We need to see what works… It doesn’t really matter why as long as it does the thing that you want it to do.”

Here Rick is unknowingly channeling a great product manager in front of an Agile board. He is explaining product wisdom to the business owner and the technology team. It doesn’t matter why as long as it does the thing that you want it to do.

Use the right combination of resources and tools for the right functions. It does not matter why, or how. It only matters that the song or the product does the thing you want it to do. The music producer wants the song to move the listener. The software designer wants the product to move the user. The movement is the thing.

The last three Focus Signal posts explain Point Focal’s zero to one journey. From nothing to alternative data in an analytics stack. A product. And a customer with whom we shared the value generated.

This Focus Signal post explores Point Focal’s transition from zero to one to 1.1. As I wrote in Pub Sub Fit, I believe there is value in telling the story from within the story.

Zero to one is brute force. It is will. And it is creation.

1 to 1.1 is momentum. It is leverage. And it is scale.

Zero to one is longer, but 1 to 1.1. is still long. Getting to one does not need to be pretty. You can live with a lot of inefficiencies if it results in a product and a customer. Reid Hoffman said, “If you’re not embarrassed by your first version of your product, you’ve launched too late.”4 Going from 1 to 1.1 is the process of fixing many of your first version’s failings.

The objective of 1.1 is to retain your first customer while making it possible to acquire many more customers. This means rebuilding your product in the form of its promise. Zero to one is when the promise is created. Early adopters understand and believe in the promise. And early on, they will tolerate the fact that much of the promise is unfulfilled. Getting to 1.1 means closing the promise gap.

It means leaving a lot of what was created behind.

When developing a product, it is liberating to appreciate that everything you are building you will eventually destroy. The destruction is a byproduct of rebuilding your product. Zero to one is incredible because you’ve produced a product and an audience. But the learning experience it leaves in its wake is humbling. Point Focal went from zero to one despite all the mistakes we made. Surviving our mistakes is a theme that opens Solving an Earnings Challenge:

Time kills startups. Solving problems creates businesses. Startups are in a race to solve problems before running out of time.

Subscribe now

Rick on risk-taking, “Oh this doesn’t sound like it’s gonna do as well as your last thing, that’s not the point.”

Keep playing. Keep producing. And keep building.

Every band and every business need to avoid becoming a one-hit wonder. This means growing beyond your original vision. It means doing what you do in a different way that attracts a larger audience. It means pleasantly surprising your original audience by exploring new territory, living uncomfortably, and taking risk.

Going from 1 to 1.1 Is not only additive, it is also subtractive. Rick spoke about identifying song elements that upon removal helped reveal the music’s fundamental essence. Improving a product requires the same elimination of anything not helping the audience.

Removing noise clarifies signal.

Undoing mistakes is creative destruction. It allows you to do more with less to improve your product. From a stronger baseline, adding to your product becomes non-linear. 1 + 1 > 2.

It is beautifully rewarding to optimize a product through iteration to make version 1.1 far better than version 1. Music producers are optimizers. In financial technology, less latency and higher performance is the product equivalent of a band’s improved timing.

Our version 1.1 doubles Point Focal data sets. And it doubles them in a completely re-architected environment. R scripts previously run in BI tools are now python lambdas running in AWS. Flat data files have become DynamoDB and MySQL databases. Portfolio updates have gone from 24-hours to real-time. User on-boarding has gone from people and days to technology and minutes. Production reporting scale has increased by orders of magnitude. My sleep schedule has gone from untenable to livable.

1.1 is much further in product distance from 1 than it is in numerical distance. It is the payoff of going from zero to one and avoiding one-hit wonder stagnancy.

Rick on suffering, “The same things that make an artist uncomfortable other people don’t feel at all.”

If all creativity is art, then product developers and entrepreneurs are artists. And in a small, humble way I can empathize with the suffering Rick describes. Living in an environment that makes one uniquely uncomfortable can be a great motivator for change. Envisioning the environment transformed can lead one to create an album or build a product.

Becoming comfortable being uncomfortable can result in sustained, high-quality production. The discomfort can drive one to continuously create the next new thing. Change becomes normal. New space is explored. Boundaries are stretched. Constraints are re-imagined. Creative destruction repeatedly displaces every new thing with the next new thing.

Prince, Madonna, Coldplay.

Apple, Amazon, Facebook.

Artists and businesses that went from zero to one and from 1 to 1.1.

Juxtaposing music production with product design is not analogy or metaphor. Creativity and suffering are universal. Art is art.

When a musician loses touch with the things that make an artist uncomfortable, creative change ends. And the audience knows. The next album is not a new thing. It sounds like the last album. The old thing.

When an entrepreneur loses touch with the things that make an artist uncomfortable, creative change ends, the audience knows, the next release is not a new thing. It feels like the last release. The old thing.

The uncomfortable nature of the creative process is the suffering and the reward.

Rick, on understanding our body’s head and heart, “I believe we know close to nothing about anything.”

Rick is an absolute master of his craft. Endlessly sought out by the world’s greatest artists, his work can be found at the Grammy’s, in the Rock and Roll Hall of Fame, and on Hollywood Boulevard5. His work is probably on your phone right now. And yet he starts from a place of spectacular humility.

Accepting our lack of omnipotence does not mean we must lack conviction. It leads to an honest, critical assessment of when one decides to act with conviction. It means listening to designers, engineers, and producers to understand their ambition. And connecting that ambition with an audience.

Navigating the 1 to 1.1 space is very different from navigating the zero to one space. It requires different human, technical, and capital resources. To reach 1.1, artists must adapt or change or be replaced. The next new thing demands it.

Rick is the momentum, the leverage, and the scale that lives between 1 and 1.1.

Any success Point Focal has had is the result of beginning with humility, listening to our audience, and thinking critically about when to act with conviction.

Appreciating the unknown is a healthy disposition. It is conducive to skepticism and unconventional methods. Whether producing music or financial technology, spectacular humility is a good place to start.

Elicited by Lex’s curiosity, Rick’s words describing what makes the human state real may capture the space between 1 and 1.1 best. “A place where you can be your most vulnerable without questioning it.”

If learning to live with discomfort is going from zero to one, balancing the highs and lows of suffering and reward is going from 1 to 1.1.

We have listened to and learned from everyone who has said yes and no to Point Focal. We have listened to and learned from advisors, investors, partners, and supporters. We try to balance this input the way Rick balances sound to optimal levels and frequencies. Our inputs are our momentum, scale, and leverage.

1.1 is a different journey than zero to one. There is more to lose. Becoming comfortable being uncomfortable helps you suffer through continuous next new things.

Listening to Lex, Rick described the 1 to 1.1 path from concept to completion: believing we know close to nothing about anything, understanding the magic is making it do the thing you want it to do, not worrying if it may not be as good as your last thing, and feeling uncomfortable.

This belief, this understanding, this not worrying, and this feeling are what has moved Point Focal from 1 to 1.1. It has produced bigger vision, new capital and resources, stronger partnerships, new clients, and greater distribution.

Momentum, scale, and leverage.

Rick’s description of producing great art completed the connection between music and tech production, “Make your favorite things… you be the audience. Make the thing for you, the audience.”

The analytics we’re building at Point Focal are the analytics I wanted when I was trading. They create the insight I wanted when I was an analyst. And they use the analytics stack I wanted when I was designing financial technology.

