Get live statistics and analysis of Concretum Research's profile on X / Twitter

How would you feel if you spent fifty years studying stats maths coding geopolitics accounting econometrics microeconomics macroeconomics technical analysis fundamental analysis only to finally discover that to consistently make money in trading, it would have been sufficient

While writing the introduction to our upcoming paper “The Volatility Edge” (coming out in the next few weeks), I stumbled upon this gem: an interview by the AQR team with the father of quant trading, Edward Thorp. I wasn’t aware of this piece before, and I strongly recommend giving it a read. Thorp shares timeless insights on option mispricings, volatility trading, dynamic hedging, and systematic investing... all decades ahead of his time. You can find the link in the comments

What exactly is the "Intraday Index Manipulation" strategy that Jane Street allegedly deployed in India’s markets—and why has SEBI, the country’s market regulator, accused the firm of generating over $500 million in profits through it? Over the past few days, I’ve received many messages asking about the mechanics behind this case. So I decided to write a short explainer that hopefully sheds light on how the strategy worked and why it triggered regulatory action. I was also lucky to discuss JS strategy with Eugenio Mazzetti, a great equity derivatives quant, whose insights helped me piece together how this trade may have unfolded Here’s what allegedly happened: On options expiry days, Jane Street would begin the session by buying large quantities of Bank Nifty stocks and futures, putting upward pressure on the index. At the same time, they accumulated a massive portfolio of put options—which gain value if the index falls. The early rally made these puts appear cheaper, allowing Jane Street to build its position at favorable prices. To finance the put purchases, Jane Street simultaneously sold call options, which had become overpriced due to the index’s sharp rise. This combination—a long put and short call—is economically equivalent to a synthetic short position. Here’s where it gets interesting: most traders in Bank Nifty options, especially retail investors, don’t participate in the underlying stock or futures markets. They trade options based solely on the index level, without understanding what’s moving it. So when Jane Street aggressively pushed the index higher, it created a false sense of bullish momentum, luring retail participants into trading options at prices distorted by this artificial move. SEBI argues that many of these unsophisticated investors were willing sellers of put options, not realizing the market had been temporarily inflated. Since retail traders typically don’t hedge their delta exposure like institutional dealers do, their option selling had no countervailing impact on the value of the index. This allowed Jane Street to continue buying puts without making them more expensive—an unusually efficient execution for such a large trade. Later in the day, Jane Street reversed its futures and stock positions, causing the index to fall. The puts they had accumulated surged in value, while the calls they had sold expired worthless or lost value. The result: huge profits on the options, with only minor losses in the cash market. According to SEBI, the notional size of Jane Street’s options exposure was >10x larger than their exposure in the cash market. A helpful illustration from SEBI (shown below) highlights the strategy’s mechanics. The blue line shows the cumulative P&L realized in January 2024, plotted alongside net buying/selling activity in put and call options as well as index futures. It visually captures how profits were generated through coordinated moves across both the derivatives and cash markets. To put this in broader context: India accounted for 60% of global equity derivatives trading volume in May 2025, according to the Futures Industry Association. At the same time, retail investors in India lost over $12.4 billion trading equity derivatives in the last financial year—a 41% increase from the previous year. While not all of these losses can be directly attributed to manipulation, it's highly likely that strategies like the one allegedly used by Jane Street played a meaningful role. SEBI says this wasn’t hedging or arbitrage—it was a coordinated effort to move the market and profit from it. Jane Street disagrees and claims it was a legitimate trading strategy. Interestingly, Jane Street recently sued Millennium Management, alleging that Millennium stole the rules behind an Indian options strategy by hiring two former Jane Street traders. So perhaps Jane Street wasn’t the only major player exploiting this intraday index manipulation framework.

