Speaker 1: Welcome to the deep dive. We're the show where we take your sources, we peel back the layers, and uh really pull out those crucial nuggets of knowledge for you. That's the goal. Today, we're plunging into well, the pretty fascinating world of financial market data. Specifically, how information from futures and options trading helps us really grasp market dynamics. Yeah, it's a complex picture. We've gathered a stack of insights, you know, ranging from those established regulatory reports all the way to some cutting edge analytics and our mission it's to show you how these different data streams well how they can give you a serious edge.

Speaker 2: a real advantage potentially. So what does all this mean for you? Maybe you're prepping for a big meeting or perhaps you're just you know incredibly curious about what actually moves the markets. Let's uh let's unpack this.

Speaker 1: Sounds good. Where do we start?

Speaker 2: Okay, let's start with the bedrock. The foundation of market transparency, at least in the futures world, the commitments of traders reports, the coot reports from the commodity futures trading commission. Right. The CFTC. These have been around for ages.

Speaker 1: Really?

Speaker 2: Yeah. Decades. Designed to shed light on who's doing what in the markets. Make it more transparent for the public.

Speaker 1: Exactly. They give you a breakdown of what's called open interest. Basically, the total number of outstanding futures and options contracts that haven't been settled yet.

Speaker 2: Okay.

Speaker 1: But, and this is key, it's only for markets where you have at least 20 traders holding positions above certain CFTC reporting levels.

Speaker 2: Ah, so there's a threshold,

Speaker 1: right? If a particular contract maybe a specific month drops below that 20 trader mark, it just well it disappears from the report for that period.

Speaker 2: Interesting. So, how does the CFTC actually classify these traders then? And what's the like the core insight we can pull from those classifications?

Speaker 1: Well, it's all based on what the traders tell the CFTC themselves. They fill out a specific form, form 40, about their main business activity.

Speaker 2: Okay. Self-reported.

Speaker 1: Yes. And what's kind of fascinating here, maybe a bit of a limitation, is that the CFTC staff They don't know the specific reasons behind any single position a trader takes. So you might have say like a company classified as a producer merchant process reserve for corn,

Speaker 2: right? Like a farmer or a food company.

Speaker 1: Exactly. All their corn futures positions get lumped together under that label. It doesn't matter if they're hedging next year's crop price or if they're just, you know, taking a punt on prices going up.

Speaker 2: Ah, so hedging and speculation can look the same within a category.

Speaker 1: Precisely. That single classification can hide a mix of motives. It's something you really need to keep in mind when you look at the data.

Speaker 2: Good point.

Speaker 1: Oh, and also confidentiality is a big deal. If there are fewer than four active traders in a specific category for a given commodity, they won't even show the count. It gets suppressed.

Speaker 2: Okay. Protects individual positions. Makes sense. So, we've talked about who the traders are broadly, but understanding how they position themselves is critical, too, right?

Speaker 1: Absolutely.

Speaker 2: Which brings us to this concept called spreading that you see in the reports. What does that actually mean for someone trying to read these numbers?

Speaker 1: Spreading uh it's essentially when a trader holds both long and short positions at the same time, usually in related contracts. Think of it like um balancing things out maybe across different delivery months for the same commodity.

Speaker 2: Okay. Like a hedge within their own book

Speaker 1: sort of. Yeah.

Speaker 2: Yeah.

Speaker 1: Or arbitrage. Because of this, if you just add up all the long positions listed or all the short positions, it won't match the total open interest.

Speaker 2: Exactly. it often won't perfectly match the total hope and interest spreading is one reason and also how option deltas are calculated plays a part too

Speaker 1: right

Speaker 2: it's a key reason why just looking at the raw sums might not give you the full picture you need to account for these balanced offsetting positions

Speaker 1: and the reports help with that

Speaker 2: yes especially the newer ones for instance the disagregated coot reports they clearly break down open interest by long short and spreading for categories like swap dealers and managed money. Gotcha. And you mentioned newer reports. These coot reports, they come in a few different flavors or versions, don't they? Each offering a slightly different view.

Speaker 1: That's right. There are generally four main types you'll encounter. You've got the legacy report. That's the oldest one. Goes way back.

Speaker 2: Simple categories.

Speaker 1: Very simple, just commercial or non-commercial. Basically, hedggers versus speculators. That's the traditional view.

Speaker 2: Okay.