We are making my favorite things. And we are our own audience. By sharing our favorite things, we have found clients, advisers, investors, and partners. And they have become part of our favorite things as we move from 1 to 1.1.

May you be so lucky as to be making your favorite things.

“You have that data on a stock-by-stock basis?” he asked.

“Yes.”

He paused and surveyed the room. Pauses are long when you’re customerless, trying to launch a product, and demoing one of the world’s largest asset service providers. It was as if he was considering a trading decision. Then, to his traders, he spoke again.

“We have to be in this beta program.”

Free data can be the most expensive kind.

When you leave everything behind, where do you start? This was the question I faced eighteen months before sitting in that room full of traders.

I had decided to build a business. I had trading knowledge, I understood data strategy and analytics, and I had learned how to create financial technology. Although this framework was valuable, it could not be used without data and technology. So, where to begin?

With data. And technology. The product of which is analytics.

The analytics stack we created is a unique combination of technology. It is the result of connecting fit-for-purpose components in an alignment that produces more value than the sum of their parts. We started with a minimally viable stack. Emphasis on minimally. Now it has been re-architected into a modern stack1.

But even the greatest technology stack is useless without data. So while we needed data and technology to begin, data is the real genesis of analytics. In financial services, and portfolio management specifically, market and reference data are minimum barriers to entry. What gamblers call table stakes.

That doesn’t mean market and reference data are easy to solve. Legacy providers have increased the barriers to entry with cost-prohibitive pricing. Sourcing market data from the NYSE or NASDAQ on a bootstrapped budget does not work. Equally impossible is obtaining even the most common industry and sector reference data. S&P Global have artfully created a monopoly on classifying businesses like United Airlines as an airline.

Our “where-to-start” question occurred as an upstart U.S. stock exchange sought to disrupt traditional exchanges. IEX was making its pricing information available via an API. For free. This was before IEX Cloud, the IEX data offering, existed.

Free works on a bootstrapped budget. IEX was playing a long term game we were happy to join.

Most product decisions are tradeoffs. And disrupting market data pricing created tradeoffs for IEX. Near-real-time is not exactly the same as real-time, for instance. And United Airlines might be classified as a transportation company rather than an airline.

Even after developing an interface to IEX (which was not free) and accepting the data product tradeoffs, this was merely the entry price to enter the world of financial technology.

We still needed data that went beyond table stakes. Preferably free.

Enter FINRA. The source of data which when captured, transformed, and visualized, prompted that head trader to interrupt our demo. It’s important to pay attention to when head traders interrupt demos.

Like so many of our product beginnings, I do not recall how I learned that FINRA was making off-exchange market activity available. Upon discovering it I realized it had already been available for years. FINRA’s website had an OTC Transparency Section reporting summarized monthly volume activity.

More than 30 U.S. crossing platforms were designated by FINRA as Alternative Trading Systems (ATSs). Each was required to report to the FINRA / NASDAQ Trade Reporting Facilities (TRFs) in real time. But the TRF only records volume and price. Traders know something transacted off-exchange, but they do not know where it transacted. This is why ATSs are commonly referred to as dark pools.

The industry is generally, if slowly, moving towards transparency. At some point FINRA began to report not only monthly summaries, but weekly details. The weekly reports attributed ATS activity by symbol.

This prompted a question: is there signal in the FINRA data?

There has always been a concentration of ATS providers in Boston. It’s likely the number 2 geographic location for ATSs, outside of NYC. As a former Boston equities trader I’d always followed the local ATS firms. They were all competing to capture the same 30% - 40% of U.S. equities volume that trades off listed stock exchanges.

I was working at State Street when the bank acquired Pulse Trading, primarily for its BlockCross ATS. The acquisition, integration, and subsequent sale of BlockCross to Instinet could be a book by itself spanning M&A, corporate politics, trading scandals, and the glass box in which the BlockCross team operated2.

FINRA’s monthly summaries of BlockCross and all ATSs are released on a one-month lag. January data, for instance, is not released until March 1st. FINRA’s weekly detailed reports are released on a 2-to-4-week lag depending on how FINRA classifies the company being traded. Two tiers distinguish between more liquid (Tier 1) and less liquid (Tier 2) stocks.

Information value decays. FINRA reports can be perceived as not useful because of the time lag between activity and reporting. Surely the data would be more valuable if it was real time. But I wasn’t convinced the lag removed all information value. In fact, with some data sets, historical information can actually accrue value when augmented with other data or processed in novel ways.

Intelligent analytics can slow, stop, or reverse information value decay.

We began manually downloading FINRA reports a couple of years before Mark Enriquez became a Point Focal angel investor. Mark was the founder of Pulse Trading, the firm that created BlockCross.

Some things really do come full circle.

FINRA data is free, which is one reason we started with it. Usually, you get what you pay for in life, but not all the time. Stock markets illustrate exceptions to this rule every day.

The value of free data can exceed its cost.

FINRA’s reports were manual and messy. The time it took to source and organize the reports nearly doubled the 2–4-week reporting lag. And while the FINRA data was the most granular off-exchange information available, it was still aggregated by week. We began to question whether it had value.

Then we realized FINRA had made the reports programmatically accessible.

We had already made sense of the monthly ATS reports, thanks in part to the Tabb Group. It was Larry Tabb who interrupted a very early demo of what we were building to ask, “How do we partner?”. Ever since that question we have supported Tabb Forum’s Equities Liquidity Matrix.

Converting Tabb Group’s PDF reports into a web application taught us a lot about exchange and ATS data. It also helped teach us how to automate a fintech data operation with our analytics stack. What used to take several analysts many days to produce static reports now takes zero analysts zero days to produce interactive analytics.

TABB Forum Equities Liquidity Matrix (Free TF Subscription Required)

This understanding of off-exchange data and analytics helped shape the first piece of our Focus List, Dark Venue Analytics.

We had enough market structure information to start telling a new story. Our narrow, but deep analytics premise started with free, off-exchange venue data. We coupled that with volume by venue and large trade data from IEX Cloud. Combining traditional and alternative market structure content created a new liquidity perspective. A new market structure story. New analytics.

In real-time, we can see what’s happening on lit exchanges. We can also see what is happening in aggregate off-exchange, via the trade reporting facilities. Now we could also see what happened historically off-exchange across alternative trading systems and brokers.

We had three pieces of the market structure picture. Real-time on-exchange detail, real-time aggregate off-exchange summaries, and historical off-exchange detail. The question was whether these three components could be used to derive meaningful information about real-time, off-exchange detail.

Could the location of where liquidity was most likely trading off-exchange be determined before FINRA reported the answer? We are still writing the story and analyzing the answer, but one thing we can say with confidence is three pieces of market structure data are better than two.

Subscribe now

Free FINRA data is valuable partly because it’s messy, incomplete, and difficult to access. These characteristics have obscured the data from financial technology.

At Tabb Group’s Fintech Festival in 2019 we had a prototype of single-stock off-exchange analytics cobbled together by Northeastern interns. Clearpool, a dominant algo-provider with TCA analytics had a booth across from Point Focal.

Vendor halls are a petri dish of competitive intelligence. When we visited the Cleearpool booth they smiled and waved at our off-exchange analytics. They said they were working on building something similar. Upon further inquiry this meant someone at Clearpool had a spreadsheet somewhere populated with some FINRA data.

