Three green sessions in a row and suddenly the stock feels unstoppable. Two bad trades on Tuesday and Wednesday and you’re sure your setup has stopped working. That jump, from a handful of recent outcomes to a confident view about what comes next, is recency bias in trading, and it’s probably costing more traders more money than any indicator setting ever has.
This piece looks at what the bias actually is, where it comes from in the psychology research, how it shows up on an Indian trading screen, and a worked example (with made-up numbers) showing how it quietly bends position sizes and exits. Then we get to the part that matters: a few checks that keep the last week in proportion.
In this article
- What recency bias in trading actually means
- Where it comes from: the availability heuristic
- How it shows up on an Indian trading screen
- A worked example: what recency bias does to position sizing
- Why streaks are less meaningful than they look
- Practical checks to keep recency bias in its place
- When recent information genuinely matters
- The bottom line
What recency bias in trading actually means
Psychologists describe recency bias as a cognitive bias that favours recent events over older ones. Wikipedia’s summary of the bias calls it a type of memory bias that gives greater importance to the most recent event. The same entry has a short section on investing, which notes that the bias can keep investors in a rising market long after they should be getting cautious, and keep them away from buying in a falling market because they stay pessimistic about recovery.
Put in trading terms, the last few candles, the last few trades and the last few headlines get far more weight in your head than they deserve. Not because they’re more informative. Just because they’re fresh.
It helps to be clear about what it isn’t. The same Wikipedia entry separates recency bias from anchoring and from confirmation bias. Anchoring is about getting stuck on a particular number, like your buy price. Confirmation bias is about hunting for evidence that agrees with you. Recency bias is about time: whatever happened last feels like the most important thing that ever happened.
Where it comes from: the availability heuristic
The roots go back to work by Amos Tversky and Daniel Kahneman. Their 1974 paper in Science, Judgment under Uncertainty: Heuristics and Biases, described three mental shortcuts people use when judging under uncertainty: representativeness, availability, and adjustment from an anchor. The abstract makes a point every trader should pin above the monitor: these shortcuts are economical and usually effective, but they lead to systematic and predictable errors.
Availability is the one that matters here. According to the availability heuristic entry, it’s the habit of judging how likely or important something is by how easily examples come to mind, and it is inherently biased toward recently acquired information. Tversky and Kahneman first studied and named it in 1973.
Markets are almost designed to feed this. Yesterday’s move is on every business channel. This morning’s gap is in every WhatsApp group. The trade you lost on Friday is the one you replay on Saturday. What happened three years ago in a similar setup? You have to go and dig that out deliberately. So the recent stuff wins by default.
The last five trades are a tiny, noisy sample. Your trading plan was built on hundreds. When the two disagree, the plan deserves the benefit of the doubt.
How it shows up on an Indian trading screen
You don’t need a psychology degree to spot it. A few patterns come up again and again.
- Chasing the hot sector. A theme runs for a few weeks, defence or railways or PSU banks, and it starts to feel like the only place to be. The longer the run, the more natural it feels to buy the next breakout, even as the entries get later and the stops get wider.
- Abandoning a working system after a losing streak. Every rules-based strategy has rough patches. A trader with recency bias treats four or five losses as proof the edge is gone and switches to something new, usually something that has just had a good run, right before the original method recovers.
- Oversizing after a winning streak. A string of wins feels like skill confirmed. Position sizes creep up, the risk per trade drifts past the plan, and the next normal losing trade suddenly does real damage.
- Treating one day’s volatility as the new normal. After a sharp fall, every small dip looks like the start of a crash, so you exit good positions early. After a calm month, you size up as if nothing could go wrong.
- Overreacting to the latest result or headline. One weak quarter or one strong one reshapes the whole story about a company, even when the longer record says something different.
Notice what ties them together. In each case, a short recent window is being treated as if it represents the long run.
A worked example: what recency bias does to position sizing
Let’s make this concrete. The numbers below are purely hypothetical and only illustrate the arithmetic; they don’t describe any real trader, stock or strategy.
Say Meera trades swing setups with a capital of ₹5,00,000. Her plan says she risks 1% of capital per trade, which is ₹5,000. Her back-tested system wins about 45% of the time, with average winners twice the size of average losers. Over a long run of trades, that’s a positive expectancy: per ₹1 risked, she expects roughly 0.45 × 2 − 0.55 × 1 = ₹0.35.
Now look at what happens over a few weeks if she lets recent results steer her.
Weeks one and two: a hot streak. She wins five of her first six trades. That’s far above her 45% long-run rate, and it’s exactly the sort of run a 45% system throws up now and then by chance. But it feels like a breakthrough. She doubles her risk to 2% per trade, ₹10,000.