Speaker 1: Then came the supplemental report, I think around 2007. That one focuses just on 13 agricultural commodity contracts and adds a category for index. traders.

Speaker 2: All right. Tracking commodity indices.

Speaker 1: Exactly. Then more recently around 2006, we got the disagregated reports. These cover agriculture, energy like petroleum and natural gas, metals,

Speaker 2: more granular,

Speaker 1: much more. They break traders into categories like producer, merchant, processor, user, swap dealers, managed money, and other reportables. Gives you a better sense of the different types of players.

Speaker 2: And the last one,

Speaker 1: the traders in financial futures or TFF report also from around 2006 that focuses on and financial contracts, currencies, US treasury securities, stuff like that.

Speaker 2: Similar breakdown,

Speaker 1: similar idea. Yeah. Categories like dealer, intermediary, asset manager, institutional, leveraged funds, and other reportables.

Speaker 2: Okay. And I think I saw something about short and long formats too.

Speaker 1: Ah, yes. Each of these reports typically comes in a short format, just the main numbers and a long format. The long one gives you extra details like the concentration of positions held by the largest four and eight traders in each category.

Speaker 2: So, you can see if a few big players dominate

Speaker 1: precisely. It adds another layer of understanding market structure.

Speaker 2: Okay, so we've got these different reports, different breakdowns showing who holds what, but when do we actually get to see them? What's the timing like?

Speaker 1: It's uh pretty rigid, predictable mostly. The data for the COT reports is usually based on positions held at the close of business on Tuesday.

Speaker 2: Tuesday is close. Okay.

Speaker 1: But they don't release the report until Friday afternoon, 330 p.m. Eastern time.

Speaker 2: Friday, so there's a lag.

Speaker 1: Definitely it takes the CFTC those three days Wednesday, Thursday, Friday morning to collect the data from all the reporting firms, process it, check for errors, verify things.

Speaker 2: So it's not real time by any stretch.

Speaker 1: Yeah, not at all. It's a snapshot from Tuesday delivered Friday. T plus3 for processing, T+4 effectively for the public release. And holidays naturally can push that release date back. You always need to check their official schedule for holiday weeks,

Speaker 2: right? So the CFTC gathers all this, verifies it, publishes it. But do they offer any analysis, any guidance on how to actually use this data or are they just the messenger?

Speaker 1: They're very clear on this, purely the messenger. They explicitly state they do not analyze the data themselves or make any recommendations based on it.

Speaker 2: They just publish the information they receive from the reporting firms, which, you know, for many active market participants immediately raises a question,

Speaker 1: if this weekly snapshot based on Tuesday's data isn't timely enough.

Speaker 2: Exactly. If it's not giving you that immediate actionable depth you need in today's markets. Where do you go for that edge?

Speaker 1: And that is a perfect segue because yeah, while those coot reports are foundational, absolutely the financial world today just moves incredibly fast,

Speaker 2: lightning speed sometimes.

Speaker 1: So to get a real edge, a lot of people are looking beyond those weekly snapshots. They need more immediate, more granular data. And this is where it gets really interesting. This is where companies like uh Xtech come into the picture. Yeah. Creating next generation futures analytics,

Speaker 2: right? Xtech, founded by Morgan Slade, is a really prime example of this push for faster, deeper data.

Speaker 1: And Slade has quite the background based on our sources.

Speaker 2: Oh, yeah. Described as a technology visionary, deep knowledge in AI, high performance computing, networking, holds degrees in engineering and finance from MIT.

Speaker 1: MIT. Wow.

Speaker 2: Plus over 25 years of experience at top tier institutional investment firms, places like Citadel, Mel Lynch, he was head of HFT business at one point, head Quat trader

Speaker 1: aviator

Speaker 2: definitely and he's also a successful entrepreneur two previous company exits mentioned his team is packed with experts in real-time data infrastructure quat research and their board even includes Travis Olifant

Speaker 1: the creator of numpai and scypi

Speaker 2: the very same so yeah you're talking about a serious brain trust there focused specifically on tackling these complex financial data challenges

Speaker 1: okay with that kind of firepower what exactly is exto flow and how does it you know fundamentally differ from those CFTC reports we just spent time on Xstack Apollo flow aims to provide well unparalleled insights into global futures trading flows aggregated by investor type similar in concept to CFTC but the execution is different

Speaker 2: how so

Speaker 1: the core innovation as you hinted at is really the frequency and the timeliness unlike the CFTC coot report weekly based on Tuesday out on Friday that T+4 delay we mentioned right

Speaker 2: axt Apollo flow is offered on a daily basis end of day data is delivered by 7 a.m. UTC the next day. So t+1

Speaker 1: daily versus weekly. That's a huge difference.

