Five green trades in a row. Then six. By the seventh you’re not checking the setup as carefully, the lot size has crept up, and a voice in your head says you’ve finally “got it”. That feeling has a name. The hot hand fallacy in trading is the belief that a run of wins makes the next trade more likely to win, when often the streak is just what randomness looks like up close.
We want to be careful here, because the research on hot hands is more interesting than the usual “it’s all luck” lecture. Streaks are sometimes real. The problem for traders is that we almost never have enough data to tell a real edge from a lucky patch, and the cost of guessing wrong falls on our position size.
In this article
- Where the Hot Hand Fallacy Comes From
- The Twist: Some Streaks Turned Out to Be Real
- Why the Hot Hand Fallacy in Trading Is Worse Than on a Basketball Court
- How Long a Lucky Streak Can Be: A Worked Example
- What the Hot Hand Does to Position Size
- The F&O Angle: Why This Matters in India
- How to Tell a Real Edge From a Hot Streak
- Hot Hand vs Gambler’s Fallacy: Two Sides of One Mistake
- A Simple Routine After Any Winning Run
Where the Hot Hand Fallacy Comes From
The idea started in basketball. A shooter who has made his last few shots is said to have a “hot hand” and is supposedly more likely to score again. In 1985, Thomas Gilovich, Amos Tversky and Robert Vallone tested this in a paper titled “The Hot Hand in Basketball: On the Misperception of Random Sequences”. According to the summary on Wikipedia’s hot hand entry, they ran several studies: a questionnaire of 100 basketball fans from Cornell and Stanford, the shooting records of the 1980-81 Philadelphia 76ers, free-throw data, and a controlled shooting experiment.
Their conclusion was blunt. Hits and misses looked independent of each other, and the feeling of being “hot” didn’t predict the next shot. Fans believed in streaks; the numbers didn’t back them.
The paper also explained why people get this wrong. When people judge a coin toss, they expect even short sequences to come out close to half heads and half tails. Real random sequences are lumpier than that. They have runs. And when we see a run, we tend to decide something other than chance must be behind it. Gilovich offered two explanations: we go looking for streaks and then remember the ones we find (confirmation bias), and we expect chance to alternate more than it actually does, so clusters look meaningful (the clustering illusion).
The Twist: Some Streaks Turned Out to Be Real
For about three decades the hot hand was the textbook example of a cognitive error. Then the statistics got a second look.
In 2018, Joshua Miller and Adam Sanjurjo published a reanalysis of the original 1985 data in the journal Econometrica and found significant evidence of streak shooting in the 76ers’ records. Their point was subtle. If you look at a short, finite sequence and only count the shots that come right after a streak of hits, you introduce a sampling bias. In short sequences, that method naturally shows fewer hits after hits than the shooter’s true average. So a shooter who hits after a streak at the same rate as after a miss may actually be running a little hot.
You can check the direction of this bias yourself with four coin flips. There are 16 possible sequences. In 14 of them at least one of the first three flips is heads. For each of those 14, work out what share of the flips that come straight after a head are also heads, then average those 14 shares. We did this enumeration and the answer is about 40.5%, not 50%, even though the coin is perfectly fair. That’s the small-sample trap Miller and Sanjurjo described, in miniature.
Later work has been mixed. Some studies found hot-hand evidence in specific settings, such as NBA free throws and the NBA three-point contest. Others found none. One 2022 study in PLOS ONE found that only a small subset of players showed a hot hand, and where it existed the effect size tended to be small. Another reanalysis suggested the effect may show up mainly when shots are taken from the same spot again and again.
Even in a sport with clean, repeatable shots and decades of data, researchers argued for thirty years over whether the hot hand exists. A trader with forty trades in a notebook is not going to settle it in a week.
Why the Hot Hand Fallacy in Trading Is Worse Than on a Basketball Court
A free throw is about as controlled as a skill test gets: same distance, same hoop, same ball. Markets give you none of that. The “shot” changes every day. Volatility shifts, a result season comes and goes, liquidity dries up on a holiday-shortened week. A streak of wins in a trending month says very little about how the same rules will do in a choppy one.
There’s also research tying the hot hand directly to how people buy and sell. A 2005 study by Joseph Johnson and colleagues in the Journal of Consumer Research looked at how people responded to stocks with rising or falling earnings trends. The results suggested that buyers, for both short and long trends, and sellers, for short trends, behaved in line with the hot-hand belief. On average, participants bought winners and sold losers. Its mirror image, the gambler’s fallacy, expects a run to reverse rather than continue, and it showed up more in longer sequences.
So the same trader can swing between both mistakes. After three winning trades on a stock, we expect a fourth. After eight up days in an index, we start calling the top because “it can’t keep going”. Both are stories we tell about sequences that may carry no information at all.
How Long a Lucky Streak Can Be: A Worked Example
Here’s the part that surprises most people. Take a strategy with no edge at all, where each trade is a coin flip: a 50% chance of a win. Run it for 100 trades. How likely is it that somewhere in those 100 trades you get at least six wins in a row?