Each prospect we’ve visited who uses Clearpool expresses that off-exchange data and analytics are coming to Clearpool. In 2022, we haven’t seen it yet. Maybe it will come, maybe it won’t. Lacking FINRA data did not prevent Bank of Montreal (BMO) from acquiring Clearpool for an undisclosed sum in 2020.

It is worth paying attention to what firms choose not to do. Clearpool had decided the cost of integrating FINRA data was not worth the benefits. While we can’t know the calculus of Clearpool’s cost-benefit analysis, the product decision was clear. And if you’re trying to be honest with your business, it’s important to understand why successful firms in your space are ignoring the same data into which you are investing resources.

I had more reason to pause and reflect on our efforts when I explained to the head of an ATS provider what we were doing with the FINRA data. How we were sourcing, organizing, analyzing, and delivering off-exchange analytics. It was an unexpected encounter on the street. My explanation was a response to the “What are you up to?” question. I thought of all the people who might care about off-exchange analytics, the head of an ATS had to be one.

I’ll never forget his response. Smiling, but seriously, and slowly, he asked, “Why would you do that?”

Uh-oh.

I nervous-laughed and gave a short answer. The voice in my head said, “one of us is crazy”. I spent the rest of the night assessing the odds it might be me.

Then Bridgewater dropped their name into our website. They wanted a demo of our off-exchange analytics.

When the largest hedge fund in the world contacts you unsolicited, seeking a demo, you wonder if it’s a signal. With $140B in assets under management, Bridgewater was not our target market. Point Focal aims to help asset managers who do not have Bridgewater resources and capabilities. Further, we are not focused on quantitative hedge funds who want raw data from data-as-a-service (DaaS) providers.

Nevertheless, we were excited to understand Bridgewater’s interest and show them what we were building. Three people from Bridgewater joined our demo. A portfolio manager, an analyst, and, according to his LinkedIn profile, Bridgewater’s “Global Market Data Leader; Data Ranger & Finder”. This was a new persona.

The demo was brief, but thorough. The communication was brief and direct. There was real interest in our off-exchange analytics. In hindsight it may have been more intrigue than interest. Our Estimize post-earnings drift views also warranted demo time and discussion.

Unsurprisingly, Bridgewater intent and ambition was poker-faced and close-to-the-vest. They did reveal they were building similar off-exchange analytics. The data ranger contemplated whether it would be prudent for Bridgewater to subscribe to our analytics rather than continue to build something similar with internal resources. The PM and the analyst weren’t sure.

We set up trial access for the PM, the analyst, and the data ranger. There was a bit of back and forth and a second demo. At the time, our platform was too immature to track user activity. We suspect they never logged in.

In our last exchange, the data ranger said two things.

1) They were probably going to build it themselves.
2) Come back when you have 605 and 606 data.

Our favorite objection to a Point Focal subscription is “we’ll build it ourselves”. This has become such a common objection, we preempt it in demos. We explain to head traders that when we leave the trading floor, their IT department will tell them they can build it themselves.

We go on to explain that they’re right, they could build it. But if they were actually going to build it, it would already be built. And the fact that it will take thousands of man hours to build means it will not be prioritized because it’s not their IT mandate, not their business value proposition, and not aligned with corporate incentives that allocate resources3.

This we’ll-build-it-ourselves notion is a fallacy for the vast majority of asset managers. Even large asset managers. But when Bridgewater says they’re going to build it themselves, they’re probably going to build it themselves.

And they should. They are the exception to the rule.

But this demo was not noise. It was signal.

The head trader who said, “We have to be in this beta program” put his team in our beta program. They had deep domain knowledge of on and off-exchange market structure. But they also had something we would learn is rare in capital markets. They understood how to apply market structure analytics to trading strategies.

This is why the light bulb went off when they saw our analytics. It wasn’t merely information reporting or theoretical scenario analyses they envisioned. It was the practical application of integrating analytics into their trading workflow.

When we get past the “We’ll build it ourselves” objection, we get to the real objection, “We don’t understand how to get your analytics into our workflow”. This is the challenge all portfolio analytics firms face when selling into asset managers and service providers.

The financial technology stack is difficult to break into. Application real estate is sacred. Legacy technology players are entrenched with sticky services protected by moats of high barriers to entry. To insert your product into this stack requires real innovation. Not the kind generally trumpeted from Chief Innovation Officers. Innovation that demonstrates growth through improved performance and risk management.

On this trading desk we had found a team who understood our analytics, how to integrate them into their workflows, and how to use them to grow their business. This was the product market fit we were chasing.

This firm was engaged. We met every other week to exchange feedback. They provided input that informed product views. We created workflows that made it easy for them to use our product alongside their existing fintech stack. When we exited beta they became our first paying customer.

The combination of their domain knowledge and our analytics produced knowledge that improved their trading strategy. And if you can improve a global asset service provider’s trading strategy, you have a business. It doesn’t matter how much the underlying data costs, how messy it is, or how challenging it is to organize.

All that matters is whether the analytics tell a story that improves performance and risk management4.

We’ve learned a lot from when we first manually downloaded FINRA data to when we used it to win our first customer. Here are the lessons:

1. Data cost is not always positively correlated with data value.

2. To integrate analytics into trading workflows, talk listen to traders5.

3. Invest time and resources into engaged prospects. Like so much in life, input drives output. I explain the exception to this rule in the Signals pod “Crossing the Chasm”.

4. The timing of head trader demo interruptions matter.

5. If people are asking if you’re crazy, you may be on to something.

In the world of data and analytics, little insights can make big differences. They lead to business knowledge, which produce better decisions, which improves performance and risk management, which wins new business.

The analytic value chain inevitably leads to growth. If the chain breaks down before growth occurs, the insights are insufficient. And even with meaningful insight, analytics must be integrated into investment management workflows to link the chain and capture growth. Isolated analytics cannot break into the chain. Stranded outside of workflows, they end up in the analytic grave of wasted resources.

Identifying valuable data is necessary, but not sufficient. Producing meaningful analytics from valuable data is necessary, but not sufficient. Integrating meaningful analytics from valuable data into investment management workflows is sufficient. This is where the analytic value chain is connected end-to-end and the magic happens.

It is how little insights lead to business growth.

Afterword

Shortly after Bridgewater told us to come back when we had 605 and 606 data, Tim Quast from Modern IR requested a demo. Modern IR provides market structure insight for corporate investor relations teams. (Investor relations is the IR.) Tim thought our analytics might help his IR reporting.

We met late on a Friday, late in 2021. Tim’s breadth of market structure knowledge was overwhelming and offset by a sharp sense of humor with an untraceable accent6. We spent eighty minutes sharing product stories, market structure humor, and platform demos.

There was something that made sense between Modern IR and Point Focal, some sort of complementary translation that begged exploration. If our firms were a Ven diagram, the overlapping section was truly novel, even if at the time we didn’t quite understand the overlap.

Tim and I agreed there was something here worth exploring. Then 2022 arrived, we both got busy with our businesses, and we fell out of touch. I followed Tim’s Modern IR progress and his Market Structure EDGE business launch from a distance.

In EDGE Tim had created a market structure sentiment model of demand and supply. Underneath the model was 605 and 606 data. Best execution and short volume information. Our business overlap was becoming clear.

Tim and I remained focused on our respective products in ‘22. But I kept thinking about our overlap. Tim’s content and domain knowledge and our client workflows and institutional analytics.