Week three: normal service resumes. The next five trades go loss, loss, win, loss, loss. Nothing unusual for her system. But at 2% risk, the four losses cost ₹40,000 and the one 2R winner brings back ₹20,000, a net hit of ₹20,000. At her planned 1%, the same sequence would have cost ₹10,000.
Week four: the opposite overreaction. Rattled, she decides the system is broken and cuts risk to 0.5%, ₹2,500, or skips setups entirely. If the next stretch is a good one, she earns a fraction of what the plan would have delivered.
See the trap? She sized up just before an ordinary losing patch and sized down just before an ordinary winning one. Neither change was driven by new information about the system. Both were driven by how the last few trades felt. Over time, this pattern can turn a positive-expectancy system into a flat or losing one, without a single rule of the strategy itself being wrong.
A fixed fractional rule, the same percentage risk every time, doesn’t make you smarter. It just stops the last week from deciding how much you bet. Our pillar guide on risk management in trading walks through the sizing maths in more detail.
Why streaks are less meaningful than they look
Here’s a simple way to see it. If a strategy wins 45% of the time, a run of five or six losses in a row is not rare at all over a long series of trades. Neither is a run of four or five wins. Randomness clumps. It doesn’t alternate politely between wins and losses.
Our brains, though, read clumps as signals. Five wins in a row must mean something. Five losses must mean the market has changed. Sometimes it has. Most of the time, it’s just the normal spread of outcomes for that edge, and the only real way to tell the difference is to compare the recent run against a large sample of past results, not against your mood.
This links neatly to outcome bias in trading, the habit of judging a decision by how it turned out rather than by whether it followed the process. Recency bias makes the latest outcomes loud; outcome bias makes you grade your decisions by them. Together, they can push a perfectly good trader into rewriting a perfectly good plan.
Practical checks to keep recency bias in its place
You won’t switch the bias off. Nobody does. What you can do is build a few speed bumps between the feeling and the order window.
1. Fix your risk per trade in writing
Decide the percentage of capital you’ll risk per trade and write it down. Change it only on a set review date, not mid-streak. If you want to adjust risk with performance, use a mechanical rule decided in advance, such as reviewing size only after a defined number of trades, so the decision isn’t made in the heat of a good or bad week.
2. Judge the system on a big sample
Keep a trade journal with every trade, not just the memorable ones. Before you decide a strategy has stopped working, look at its record over a long stretch and ask: is this losing run outside what the system has produced before? If it isn’t, the run is noise until proven otherwise.
3. Zoom out on the chart before you trade
A simple habit: before entering, look at the weekly chart and a longer daily window, not only the last few sessions. A stock that looks like it’s breaking out on a 15-minute chart may be running straight into a level that has capped it for months.
4. Write the reason before you click
One line is enough: “Buying because the setup matches rule 3, stop below the swing low.” If the honest reason is “it’s been going up all week” or “I’ve lost three in a row, I need this one”, that’s recency bias talking. Wait.
5. Review on a calendar, not on emotion
Pick a fixed review time, say the last trading day of each month, to look at performance and tweak the plan. Changes made after a planned review tend to be better than changes made at the end of a painful Thursday. Our piece on hindsight bias covers why reviewing trades soon after the fact can also distort what you learn from them.
You can’t stop the last few sessions from feeling important. You can stop them from changing your position size.
When recent information genuinely matters
None of this means ignoring what just happened. Markets do change. A company can announce something that genuinely resets its prospects. A regulation change can alter how a product trades. Volatility regimes shift, and a stop that made sense in a calm month may be too tight in a choppy one.
The question to ask is whether the recent information is new information about the situation, or just a recent outcome of the same situation. A fresh SEBI circular on derivatives, a change in a company’s guidance, a sharp shift in a stock’s average daily range: those are new facts, and your plan should account for them. Three losing trades on the same setup you’ve traded for two years? That’s an outcome, and on its own it doesn’t tell you much.
The bottom line
Recency bias is not a character flaw. It’s the brain doing what Tversky and Kahneman described back in the 1970s: using a quick shortcut that usually works and sometimes misleads in predictable ways. In trading, the predictable way is this: we oversize after good runs, give up after bad ones, and chase whatever moved last.
The fix is boring, and that’s the point. Fixed risk per trade. A journal with every trade in it. Decisions about the system made on review dates, using a big sample. If the last few sessions still feel like the whole story after all that, give it a day. The market will still be there tomorrow.
Disclaimer: This article is for educational purposes only and is not investment advice. The worked example uses hypothetical figures. 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