We worked it out exactly, treating each trade as independent. The probability comes to roughly 54.6%. More often than not, a strategy with zero edge will hand you a six-trade winning streak somewhere in a hundred trades. The chance of at least five in a row is about 81%. Even a run of eight shows up around 17% of the time.
Now make the strategy worse. With a 40% win rate, the odds of at least six straight wins in 100 trades are still about 21%. Roughly one trader in five running a losing-rate system would see a streak long enough to feel like genius.
These figures are our own illustrative arithmetic on a simplified model, not results from any real strategy or dataset. Real trades are not perfectly independent coin flips. But the lesson holds: streaks are a normal feature of random sequences, not proof that something has changed.
What the Hot Hand Does to Position Size
The danger isn’t the belief by itself. It’s what we do with it. A trader who feels hot usually does one of three things: raises size, loosens entry rules, or skips the stop because “this one will work too”.
Let’s put rupees on it. Say you trade with a capital of ₹5,00,000 and risk 1% per trade. That’s ₹5,000 at risk each time. After a six-trade streak you decide you’re on fire and double the risk to 2%, or ₹10,000 per trade. If the next three trades lose, the damage is ₹30,000, which is 6% of the starting capital. At your normal 1% it would have been ₹15,000, or 3%. The streak didn’t change the odds of the next three trades. It only changed how much they cost you.
And losses usually come right when size is highest, because size went up because of a good run, and good runs end. This is a close cousin to the house money effect, where recent profits feel like free money to gamble with.
A winning streak doesn’t change the odds of your next trade. It changes how much you’re willing to bet on it, and that’s where the account gets hurt.
The F&O Angle: Why This Matters in India
Derivatives magnify everything above. Leverage makes a raised position size bite harder, and the quick feedback of intraday options trades produces streaks fast. SEBI’s own data gives a sense of the stakes. In a press release dated September 23, 2024, SEBI said 93% of individual traders incurred losses in equity F&O between FY22 and FY24, with aggregate losses exceeding ₹1.8 lakh crore over those three years.
We can’t say how much of that comes from the hot hand specifically; SEBI’s study doesn’t break it down that way. But it does tell us that the typical retail derivatives trader is not running an edge, and when the average outcome is a loss, a winning streak is far more likely to be variance than proof of skill.
How to Tell a Real Edge From a Hot Streak
None of this means you should ignore good results. It means you should judge them properly. A few habits help.
Count trades, not days
Ten winning trades feel like a lot. Statistically they’re very little. Judge a system over a large sample of trades taken with the same rules, and be suspicious of any conclusion drawn from a few dozen.
Compare the streak to what chance would give
Before you celebrate a run, ask what a coin would do. As the worked example showed, a six-win streak is more likely than not over 100 trades even with no edge. If your streak fits comfortably inside what randomness produces, it’s not evidence of anything.
Check whether the conditions were special
Did the streak happen in a strong trending month? During a single sector rally? If the wins cluster in one market type, you may have a strategy that works in that regime and struggles in others. That’s useful to know, but it’s not a hot hand.
Fix position size before the streak, not during it
The cleanest defence is a sizing rule you set in advance and don’t touch after a few wins. If you want size to grow, tie it to account equity on a schedule, not to how you feel after Thursday. Our guide to risk management in trading walks through fixed-fractional sizing in more detail.
Write down why each trade won
A trade journal that records the setup, the entry reason and the exit separates “won because the plan worked” from “won even though I broke my rules”. A streak built on broken rules is a warning sign, not a credential. This links to the illusion of control, the feeling that more screen time and more activity give us more say over random outcomes.
Hot Hand vs Gambler’s Fallacy: Two Sides of One Mistake
It helps to see the two errors side by side. The hot hand fallacy expects a streak to continue. The gambler’s fallacy expects it to reverse, because random numbers “should” balance out. Both come from the same root: believing that a small sample must look like the long-run average, sometimes called the law of small numbers.
In trading, the hot hand pushes us to add size after wins and chase stocks that have run. The gambler’s fallacy pushes us to fade a strong move because it has “gone too far” or to double down after a string of losses because a win is “due”. A disciplined process ignores both and asks only one question: does this trade fit my rules, at my normal size?
A Simple Routine After Any Winning Run
Here’s what we’d do after a run of five or more winners:
- Keep the next trade at your standard risk. No exceptions for “conviction”.
- Re-read the last few trades in your journal and mark any rule you bent.
- Check the market backdrop. If it was a one-way trend, expect the next phase to be different.
- Look at your stats over the full sample, not the streak. Win rate and average win against average loss over many trades tell you far more.
The hot hand isn’t a myth in every setting. The research says small, real streaks can exist in tightly controlled tasks. But markets are not a free-throw line, and a trader’s sample is almost always too small to prove a streak means anything. Treat a winning run as a pleasant result, not a signal. Your position size will thank you.
Disclaimer: This article is for educational purposes only and is not investment advice. The streak probabilities and position-size example use illustrative figures based on a simplified model, not data from any real strategy. Trading in derivatives involves substantial risk. Please consult a SEBI-registered investment adviser before making investment decisions.





Leave a Reply
Join the conversation
Commenting is open to registered readers. Sign up with just your email — no password needed, and it takes under a minute.
Sign up to comment Log in