In October I invited Tim to Boston to join me on a Boston Securities Traders Association (BSTA) panel. I knew he would be a wildly entertaining and educational panelist. A perfect misfit. But selfishly, I wanted to show Tim the institutional community in which we are embedded. I wanted to understand Modern IR and Market Structure EDGE. I wanted to understand our overlap and bring it to market together.

Tim graciously said yes and arrived in Boston in cowboy boots. The panel was a blast. Turnout was great and there was a feeling of post-COVID normalcy. Tim was hysterically informative in the greatest way.

I made it clear to Tim I wanted to help bring EDGE to institutions. He smilingly pondered the thought while absorbing the portfolio management and trading community before us. Afterwards, he returned to Colorado and we both got crazy again developing our products.

Then a month later I got an email from Tim.

Subject line: “EDGE for Institutions”.
Message: “Let’s figure this out.”

Figuring this out has led to an exciting collaboration between ModernIR, Market Structure EDGE, and Point Focal. Together we’re enriching market structure sentiment analytics. And we’re designing a more complete investor relations dashboard combining market structure with earnings, ESG, and news information.

Like our BSTA panel, our goal is to educate and help those who can benefit from a truer understanding of how market structure explains market behavior7.

The psychological equivalent to demand and supply is greed and fear. More base human forces that manifest in market behavior. From Jesse Livermore to Michael Saylor and everyone in between. Greed and fear drive markets entirely.

For as long as markets have been observed, humans have tried to measure greed and fear. This measurement is best classified as sentiment. Sentiment is everywhere. Political approval and disapproval polls, product reviews, and call center responses are all sentiment. The prose of this Substack can be converted into sentiment.

While sentiment is not exclusive to financial markets, it is maybe most obviously prevalent in financial markets. Measured in many forms.

• The VIX (CBOE Volatility Index)

• CNN Fear and Greed Index

• High Low Sentiment Ratio

• NYSE Bullish Percentage

• Consumer Confidence

• The Buffet Indicator

• NAAIM Survey

• StockTwits

Language is sentiment. Which means everything is sentiment. Including, of course, social media and the great bird application, Twitter.

Like earnings, sentiment is almost universally relevant. This is why we identified sentiment as an important input when we started building portfolio analytics. I wish I could say we started with PsychSignal, but we didn’t. Many of the mistakes we’ve made building Point Focal have been about failing to properly choose what not to do. And so it was when we started building a natural language processing (NLP) sentiment model to create analytics that would capture the most fundamental underpinning of market forces. Greed. And fear.

Three years ago, while teaching in Northeastern’s masters of analytics program I had a couple of capstone students helping Point Focal. Another Focus Signal post will likely explain why I was teaching at Northeastern. But I was. And one of the many benefits of the experience has been creating a rotational internship program with the best students from the capstone. By “best students”, I do not mean highest GPA. I mean the most ambitious and creative students who understand apprenticeship value. In my experience, this profile has not been correlated with the highest GPA.

Enter Zhoujun. A smart, creative capstone student who decided the benefits of working with Point Focal were greater than its costs. Zhoujun helped us build a simple model that could process Twitter content associated with stock symbols using R sentiment packages.

Many great models are simple. But not all simple models are great.

Our model was like a single blade of grass in an endless field of sentiment. Despite the limiting content constraints of the Twitter development environment, we were overwhelmed by tweet volume. The sentiment models we used classified language into emotions - none of which were greed or fear. They were more like the happy / sad faces of on-line surveys no one completes.

I recall Zhoujun trying to decode Apple product sentiment with the model and thinking that despite its tiny scale, we were hypothesizing that our model might explain iPhone sales. If a sample of Apple tweets could be converted into emotions that produced meaningful information through an off-the-R-Studio-shelf sentiment model, it could explain anything. And a model that explains anything explains nothing.

Weeks into our effort it was clear the challenge was beyond us. We lacked the financial capital the raw data required, the natural language processing expertise the model required, and the technology infrastructure the data operation required. Not least of all, we lacked the human capital the execution would require.

Otherwise, we were ready to roll.

All we needed was zero cost, post-NLP-processed data, a near-zero cost tech stack, and automation instead of humans. It is moments like these when many rational actors might stop. But solving problems is also rational. And this was a problem to solve.

Enter PsychSignal. I found PsychSignal online and remember thinking they were mature enough to produce the content we needed and small enough to be approachable. Indeed, they seemed to have a few employees and a few customers - one of whom evidently funded the operation. Today, Crunchbase reports PsychSignal has raised $230K and have 3 employees with the following company summary:

PsychSignal quantifies the real world psychology of the investment crowd by listening to social media conversations. They provide the most granular real time financial sentiment APIs in the world. They do so under the leadership of a team of experts in psychology, engineering, and data science, plus a few legendary traders from Wall Street.

I dropped a note into the PsychSignal website explaining why Point Focal wanted to access their data.

“We believe our product would benefit from enabling PMs and traders to view PsychSignal data… We are unlikely to have the capital required to subscribe to your data… However, I don’t think we need the fire-hose version of your data to create value for our users. A fairly limited set would likely be sufficient.”

James Crane-Baker responded immediately. He was receptive to engaging and simply asked what we could contribute to help defray resource allocation costs while allowing us time to get to market. In the end, what we and their other customers could contribute would not be enough. But it was enough to begin, and for that I remain grateful to James and those like him who are willing to engage with early stage, bootstrapping founders.

PsychSignal covered thousands of global securities. Many more than we needed with our U.S. equities scope. They offered real time intraday content and a single, overnight file. There is a huge difference in complexity between processing streaming data and processing static data. We started with the daily static files from PsychSignal. Choosing not to start with streaming data was one our better PsychSignal decisions.

We parsed the daily data into more than 13,000 individual security files. Then we migrated a few years of history and created a process to append daily data from the API to our stock files. A separate process produced a single, filtered, merged file of PsychSignal data for the roughly 3,500 stocks in our beta universe.

PsychSignal had a feed from Twitter sourced by cashtags ($TSLA) from which they produced the following fields: Bullish Intensity, Bearish Intensity, Bull-Minus-Bear, Bull-Scored Messages, Bear-Scored Messages, Total Scanned Messages, and Bull-Bear-Message-Ratio.

We loaded a few million rows of PsychSignal data joined with reference and pricing data into Spotfire. Finally, we had sentiment analytics.

There was a lot we wanted to analyze with our new PyschSignal data. Our primary goal was to understand the effect of sentiment on prices - analyzing correlation and its materiality or spuriousness. But beyond an unlikely golden-sentiment-bullet that might help forecast future prices, we were interested in what else we could learn from the flow of social emotions.

We wanted to understand the relationship between extreme sentiment and future volatility, regardless of price direction. We wanted to know over what time frames sentiment signals are most valuable. We wanted to design sentiment model parameters and generate back-tested results with historical performance attribution. We wanted to reveal currently actionable signals. And we wanted to know if creating and rebalancing long / short baskets of positive / negative sentiment stocks could ourperform a benchmark. We were excited. But first we wanted to do something we thought should be simple. Color stock prices by intensity of sentiment.

A typical line chart in visualization software is designed to plot lines based on continuous variables over time. And they allow the lines to be colored by a separate categorical value. Consider 2 lines plotted where Y is performance, X is time and each line is colored by a category, portfolio.