🚫 Stop Chasing the Perfect Strategy 🚫 I often receive messages from readers searching for the perfect set of parameters to optimize a trading strategy. The truth is that adding more filters and tweaking parameters usually improves the in-sample Sharpe ratio—but almost always leads to much worse out-of-sample results. So my sincere advice to all aspiring quants is this: don’t waste time chasing the holy grail in a single strategy. Instead, focus on combining multiple simple strategies and aim to capture the essence of the edge, not its most fragile historical outliers. As highlighted in a recent research by MAN AHL, there is significant dispersion of performance among leading CTA funds each year, often due to parameter design choices. Fast trend signals may thrive in crisis periods but lag in calmer markets. Trying to forecast which parameter set will dominate next has proven challenging even for the largest, most experienced funds. A better approach is to build an ensemble of trading strategies, even within the same style. For example, in the intraday trend-following space, we are currently adding a new simple trend model to our portfolio, which already includes seven different intraday solutions. These eight strategies differentiate themselves by the assets traded, rebalancing frequency, entry and exit rules, and trailing-stop mechanisms. Within this bucket, we don’t expect strategies to be uncorrelated (after all, we want to be fully long/short when strong intraday trends emerge). But with an average pairwise correlation of around 0.50, we managed to: ✅ Improve the aggregate Sharpe Ratio of the portfolio from 1.06 to 1.50 ✅ Reduce the risk of being overly concentrated in a single strategy that might lose its edge Attached you’ll find three charts: 1️⃣ Correlation matrix showing pairwise correlations among strategies 2️⃣ Rank board: each year strategies are sorted by return, with fixed colors per strategy—notice how some of the worst performers in one year often become top performers the next 3️⃣ Equity curves of each strategy, along with the combined portfolio (black line) Bottom Line: Long-term success in systematic trading doesn’t come from chasing the “perfect” strategy, but from building robust portfolios of simple, complementary edges.

When constructing a portfolio made up of systematic strategies, incorporating uncorrelated—and ideally orthogonal—trading systems can significantly improve the risk-reward profile. Intraday momentum and short volatility systems are straightforward yet effective examples.

**** My Favorite Novels On Quant Trading **** After six months of intense research and trading, I’m finally taking a well-deserved break in stunning Sardinia. While waiting for the next kite session, I want to share my favorite novels on quant trading—perfect reads for a vacation! These books are great for anyone passionate about algorithmic trading, offering inspiration without heavy technical jargon. Enjoy these reads, and feel free to suggest more titles!

I often hear traders rejecting volatility sizing approaches because they dislike using standard deviation as a proxy for risk. But this criticism is usually based on a flawed assumption: 🚫Volatility = Standard Deviation 🚫 In reality, standard deviation is just one way to estimate volatility... and volatility itself is a latent variable: it cannot be observed directly and must always be estimated. There are many alternative estimators — some parametric (such as GARCH, EWMA, ARCH, HAR), others non-parametric (like ATR, realized volatility (RV), or range-based methods). Some rely solely on daily closing prices, while others make use of high-frequency or intraday data. Among trend followers, one of the most widely adopted alternatives is the Average True Range (ATR) — easily expressed also in percentage terms. ATR is non-parametric, intuitive, and doesn’t rely on distributional assumptions. It captures total price movement, not just close-to-close returns, making it a practical and often more realistic proxy for realized risk. In our experience, the most effective tools to capture time-varying volatility combine intraday data with simple math or regression-based models. While these techniques can’t always be applied to long historical backtests due to data limitations, they are highly practical for real-time implementation. 🔍 Bottom line: we can debate which method best captures risk and how often that risk should be re-estimated but if you're scaling positions based on ATR, you're still doing volatility sizing, just through a different estimation lens. 📄To explore this topic further, check out Appendix A of our latest research paper "The Volatility Edge", where we highlight recommended resources on volatility estimation. 👇 Link to the paper in the comments