Speaker 2: It's an absolute gamecher for institutional strategies. It allows for much more reactive decision-m adapting to flows almost as they happen.

Speaker 1: Okay, T+1 is impressive. Where does this daily data actually originate? Is it survey based like CFTC?

Speaker 2: No, that's another key difference. It's described as a completely new data set sourced directly from a clearing system.

Speaker 1: Uh from the clearing house.

Speaker 2: Exactly. Which they claim ensures it's complete. No missing data. No incorrect data critical for reliability when you're trading on it. It's anonymized, then aggregated to characterize daily trading behavior.

Speaker 1: And how much of the market are they capturing? Is it a good sample?

Speaker 2: They boast pretty significant market share. Our sources say up to 45% in some products. That allows for what they call robust sampling of each investor type, giving a pretty high fidelity picture of the overall flow.

Speaker 1: Okay, 45% is substantial. So, how does X-Tec classify its investor types then? Is it the same as the CFTC categories or is something different?

Speaker 2: They use their own proprietary methods. They analyze the trading activity itself to categorize traders into what they call Xtech inferred investor type labels or groups.

Speaker 1: Inferred.

Speaker 2: Yeah. The groups aren't identical to the CFTC's, but they provide equally useful, maybe even more trading focused distinctions. Producers, hedgers, fund managers, broker dealers, and other

Speaker 1: similar concepts, different labels

Speaker 2: pretty much. And they update these group labels annually at the end of the calendar year. They use 5 years of past trading data to build the models, which helps ensure consistency and relevance over time.

Speaker 1: And the confidentiality aspect,

Speaker 2: same principle as CFTC applies. If there aren't enough active accounts in a particular product within one of their investor groups to meet what they call K aggregation requirements for confidentiality,

Speaker 1: they withhold the numbers.

Speaker 2: Exactly. The numeric values for that specific entry are intentionally withheld. Standard practice to protect anonymity.

Speaker 1: Right. Now, our sources mentioned something Beyond just the investor type, these green buffalo versions of Apollo flow. That sounds intriguing. What do those add?

Speaker 2: Ah, yes, the green buffalo versions. So, besides the basic Apollo flow, which gives you daily flows by investor type, they offer versions that integrate specific trading styles. This lets you slice the data even finer, get more granular.

Speaker 1: Trading styles like

Speaker 2: Well, there's position size. This segments each investor group and each product into equal quantiles based on the average position size held. small, medium or large.

Speaker 1: Okay. So, you can see if large fund managers are driving flow versus small ones.

Speaker 2: Precisely. Then there's trade size. Does the same thing but based on the average trade size. Yes. Small, medium, large trades.

Speaker 1: Distinguishing between lots of small trades versus fewer big block trades.

Speaker 2: Exactly. And finally, there's turnover rate. This classifies activity by low, medium, or high turnover within each group and product. Are they frequently flipping positions or holding for longer.

Speaker 1: Wow. Okay, that adds a lot more texture.

Speaker 2: It really does. These style labels, like the main group labels, are also updated annually. They build on those inferred investor types to give you a much deeper, richer understanding of how different types of traders are behaving in the market dayto-day.

Speaker 1: So, let's connect the dots with all this incredibly detailed highfrequency data. Daily updates categorized by investor type, even sliced by position size, trade size, turnover rate, What are the actual practical applications? Why should someone managing money or analyzing markets care about this level of detail?

Speaker 2: Well, the potential uses are pretty extensive. It moves beyond just general sentiment into actionable signals potentially.

Speaker 1: like what? Give us some examples.

Speaker 2: Okay, you could potentially identify critical support and resistance levels, not just on a price chart, but seeing where specific types of investors, say large fund managers, are consistently buying or selling.

Speaker 1: Ah, seeing their lines in the sand. Exactly. You could estimate market impact functions. Who actually moves the price when they trade and by how much? Is it the broker dealers?

Speaker 2: The hedgers understanding who has the market muscle,

Speaker 1: right? You could visualize cross asset flow correlations in near real time. Is money flowing out of bonds and into equities today driven by leveraged funds?