Or 6 lines in the same space where lines are colored by symbol.

This design solves most line chart use cases.

Now consider a line in the same space and our desire to color the line by a different dimension - a continuous scale of sentiment rather than a categorical scale of symbol or sector. This is far more complex and not the use case for which visualization software is designed. It is what project managers call a corner case. Smart product owners tend not to allocate lots of development resources to corner cases.

So we bastardized the scatter plot in Spotfire. Instead of drawing the price of the stock as a line, we drew price as a scattering of plots. Then we made the plots very small and connected them so that they appeared as a line. Finally, we colored the plots that made up our line by sentiment.

Now we could see greed and fear. And for the moment, it was thrilling to visualize the sentiment our social media world was ascribing to companies.

Afterwards, our technology lead informed me I was using the visualization software incorrectly. I explained that the software was not designed correctly. Of course, both of us were right. Risk is introduced when using software in a manner for which it is not designed. But with risk comes reward. And extending software to do something beyond its original design can lead to great advances.

While our sentiment views lacked maturity, they increased the number of alternative data sets in our product from two to three. We had off-exchange activity, we had Estimzie earnings, and now we had PsychSignal sentiment. The off-exchange activity was the first data we brought into Point Focal and therefore had the most mature analytics. PyschSignal was last of our initial three and therefore had the least mature analytics. But three feels more like a product than two. Heck, with three we were a self-described alternative data platform.

As we explained to prospects, narrow, deep analytics are superior to wide, shallow analytics. This is fundamentally true and an important premise on which we build. We had the narrow part down. We were working on the deep part.

Our sentiment views evolved. They became more complex and then simpler. The demo response to our sentiment analytics was heightened interest. People had not seen their portfolios colored by social sentiment. But interesting analytics are only necessary, not sufficient. People do not pay for interesting. They pay for compelling. And the bar for compelling analytics in asset management is improving performance and risk management decisions.

Eventually we overlaid moving averages and Bollinger bands onto the sentiment-colored price history. And we created parameters that could be configured to identify moments in time when pricing and sentiment information converged in a meaningful manner.

But what is meaningful? Our curiosity was based on whether extreme price and sentiment conditions created a propensity for stocks to extend or reverse direction. Momentum and mean-reversion. We populated greed and fear flags over scenarios isolated by model parameters.

With psychological labels applied to our visuals, stock behavior could be observed before and after the labels. Thus, from a set of configurable price and sentiment parameters, we could analyze volatility and performance within the context of social sentiment. We calculated metrics that quantified performance from the signals produced by the parameters.

This was the beginning of creating a discretionary framework from which to analyze sentiment impact on stocks and portfolios. The combination of outputs that could be produced from the model was large. There was so much to do.

Subscribe now

“Everyone has a plan ‘til they get punched in the mouth.” - Mike Tyson

On the morning of April 14th, 2020 I received an email from James at PsychSignal with the subject line, “Data Subscription Termination Notice”.

“Unfortunately due to increasing upstream data costs associated with producing this data, we have decided to terminate the live data feed, effective immediately.”

After a brief email exchange with James, I confirmed we had a problem. All the sentiment views and functionality in Point Focal no longer had a data feed. The greed and fear we had worked so hard to capture in our portfolio analytics were gone.

Beyond the obvious pain this inflicted on our morale and the lost sunk cost of our sentiment efforts, it also forced us to drop everything we were doing and figure out how to save our sentiment.

This is the opportunity cost of time that can kill startups. Projects across infrastructure, data analysis, strategic partnerships, outreach, and everything else we were doing stopped to focus on sentiment. The thought of our demos going from 3 alt-data sets to 2 was demoralizing.

We began to consider alternative sentiment data sources and realized that IEX Cloud, the new financial data provider whose API we used for market and reference data had brought StockTwits sentiment data into their platform. We had seen the IEX Cloud API mature technically and expand its content over the years, but we had not realized sentiment was available. The API endpoint contained daily scores and activity which we could, at least in theory, use to replace our now legacy PsychSignal feed.

IEX Cloud knew Point Focal. Probably because we had been there from the start. We were using their API before they productized it. I’m confident we were early enough to have provided helpful product input. Over time we got to know their team and shared our own product with them. We published a joint case study and even convinced them to let Charlie and me ring the IEX bell.

Of course, it’s not actually a bell. It’s a button which “closes” the market. It’s a great gimmick because everyone knows it’s a gimmick and because IEX uses it to demonstrate how milliseconds matter. No one in practice ever presses the button at exactly 16:00 EST because the press is measured with incredible precision. This closing ceremony is messaging at its best. Reminding the world that IEX is designed to neutralize speed advantages sought after by high frequency trading operations. See Michael Lewis’s Flash Boys for the spellbounding details.

So when I wrote Josh Blackburn, head of IEX Cloud, to ask for help, he at least knew who we were. I explained that PsychSignal was immediately terminating their data feed and we wanted to rapidly convert to StockTwits through their API. I also asked for help organizing the historical data required to completely replace PsychSignal. Despite not knowing whether the signal in StockTwits was better, worse, or indistinguishable from PsychSignal, we knew we could not have sentiment history that began with one source and ended with another.

Josh responded in eleven minutes with “We can get this done for you today”. In less than 24 hours, IEX Cloud delivered a flat file of sentiment data to us that kept our product from regressing. We backed up PsychSiganl data and stripped it out of our product. When we loaded StockTwits data we didn’t even change the data pipeline. We morphed it into the PsychSignal stream with unchanged column headings that tricked our product into thinking StockTwits was PsychSignal (to antrhopomorphize technology).

When this occurred in April of 2020, our product was still in Beta. And with a minimal amount of pain we were back in business. Beta user experiences and demos were largely uninterrupted. Having the support of a larger partner with more resources than Point Focal was wonderful. It’s why in our case study that would be published one week to the day after this event, we wrote:

“IEX Cloudʼs responsive support makes us feel more like a partner than a client – they are invested in our success.”

Abraham Thomas explains in The Economics of Data Businesses:

A common failure mode is to build a business on top of somebody else’s data. If you depend on a single upstream source for your data inputs, they can simply raise prices until they capture all of the economics of your product. That’s a losing proposition. So you should try to build your own primary data asset, or work with multiple upstream providers such that you’re not at the mercy of any single one.

Indeed, PsychSignal could raise prices, but they could also terminate their feed! Both situations are undesirable. We are doubling our upstream data providers this quarter from 3 to 6, one of which has their own aggregated content. Between this data diversification and the maturing IEX Cloud offering, we are becoming less susceptible to the failure mode identified by Thomas, who goes on to say:

You should also try to add proprietary value of your own, lest either your suppliers or your customers encroach and disintermediate you. A sufficiently large transformation of your source data is tantamount to creating a new data product of your own.

This is critical for Point Focal. Across data sets, and specifically with sentiment, we are overlaying analytics onto the raw data to transform the original content into something new. This IP layer creates usable insight from raw information and is how we are creating a new data product of our own.

Sourcing and managing alternative data, producing signals with quantified performance for custom portfolios delivered with visual analytics, natural language narratives, and automated reporting… this is our new data product.