Systematic trading is often considered as an approach to trading that is free from emotion. While it can greatly reduce the emotional rollercoaster faced by active discretionary day traders, quantitative trading also demands considerable self-discipline and control. Simulating a 20-year backtest in just a matter of seconds can instill a sense of impatience and invincibility in quants. This overconfidence may become a liability when a strategy goes into production, as the quant then faces the slow progression of time and the temporary drawdowns that the strategy might encounter. To address this issue, when we explore new trading strategies, we create an animated version of the backtest. The performance trajectory is revealed gradually, helping us prepare emotionally for the daily volatility our portfolio might face once it goes live. The animation attached comes from a volatility trading program we developed in partnership with Prof. Antonio Mele, the brilliant mind behind fixed-income VIX indexes. Although historical tests were impressive, our nerves would have been tested during the inevitable drawdowns. Looking ahead, we plan to co-publish with Prof. Antonio Mele a practical paper on volatility trading in the summer of 2025. This resource will offer traders and speculators an accessible guide to the fundamentals of volatility markets, complete with real-world insights gleaned from our own research and experience

How would you feel if you spent fifty years studying stats maths coding geopolitics accounting econometrics microeconomics macroeconomics technical analysis fundamental analysis only to finally discover that to consistently make money in trading, it would have been sufficient

🚫 Stop Chasing the Perfect Strategy 🚫 I often receive messages from readers searching for the perfect set of parameters to optimize a trading strategy. The truth is that adding more filters and tweaking parameters usually improves the in-sample Sharpe ratio—but almost always leads to much worse out-of-sample results. So my sincere advice to all aspiring quants is this: don’t waste time chasing the holy grail in a single strategy. Instead, focus on combining multiple simple strategies and aim to capture the essence of the edge, not its most fragile historical outliers. As highlighted in a recent research by MAN AHL, there is significant dispersion of performance among leading CTA funds each year, often due to parameter design choices. Fast trend signals may thrive in crisis periods but lag in calmer markets. Trying to forecast which parameter set will dominate next has proven challenging even for the largest, most experienced funds. A better approach is to build an ensemble of trading strategies, even within the same style. For example, in the intraday trend-following space, we are currently adding a new simple trend model to our portfolio, which already includes seven different intraday solutions. These eight strategies differentiate themselves by the assets traded, rebalancing frequency, entry and exit rules, and trailing-stop mechanisms. Within this bucket, we don’t expect strategies to be uncorrelated (after all, we want to be fully long/short when strong intraday trends emerge). But with an average pairwise correlation of around 0.50, we managed to: ✅ Improve the aggregate Sharpe Ratio of the portfolio from 1.06 to 1.50 ✅ Reduce the risk of being overly concentrated in a single strategy that might lose its edge Attached you’ll find three charts: 1️⃣ Correlation matrix showing pairwise correlations among strategies 2️⃣ Rank board: each year strategies are sorted by return, with fixed colors per strategy—notice how some of the worst performers in one year often become top performers the next 3️⃣ Equity curves of each strategy, along with the combined portfolio (black line) Bottom Line: Long-term success in systematic trading doesn’t come from chasing the “perfect” strategy, but from building robust portfolios of simple, complementary edges.

**** My Favorite Novels On Quant Trading **** After six months of intense research and trading, I’m finally taking a well-deserved break in stunning Sardinia. While waiting for the next kite session, I want to share my favorite novels on quant trading—perfect reads for a vacation! These books are great for anyone passionate about algorithmic trading, offering inspiration without heavy technical jargon. Enjoy these reads, and feel free to suggest more titles!

When constructing a portfolio made up of systematic strategies, incorporating uncorrelated—and ideally orthogonal—trading systems can significantly improve the risk-reward profile. Intraday momentum and short volatility systems are straightforward yet effective examples.