Speaker 2: This data could help pinpoint momentum signals. Are hedgers suddenly so short? Or reversal signals. Are large position holders starting to unwind? You could decode complex spread dynamics. much faster than waiting for the coot report.

Speaker 1: Makes sense.

Speaker 2: You could even overlay it on say statistical arbitrage models. Maybe your model sees a price divergence, but the flow data shows it's driven by temporary dealer hedging.

Speaker 1: So, you know, it might revert quickly rather than being a fundamental shift.

Speaker 2: Exactly. Avoid getting caught on the wrong side of temporary pressure. It might help identify who's behind unexplained maybe curious highfrequency trading patterns. You see

Speaker 1: unmasking the HFTs

Speaker 2: potentially. Potentially. Yeah. And you can analyze concentration risk much more dynamically. Plus, understand things like futures option, short gamma behavior of dealers. That's a big one.

Speaker 1: Short gamma. Explain that briefly.

Speaker 2: Basically, when market makers sell options, they often need to hedge by trading the underlying future. As the price moves, especially near the strike price, the amount they need to hedge, their delta, changes rapidly. That's gamma. If they're short gamma, they have to buy when the market goes up and sell when it goes down, potentially amplifying moves. Seeing dealer flow can give insights into that pressure.

Speaker 1: Okay, that's powerful. So, strategies could emerge from this.

Speaker 2: Absolutely. Maybe you enter a trade after you see a large position change confirmed by a specific investor type in the Xtech data or you combine price thresholds with confirmed position increases from say fund managers. It adds confirmation and timing elements.

Speaker 1: Got it. A much sharper view of the futures landscape. Now, we've covered futures pretty thoroughly, but what about the options market? Options are huge, right? Is Xtech bringing this same kind of det detailed flow analysis there too.

Speaker 2: Yes, they are with a product called Xc equity flow US option flow analytics.

Speaker 1: Equity flow. Yeah,

Speaker 2: this one's based on the OP feed. That's the main feed for US options trades combined with complex Greek calculations.

Speaker 1: The Greeks again delta gamma.

Speaker 2: Exactly. Delta gamma vega for volatility, theta for time decay, row for interest rates. And importantly, these calculations are done on a point in time basis using the market conditions that existed right when the trade happened.

Speaker 1: Crucial for accuracy

Speaker 2: very it applies really intricate knowledge of option market microsstructure basically the nitty-gritty mechanics of how option trades actually happen and orders interact to try and pinpoint the direction of the active risktakers the ones revealing information through their trades

Speaker 1: so figuring out if a trade was buyer initiated or seller initiated essentially

Speaker 2: that's a core part of it yes then this directional information is filtered through the lens of those Greek exposures how much delta risk was traded how much Vega risk etc And it's aggregated on a minutely basis.

Speaker 1: Minutely. Wow. That's granular.

Speaker 2: Extremely. That level of detail makes it potentially very powerful for developing institutional investment strategies, especially shorter term ones.

Speaker 1: Okay. Minutely data on directional option flow broken down by risk exposure. Why is this so powerful? It sounds like it might go beyond just revealing sentiment. Maybe

Speaker 2: this is where according to the sources it gets really interesting. There's evidence suggesting this option flow data can actually predict underlying stock returns.

Speaker 1: Predict stock returns. How?

Speaker 2: Think about how options markets work. When customers could be institutions, could be retail execute option trades. Who takes the other side?

Speaker 1: Usually a market maker,

Speaker 2: right? And what does that market maker do immediately to manage their risk from that option trade?

Speaker 1: They hedge usually by trading the underlying stock.

Speaker 2: Delta hedging.

Speaker 1: Exactly. They perform delta hedging in the stock market. This effectively transfers any imbalance in stock exposure created by the option trans transactions directly into the stock market itself.

Speaker 2: Ah so in foreign option trades create a footprint in the stock market via market maker hedging. Precisely approach as described decomposes the total equity and equity option order imbalance into two parts. The imbalance specifically induced by option transactions and the imbalance just in the underlying equity itself.

Speaker 1: Okay. Separating the two sources of imbalance.

Speaker 2: Yes. And their analysis according to these sources reveals something quite striking. The option induced order imbalance significantly predicts future stock returns.

Speaker 1: Wow.