With our sentiment disaster averted, we began to analyze StockTwits sentiment performance relative to PsychSignal. And we returned our focus to our beta program. Then, three weeks later on May 20th, we received another sentiment email, this time from IEX Cloud:

We’re reaching out to let you know about upcoming changes regarding our social sentiment data provided by Stocktwits. On June 1, 2020, Stocktwits data will be moving off our Core Data offering.

We had just saved our platform from losing sentiment. Now we were losing sentiment again.

One thing starting a business teaches is when everything is a surprise, nothing is a surprise. While it may not be possible to expect the unexpected, it does seem possible to become desensitized to surprises. To become unsurprised.

Another startup lesson is understanding that the game is to continuously solve problems until you generate revenue. If you cannot solve an important problem, or if you run out of time (money) while solving important problems, the game is over. Having our sentiment feed shutoff twice in two months was just another problem to solve.

I spent the next month being rejected by StockTwits and RavenPack.

I figured if we couldn’t source StockTwits data from IEX Cloud, maybe we could obtain it directly from StockTwits. In fact, seeing how StockTwits was using their own sentiment data, I thought we could probably help them too. No one at StockTwits would respond to me - even after Linking-In with me.

Something about the nature of StockTwits made me not want to chase them. While retail / social sentiment could be a valuable alt-data input, it felt like a niche of a niche. Even their name, StockTwits, turned me off1. Of course, it’s easy to feel turned off by a firm that won’t engage with you.

I met RavenPack at a Tabb Forum fintech event in 2019. RavenPack was an established company out of Spain with a set of NLP analytics across text, news, insider transactions, and corporate events. After an exchange with their product expert from the event, I met with their Head of Strategy who reported to their CEO.

RavenPack appreciated what we were doing and thought they could help us. But they very nicely informed me that without significant upfront capital from Point Focal, they would not engage with us. They had tried revenue share deals with small firms before and determined that the risk-reward, cost-benefit equation for RavenPack was not favorable. In fact, they expressed that through no fault of our own, their appetite to engage with us was impacted by their prior engagements with other small firms. Understandable, but frustrating.

Most startups fail. But we were fighting to become unlike most startups.

It is this risk-reward equation for established firms that prompted me to write about Estimize’s willingness to engage with Point Focal in Solving an Earnings Challenge:

From experience, I know there are many firms unwilling to work with a bootstrapped startup. This is both understandable and exhausting. I appreciate the cost-benefit analysis of such decisions from established firms. However, I also believe if there is minimal cost, low risk, and asymmetric upside to the established firm, there is value in supporting aspiring businesses. Sometimes filters should be applied to cost-benefit analyses. Leigh’s willingness to work with Point Focal is something I am excited to pay forward to another firm in the future.

I don’t know where I discovered Brain. I don’t even think I was looking for sentiment when I saw Brain. But it was June 11th, 2020, nine days after my RavenPack meeting, when I sent Brain’s CEO, Francesco Cricchio, a LinkedIn message introducing myself, our sentiment predicament, and the nature of our bootstrapped fight for survival.

Francesco responded immediately and was happy to explore an engagement. Brain was based in Europe and had a variety of NLP alternative data sets, including single-stock sentiment. After a couple of calls with Francesco and Brain’s Head of Research, Matteo Campellone, we were not only back in the sentiment business, we had a new partner with complementary capabilities genuinely aligned with our model.

This felt different and was exciting.

As I discussed on the 12th episode of Signals, Split Personality, it’s not only the things you get in life that shape your path, it’s also the things you don’t get in life. We did not get PsychSignal, StockTwits, or RavenPack.

Brain is a young firm with wonderful Italian accents based in Milan. Like Point Focal they are small, entrepreneurial, and ambitious. Unlike Point Focal, they have deep expertise at the intersection of natural language processing and data science. Beyond global individual stock sentiment, they have a growing suite of alternative data and quantitative capabilities that includes global market sentiment, sentiment volatility, thematic baskets, 10-K, 10-Q, and corporate earnings call language metrics, machine learning stock rankings, and a programmatically accessible back-testing engine. Brain was more than a sentiment feed!

Brain enabled us fast. In a matter of weeks, we replaced our sentiment data pipeline for the second time.

The best partnerships come when natural synergies exist. Showing Brain their content in a way they had not seen it before demonstrated our capabilities. Brain’s NLP data expertise and our ability to visualize their sentiment created an end-product more valuable than what either of us could produce individually. The complementary nature of our offerings was promising.

Brain’s sentiment is sourced from news rather than social media, lending itself more towards signal than noise and creating a more institutional feel to our analytics. Brain has educated our team on how to interpret sentiment scores, how to align the timing of news sentiment with stock performance, and how to objectively analyze the predictive value of sentiment information.

This last point is crucial. As we quantify the statistical meaningfulness of stock sentiment and performance, we are learning that the sheer volume of strongly scored news sentiment - regardless of whether it is positive or negative - can indicate future stock volatility. Further, the volatility of broad market sentiment has recently preceded significant risk-on, risk-off inflection points for U.S. equities.

Our Brain sentiment content will quickly grow from individual stock to portfolio level analytics. Intraday broad market sentiment and sentiment volatility will follow, creating a completely new, quantitative view of sentiment behavior. Then, we will do something more ambitious.

As we integrate more Brain content we will create a comprehensive suite of sentiment and NLP analytics. Coupling these analytics with the Brain backtesting engine will effectively productize sentiment and NLP quantitative models. Pushing timely alt-data driven signals into our insight platform creates a novel offering. A joint capability that can help a broad swath of the asset management community.

The best part will be producing performance attribution reports that demonstrate that Brain and Point Focal turn sentiment analytics from nice-to-have to must-have.

Our original vision of producing a unique sentiment lens on portfolios has evolved into something greater. With the right partner we’ve learned that sentiment is the product of an NLP capability. And using that capability across a broader set of language content creates analytic depth. It is depth that differentiates our analytics.

As a startup focused on alternative data and portfolio analytics, content is king. Again from Solving an Earnings Challenge, I wrote that content selection is paramount:

Selecting data is a critical input to building portfolio analytics. The solution’s entire worth is data dependent.

As data selection is to portfolio analytics, partners are to businesses. The most important function of any company is talent acquisition, an HR term for deciding who you work with. This is true of internal employees and external partners. Companies do not have thoughts, feelings, or sentiment. Their people do. And so with whom you build your business determines your sentiment. And your outcome follows.

We feel very fortunate to have a strong and growing ecosystem of partners working with Point Focal. We are connected by a shared sentiment. And just as an analytics stack can be built by uniquely connecting technology components to create a system more valuable than its parts, so too can a product distribution stack be built by uniquely connecting complementary businesses into a network more valuable than their individual companies.

Our partnership with Brain epitomizes this alignment. And it is the result of solving one sentiment challenge after another until you have a sentiment solution.

I wrote about error correction in Pub-Sub Fit. Here its process has produced sentiment knowledge:

1. To understand sentiment analytics, understand sentiment production.

2. To understand sentiment production, understand its NLP engine.

3. Understanding the engine leads to other uses (leverage and scale).

4. Understanding other uses creates NLP analytic depth.

   ---

This has led us to a set of NLP analytics spanning single-stock, portfolio, and market sentiment, 10-K, 10-Q, and earnings calls language metrics, and smart thematic baskets built on processed language. Quantifying this content produces new ideas, signals, and insight for anyone who wants to understand how greed and fear can impact performance and risk.