What exactly is the "Intraday Index Manipulation" strategy that Jane Street allegedly deployed in India’s markets—and why has SEBI, the country’s market regulator, accused the firm of generating over $500 million in profits through it? Over the past few days, I’ve received many messages asking about the mechanics behind this case. So I decided to write a short explainer that hopefully sheds light on how the strategy worked and why it triggered regulatory action. I was also lucky to discuss JS strategy with Eugenio Mazzetti, a great equity derivatives quant, whose insights helped me piece together how this trade may have unfolded Here’s what allegedly happened: On options expiry days, Jane Street would begin the session by buying large quantities of Bank Nifty stocks and futures, putting upward pressure on the index. At the same time, they accumulated a massive portfolio of put options—which gain value if the index falls. The early rally made these puts appear cheaper, allowing Jane Street to build its position at favorable prices. To finance the put purchases, Jane Street simultaneously sold call options, which had become overpriced due to the index’s sharp rise. This combination—a long put and short call—is economically equivalent to a synthetic short position. Here’s where it gets interesting: most traders in Bank Nifty options, especially retail investors, don’t participate in the underlying stock or futures markets. They trade options based solely on the index level, without understanding what’s moving it. So when Jane Street aggressively pushed the index higher, it created a false sense of bullish momentum, luring retail participants into trading options at prices distorted by this artificial move. SEBI argues that many of these unsophisticated investors were willing sellers of put options, not realizing the market had been temporarily inflated. Since retail traders typically don’t hedge their delta exposure like institutional dealers do, their option selling had no countervailing impact on the value of the index. This allowed Jane Street to continue buying puts without making them more expensive—an unusually efficient execution for such a large trade. Later in the day, Jane Street reversed its futures and stock positions, causing the index to fall. The puts they had accumulated surged in value, while the calls they had sold expired worthless or lost value. The result: huge profits on the options, with only minor losses in the cash market. According to SEBI, the notional size of Jane Street’s options exposure was >10x larger than their exposure in the cash market. A helpful illustration from SEBI (shown below) highlights the strategy’s mechanics. The blue line shows the cumulative P&L realized in January 2024, plotted alongside net buying/selling activity in put and call options as well as index futures. It visually captures how profits were generated through coordinated moves across both the derivatives and cash markets. To put this in broader context: India accounted for 60% of global equity derivatives trading volume in May 2025, according to the Futures Industry Association. At the same time, retail investors in India lost over $12.4 billion trading equity derivatives in the last financial year—a 41% increase from the previous year. While not all of these losses can be directly attributed to manipulation, it's highly likely that strategies like the one allegedly used by Jane Street played a meaningful role. SEBI says this wasn’t hedging or arbitrage—it was a coordinated effort to move the market and profit from it. Jane Street disagrees and claims it was a legitimate trading strategy. Interestingly, Jane Street recently sued Millennium Management, alleging that Millennium stole the rules behind an Indian options strategy by hiring two former Jane Street traders. So perhaps Jane Street wasn’t the only major player exploiting this intraday index manipulation framework.

In a room of 35 people, what is the probability that at least two people share the same birthday?

What are the best podcasts you’ve listened to lately? I’m driving down to Chia, in the south of Sardinia, and I’d love to fill the journey with some fresh, engaging content. If something really caught your attention recently, I’d love to hear it. Thanks in advance for your recommendations

While writing the introduction to our upcoming paper “The Volatility Edge” (coming out in the next few weeks), I stumbled upon this gem: an interview by the AQR team with the father of quant trading, Edward Thorp. I wasn’t aware of this piece before, and I strongly recommend giving it a read. Thorp shares timeless insights on option mispricings, volatility trading, dynamic hedging, and systematic investing... all decades ahead of his time. You can find the link in the comments

Over the years, I’ve come across many trading strategies that use the VIX Index to time short-term trades on SPY or VIX-related ETNs. But there is a subtle and dangerous issue that affects many of these backtests: look-ahead bias. 🔎 Here’s the catch... The daily closing price of the VIX Index is published at 4:15 PM, while SPY, VXX, and most US-listed ETFs officially close at 4:00 PM. This means if you are using daily VIX data to generate signals and applying them to ETFs at the same day's close, you are unintentionally using future information, a classic look-ahead bias. This may seem like a small detail, but the impact can be huge, especially if you are building momentum systems with a very short-term horizon. ✅ The best way to avoid this: Use intraday VIX data to retrieve the actual value of the VIX at 4:00 PM; or, even better, compute your signal around 3:45 PM and assume execution with a Market-on-Close order. If you do not have access to intraday data, the safer approach is to generate your signal using VIX daily close and assume trade execution at the next day's open. The chart below shows just how much performance can be inflated when this bias creeps in. Always double-check your timestamps.... The devil is in the details!