Speaker 2: While the imbalance that's unrelated to options does not show the same predictive power.

Speaker 1: That's huge. So it's not just general buying or selling pressure in the stock. It's specifically the pressure coming from the options market.

Speaker 2: Correct. And they argue this predictability is driven by a permanent information flow. Meaning informed traders are using options to express their views, not just temporary price pressures that quickly revert.

Speaker 1: That's a really remarkable insight. Opt Options flow predicting stock moves because of information not just noise. Are there specific types of options or maybe market conditions where this predictive power is stronger?

Speaker 2: Absolutely. The sources point to several factors. Firstly, the predictive power seems to come mainly from the delta exposures of at the money ATM and in the money ITM options.

Speaker 1: Not out of the money,

Speaker 2: less so. Out ofthe- money OTM options have very small deltas, right? They're typically used more by institutions looking for volatility exposure. using that high leverage OTM options offer.

Speaker 1: Okay.

Speaker 2: ATM options with moderate delta are often used for directional bets and ITM options they have the highest delta closer to one. They're favored by investors who expect pretty significant price changes and want more certainty even though they offer less leverage than OTMs. So the information seems concentrated in those ATM and ITM trades.

Speaker 1: Makes sense. What else enhances predictability?

Speaker 2: Information asymmetry seems key. The predictive power is stronger for firms where there's likely more information. Think companies with low analyst coverage or wide bides spreads in their stock.

Speaker 1: Fewer eyes on them, more potential for informed trading.

Speaker 2: Exactly. Also, firms with low institutional ownership. This might suggest higher short sale constraints, meaning informed traders who are bearish might prefer to express that view via options rather than shorting the stock directly.

Speaker 1: Interesting workaround.

Speaker 2: And finally, just higher option trading volume in general tends to boost predictive power. Informed traders often prefer deep liquid markets where they can execute large trades without moving the price too much against them.

Speaker 1: So look for high option volume in ATM ITM options, especially in stocks with less coverage or low institutional holdings.

Speaker 2: That seems to be the implication. Yes,

Speaker 1: this sounds incredibly useful for getting ahead of market moves, maybe especially around big known events like reenings announcements.

Speaker 2: Indeed, the sources state that this option induced order imbalance can predict cumulative abnormal returns around earnings announcements. really

Speaker 1: particularly when the earning surprise itself ends up being large or when there was high dispersion among analyst forecasts beforehand suggesting more uncertainty more potential for informed trading.

Speaker 2: So it could give you a real edge during earning season.

Speaker 1: That's the potential. Yes. A strategic advantage in anticipating reactions during those crucial reporting periods.

Speaker 2: Wow. Okay. What a deep dive today. Seriously, we've journeyed all the way from the uh foundational maybe somewhat slower paced CFTC commitments of traders reports

Speaker 1: understanding their classifications, the different types, the lag,

Speaker 2: right? All the way to the rapid super granular insights offered by XEX Apollo flow for futures, daily data, investor types, trading styles,

Speaker 1: and their equity flow for options with that minutely data and predictive potential based on option induced imbalance.

Speaker 2: It really raises a fundamental question, doesn't it? In this world of just information overload,

Speaker 1: constant data streams everywhere.

Speaker 2: Yeah. How critical is it to not just have data, but to have the right data at the right frequency and crucially with the right analytical lens applied to it.

Speaker 1: Absolutely. Understanding investor types, their position styles, their trade sizes, their turnover rates and doing that on a daily basis for futures.

Speaker 2: And then seeing how option flow filtered correctly might actually predict stock returns.

Speaker 1: It fundamentally changes how you can approach market intelligence and strategy development. It moves beyond lagging indicators to potentially leading ones.

Speaker 2: It truly does. So, As you listening reflect on all this, maybe consider just how quickly these market dynamics are evolving.

Speaker 1: The pace of change is incredible.

Speaker 2: What new types of flow data might emerge next? What could give traders an even more precise understanding of real time market sentiment and action?

Speaker 1: And how will things like artificial intelligence and advanced analytics continue to shape this landscape? Not just for the big institutions, but eventually perhaps for all of us.

Speaker 2: Lot to think about there. That's our deep dive for today.

Learn how to subscribe to macro data sources and forecasts: https://www.exponential-tech.ai/

Learn more about XTech Products: https://calendly.com/xtech-marketing-leads/brief-introduction-to-exponential?from=slack&month=2025-08