Our sentiment path turns language into data into language2. News stories into sentiment scores into quantitative research. Words to numbers to words.

It’s a long way from our first attempt to interpret the mood of a tweet.

And surely the product of our own greed and fear.

Challenging problems with a high difficulty level can take time to solve. And when markets want a problem solved, the more difficult the problem, the more valuable the solution. Solving challenging problems can be rewarding. If you don’t run out of time.

Point Focal was created to capture and make sense of high-value, difficult-to-access alternative fintech data. Selecting data is a critical input to building portfolio analytics. The solution’s entire worth is data dependent. Selection is particularly daunting when starting from a blank canvas of nothingness. Not a single column, row, or data element. A challenging problem to solve.

Earnings matter. Demand for future earnings information is universal. Everyone, buy-side, sell-side, institutional, and retail, want high quality information about future earnings expectations. Every portfolio manager (PM), trader, analyst, and individual with a pricing model - fundamental, quantitative, and quantamental - requires earnings information as an input.

A premise of ours is if we can identify alternative earnings data with more meaningful and verifiable value than traditional earnings data, we can help everyone who cares about earnings. In this sense ‘alternative’ means different from Wall Street. And ‘help’ means automatically overlaying quantitative earnings analytics on portfolios.

Estimize is a 10-year-old fintech firm with a crowd-sourced platform of earnings estimates. Their estimate sample is larger than Wall Street and comes from a more diverse buy-side, sell-side, academic, and practitioner community. This is the ‘crowd’ in crowd-sourced. And while the term can cause some to recoil, it shouldn’t. Especially for those who appreciate that crowds, in many contexts, tend to have more wisdom than single or central institutions. The wisdom of this crowd is not unique. Estimize estimates are quantitatively proven to be more accurate than traditional Wall Street sell-side estimates.

Over eight years of data history, relative to Wall Street, Estimize has produced more than twice the number of estimates per earnings release on average with wider estimate dispersion and three times the average number of revisions per estimate. The Estimize consensus has been closer to actual reported results 70% of the time compared to legacy sell-side only estimate data sets.

So why is everyone looking at Wall Street estimates when there is a higher quality source of future earnings? Historical inertia. In other words, my least favorite answer to the “Why do we do it like that?” question. “Because we’ve always done it like that.”

Spoiler alert: Wall Street has a bias. A bias so obvious it is taken for granted. The same investment banks that take companies public are publishing estimates on their future earnings. As can be observed in the data, it is more difficult for companies to beat Estimize estimates than to beat Wall Street estimates. A non-trivial question arises from further exploration: if a company never misses Wall Street estimates, are Wall Street estimates valuable?

Higher quality earnings estimates mean better information, improved model accuracy, and opportunities to improve performance and risk management. We wanted to bring Estimize content to the asset management community.

From what we could tell, only the most sophisticated, quantitative managers were using Estimize data to increase alpha. We believed this was due to two reasons:

1. The difficulty level of integrating an end-to-end Estimize workflow into an investment process.

2. Lack of awareness.

It is not easy for an asset manager to capture, transform, and contextualize Estimize data across their portfolios at scale. It is impossible if you don’t know Estimize exists. We considered these problems to solve. This is our Estimize story.

We prioritized Estimize. This was the first data set beyond market, reference, and free1 data that we pursued. In the end, our earnings premise was simple. Solving it has been incredibly challenging.

Our pursuit started with a cold email. I wrote Leigh Drogen, founder of Estimize, twice over a few weeks. Leigh responded, joined a call with me, and appreciated what we wanted to accomplish. Leigh empathized with the challenges of building a fintech data and analytics product. He is a believer in and a supporter of entrepreneurial efforts in the space. From experience, I know there are many firms unwilling to work with a bootstrapped startup. This is both understandable and exhausting. I appreciate the cost-benefit analysis of such decisions from established firms. However, I also believe if there is minimal cost, low risk, and asymmetric upside to the established firm, there is value in supporting aspiring businesses. Sometimes filters should be applied to cost-benefit analyses. Leigh’s willingness to work with Point Focal is something I am excited to pay forward to another firm in the future.

We began exploring Estimize data with great enthusiasm. The beauty of Estimize is its simplicity: better estimates. Yet as we analyzed Wall Street estimates, Estimize consensus estimates, actual earnings per share, revenue, estimate counts, weighted and unweighted consensus2, contributor information, and other metrics, we discovered there was a lot to learn. And as we learned, we began to understand that the implications of better estimates were greater than we initially thought.

Our first goal was to create an institutional workflow making it easy for PMs and traders to consume and understand Estimize data with descriptive earnings analytics overlaid on their portfolios. Occasionally we would get excited about something else we could3 do with Estimize data - quantitative models, contributor analytics, company KPIs - there were lots of possibilities. But each time we veered from our initial vision we set ourselves back. The art of building a business is often choosing what not to do. Time kills startups. Every delay a silent tick of the demise clock.

With initial views created, we began telling the Estimize story in product demos. It is a fantastic story, and we were surprised by how many asset managers were unfamiliar with, and amazed by, Estimize. Yes, the estimates are more accurate than Wall Street. Yes, we have the history. Yes, we can show you trends and the latest consensus and revisions for your upcoming releases.

Our views created curiosity and interest, but traders wanted to understand how to incorporate Estimize data into their workflows. This produced many discussions that led toward an answer still being shaped. The process was informative and caused us to shift analytic views from descriptive to proscriptive. We created analytics illustrating what happened after historical earnings releases to help inform current earnings releases. The historical future. Because time is relative. Specifically, we created post-earnings drift views displaying return curves of stock performance following earnings releases. We were moving from interesting to useful.

Over time we enhanced our views to display significant post-earnings drift detail. One to twenty-day performance metrics in the post-earnings drift period were aggregated by releases that beat and missed Estimize and Wall Street estimates. This is valuable information previously inaccessible to portfolio managers in the aftermath of an earnings surprise.

Changing trading desk conversations from gut-feelings to quantitative analytics is a mind shift that does not happen overnight. Change resonates most with those unconstrained by primitive “we do it this way because we’ve always done it this way” thinking. We want to work with those who want to change. To find them, we are continuously refining how we communicate the value proposition of having superior estimates leading up to and immediately following earnings releases.

Subscribe now

Another reason we love the Estimize story is the nature of earnings seasons. Four times a year, portfolios enter and exit earnings season. As we continued to explore and create Estimize value, we began to understand how the data can be useful in pre-earnings, through-earnings, and post-earnings time frames.

Ahead of earnings releases the delta between Estimize and Wall Street estimates can be tracked. The Estimize platform is far more dynamic than the traditional sell-side process, enabling tracking of new estimates, revisions and estimate dispersion leading into earnings events. Our analytics are informed by how rapidly Estimize estimates change, how dramatically they change, and how close to earnings releases such changes occur.

Some of the analysis is counter intuitive. For instance, how should one interpret the scenario where Estimize is expecting a higher earnings release than Wall Street? Does this mean the company is likely to report higher earnings than Wall Street is expecting? Yes. Does it also mean the true reflection of earnings is a more difficult standard for the company to meet and that the company may therefore struggle through earnings season? Also, yes.

Answering these questions led us to the Estimize Post-Earnings Drift quantitative model. In fact, we had begun building it before we knew what it was.

The original Estimize post-earnings drift event study was published in 2018 and explored “the relationship between the post-earnings EPS surprise and the cumulative residual returns in the days following the event”. The study’s conclusions are as follows:

• There is a positive cumulative residual return when the EPS surprise is positive and vice versa.

• Negative earnings surprises tend to produce a larger negative return than positive surprises.

• After day one, negative earnings surprises show a reversal in cumulative residual returns from negative to positive.

• Estimize EPS deltas produced more alpha following an event than the Wall Street EPS deltas.

• Sector analysis showed most sectors produced strong alpha except for Consumer Staples, Healthcare and Utilities.

• Market Cap analysis showed that Small and Mid-Cap stocks produced the strongest alpha.

For the quantitatively inclined, here is a link to the original Estimize post-earnings event study in GitHub.

A more concise summary of the study: Estimize data has alpha. With this knowledge we set out to replicate, optimize and productize the post-earnings drift model. Estimize has been quite helpful along the way and even validated from their experience that there is a market for quantitative modeling as a service solutions.

We have unwound and rewound the event study. First locally in R. Now in AWS using Python. We failed multiple times trying to implement the model. But we believed in our premise, in the value of the data, and in the insight we knew it could produce. We sought and received all the help we could and eventually implemented the model successfully. Indeed, there is alpha in the model. But damned if we knew how to calculate cumulative residual return. We are solving that equation now as part of the performance attribution views we are layering onto the model.

Processing the model from Q1-2018 through Q3-2021 resulted in analyzing 4,500 earnings releases across 1,400 symbols. The model classifies every release into one of ten quantiles based on the statistical significance of the earnings beat or miss. Importantly, earnings yield normalized EPS results are used in the classification.

The basic premise of the model is that the largest positive earnings surprises (quantile 10) outperform the largest negative earnings surprises (quantile 1). Simulating a long-quantile-10, short-quantile-1 strategy, results show an average post earnings drift performance delta between quantiles 10 and 1 of 101 basis points on day one, increasing to 217 basis points on day twenty. The median performance delta between the same top and bottom quantiles are 64 bps and 186 bps, respectively.

Of market caps, small caps realized the strongest performance. Our sector analysis is in process, and we are considering weighting the results. One challenge is the sector classification information available to us is a distant second in popularity and use. GICS4, the most common sector classification scheme, is laughably expensive. We look forward to paying for it as our business matures or identifying a superior classification system.

A lot happened in the world between the start and end dates of our analysis. COVID surfaced. Trump was banished from Twitter. And we won our first customer. The VIX CBOE Volatility Index spanned a low of 9 to a high of 82. We analyzed many cuts of the data including pre-and-post-COVID periods, low and high volatility periods, and more. We seek to identify optimal post-earnings drift model conditions and make various scenarios available for analysis.

How an asset manager uses Estimize information depends on where they exist on the discretionary-to-systematic strategy continuum. Our goal is to deliver end-to-end analytics that deliver everything from descriptive context to quantitative signals. While we have been building the Estimize post-earnings drift model, we have also been rearchitecting our analytics stack from top (ingestion) to bottom (delivery). In so doing, we are creating the ability to automatically process the Estimize model daily for custom portfolios. We will generate and surface the resulting signals at the right time to the right people. And we will provide thorough performance attribution for portfolios and their benchmarks.

Most exciting is the fact that post-earnings drift is only one leg of a three-legged Estimize quant model stool. The other two legs are pre-earnings drift and through-earnings drift. By analyzing Estimize revisions and Estimize to Wall Street deltas during the pre-earnings period, we can introduce risk management insight. The through-earnings model bridges pre-and-post earnings models adding quantitative metrics from the interval one-day-before earnings to one-day-after earnings. Estimize enables us to deliver a year-round, continuous stream of earnings analytics.

We are fortunate to also be partnered with Brain Company. We are working with their back-testing engine to integrate model results with our Brain News Sentiment Indicator analytics. But we can also use it with non-Brain content like Estimize. This means we can deliver rolling quantitative model results including cumulative return, annualized return, Sharpe ratio, volatility, max drawdown, time to recovery, and more. The end-to-end quantitative model as a service offering is becoming tangible. I will provide a deep dive into our Brain partnership and their NLP (Natural Language Processing) driven alternative data in a future Focus Signal post.

Envisioning the systematic processing of all three Estimize earnings models across custom portfolios year-round, with signal generation and performance attribution throughout, gives one an idea of the magnitude of this challenge. It is what excites me about delivering value in this space. It is why I’m writing Focus Signal.

Productizing Estimize analytics at scale, in full, is still a solution being realized through our execution. While we are confident, we are also humbly aware that time, the great startup killer, is our biggest enemy. Solving this earnings challenge is not for everyone. This earnings solution is not for everyone. But for PMs, traders, and analysts who want to monetize data, it is a challenge worth solving.

Product market fit is an infinite regression. In a bootstrapped company you shouldn’t build a product until you have a market and you cannot find a market until you build a product. Crossing the product market chasm means negotiating between the prospective product and market. This approach doesn’t seem unreasonable. Which is not the same as reasonable.

This circularity can result in building something that isn’t a product and pitching it to someone who isn’t a market. In September, 2019, in a NYC WeWork conference room at Tabb Group I showed Larry Tabb an idea in the form of an immature web application. Larry’s response was the first sign we could find product market fit.

When you’re building something from nothing, a sign is encouraging.

Since receiving this sign, Point Focal, the portfolio analytics firm I founded, has gone from idea to product to revenue. We are not yet viable and we have not yet failed. A recurring thought of mine is that waiting for the story to end before writing the story would be an error.

Error-correction is learning. This prose is an effort to error-correct. And to learn.

SaaS financial technology is to product market fit as writing online is to publisher subscriber fit. The pub-sub-fit circularity implies you cannot write to an audience that does not exist and you cannot have an audience if you do not write.

• To find portfolio analytics customers, build.

• To find an audience, publish.

Portfolio Reporting and Focus Signal

Portfolio Reporting is the product born from the engine in our analytics stack. It integrates alternative data with custom portfolios to produce quantitative insight. We believe it is the future of institutional research. Automatic, alt-data-driven, custom intelligence. Produced with leverage from code, media, and capital. Mostly code. Some media. And very little capital.

Leverage creates scale. And scale enables a seven person startup to compete with the world.

• The Portfolio Reporting engine is designed to surface insight.

• This Focus Signal Substack has the same purpose.

Content here will span two categories:

1. Portfolio analytics

2. Building a fintech SaaS company

Mark Twain said “fiction must be plausible, truth needn’t be”. What happens at Point Focal could not be made up. Such is the environment when choosing a pursuit with a high failure rate and a constant threat to survival. Each day Point Focal does not fail is a day that can be framed in two ways.

1. A day that in hindsight will appear to have been extending an inevitable demise.

2. And a day closer to viability.

Time will tell.

Point Focal’s portfolio analytics are built for PMs, traders, and analysts in U.S. equity markets. Our content spans market behavior, venues, ESG, earnings estimates, news sentiment, and performance. We make sense of this data and overlay it on portfolios.

To understand something entirely, write about it. I want to understand Point Focal, entirely. If Focus Signal is effective, it will help Point Focal error-correct and learn while it still matters. Following along may help you error-correct and learn too. And it may also be entertaining.

But these second order outcomes and any semblance of an audience, should either materialize, will be the consequence of openly writing for one.