Overconfidence bias in trading rarely shows up on a bad day. It shows up after a good week. Five winners in a row, a screenshot you almost posted, and suddenly the 1% risk rule you swore by feels a little timid. We have all met this version of ourselves. This guide explains what the bias actually is, what the research found when it looked at real brokerage accounts, and how to put numbers around your own confidence so a hot streak doesn’t decide your position size for you.
In this article
- What Overconfidence Bias in Trading Really Means
- What the Brokerage Data Showed
- Worked Example 1: The Hot Streak That Resizes Your Risk
- Worked Example 2: How Much of a Streak Is Just Luck?
- Worked Example 3: A Calibration Check From Your Own Journal
- Signs the Bias Is Already Steering You
- Practical Fixes That Actually Hold Up
- Is Some Confidence Useful?
- What We Take Away
What Overconfidence Bias in Trading Really Means
Psychologists don’t treat overconfidence as one thing. The research summary on the overconfidence effect describes three distinct forms, and each one maps neatly onto a trading habit:
- Overestimation: thinking your actual performance is better than it is. In trading terms, remembering your win rate as 65% when the journal says 48%.
- Overplacement: believing you are better than other people. The trader who assumes the person on the other side of the trade is the “dumb money”.
- Overprecision: excessive certainty that your beliefs are accurate. This is the target price you’d bet the house on, the stop-loss you don’t place because “it won’t go there”.
The same summary cites Ola Svenson’s well-known 1981 finding that 93% of American drivers in his sample rated themselves better than the median driver. Statistically, that can’t be true for everyone. Markets are no different, except the bill arrives in your trading account instead of a traffic fine.
One distinction is useful before we go further. Overconfidence is a belief problem: you think you know more than you do. Its cousins on this site are mostly about how you treat money after a result. The house money effect is about risking profits more loosely because they feel like a bonus, and the hot hand fallacy is about expecting a streak to continue. In practice they feed each other, and a winning run is where all three tend to meet.
What the Brokerage Data Showed
The most cited evidence comes from two papers by Brad Barber and Terrance Odean, both built on account data from a large US discount broker in the 1990s.
In “Trading Is Hazardous to Your Wealth”, published in the Journal of Finance in April 2000, they studied 66,465 households between 1991 and 1996. The households that traded most earned an annual return of 11.4%, while the market returned 17.9%. The average household earned 16.4% and turned over 75% of its portfolio every year. The striking part is what happened before costs: the paper reports very little difference in gross performance between frequent and infrequent traders. The gap opened up after trading costs. Frequent traders, defined as households with monthly turnover above 8.8%, earned a net 11.4% a year against 18.5% for those who traded infrequently, and the heaviest traders turned over more than 250% of their portfolio annually. The authors argue that overconfidence can explain both the high trading and the poor results.
Their follow-up, “Boys Will Be Boys”, split more than 35,000 households by gender, on the basis that psychological research finds men more overconfident than women in areas such as finance. Men traded 45% more than women. Trading cut men’s net returns by 2.65 percentage points a year, against 1.72 percentage points for women.
The traders who traded most weren’t worse at picking stocks before costs. They were worse at knowing when not to trade.
Closer home, SEBI’s own numbers tell a story that overconfidence fits uncomfortably well. Its September 2024 study release found that 93% of over 1 crore individual equity F&O traders made losses over FY22 to FY24, with average losses of around ₹2 lakh per trader including transaction costs. Aggregate losses of individual traders exceeded ₹1.8 lakh crore over those three years. Only 1% of individual traders earned profits above ₹1 lakh after costs. And the line that matters most for this topic: more than 75% of loss-making traders kept trading F&O despite consecutive years of losses.
SEBI’s study doesn’t diagnose why people kept going, and we won’t pretend it does. But continuing after repeated losses is exactly what you’d expect from people who believe their next year will be different from their last two.
Worked Example 1: The Hot Streak That Resizes Your Risk
Let’s put numbers on the scenario from the opening. These figures are illustrative, not from any real account.
A trader starts with ₹5,00,000 and risks 1% of capital per trade, so ₹5,000. Each setup has a stop 2% below entry, which means a position of ₹2,50,000 (₹5,000 divided by 2%). Targets are set at twice the risk, so a winner makes ₹10,000.
Five winners in a row add ₹50,000. Capital is now ₹5,50,000. So far, so good.
Now the bias kicks in. Feeling “in sync” with the market, the trader quadruples risk to 4% of capital per trade. Compare what three straight losses do under each approach, with risk recalculated on the shrinking balance each time:
| Approach | Risk per trade | Calculation | Capital after 3 losses |
|---|---|---|---|
| Stays at 1% | 1% of current capital | ₹5,50,000 × 0.99 × 0.99 × 0.99 | about ₹5,33,664 |
| Sizes up to 4% | 4% of current capital | ₹5,50,000 × 0.96 × 0.96 × 0.96 | about ₹4,86,605 |
The disciplined trader is still about ₹33,664 above the starting ₹5,00,000. The overconfident one has given back the whole streak and sits about ₹13,395 below where they began. Three losses in a row isn’t bad luck in any unusual sense; for a strategy that wins half the time, it’s routine. That is the real damage of overconfidence: it doesn’t just make you wrong, it makes you big right before you’re wrong. Our guide to maximum drawdown in trading shows how these dips compound over longer runs.
Worked Example 2: How Much of a Streak Is Just Luck?
Here is a simple calculation we find sobering. Suppose a strategy has no edge at all and every trade is a 50-50 coin flip. The chance of five wins in a row is 0.5 multiplied by itself five times, which is 0.03125, or about 3.1%.
That sounds small until you think about how many people are trading. Out of every 1,000 traders running a strategy with zero edge, roughly 31 will string together five straight winners purely by chance. Each of them will feel skilled. Some will write threads about it.
Give the strategy a modest real edge, say a 55% win rate, and five in a row becomes 0.55 to the fifth power, about 5.0%. Notice the gap: a genuine edge barely changes how often a five-trade streak happens. A short streak tells you almost nothing about whether you have an edge. You need a much larger sample, logged honestly, before your confidence deserves a bigger position.
A five-trade winning streak is about as informative as a coin landing heads five times. It happens to someone every day.
Worked Example 3: A Calibration Check From Your Own Journal
The best antidote to overprecision is to measure it. Before each trade, write down how confident you are on a simple scale, say 60%, 70% or 80%. After 40 or 50 trades, group them and compare stated confidence with what actually happened.
An illustrative journal might look like this:
| Stated confidence | Trades | Winners | Actual win rate | Gap |
|---|---|---|---|---|
| 60% | 15 | 8 | about 53% | 7 points too high |
| 70% | 15 | 8 | about 53% | 17 points too high |
| 80% | 20 | 11 | 55% | 25 points too high |
Look at what this hypothetical trader learns. Their “high conviction” trades win at roughly the same rate as their ordinary ones. The extra confidence carries no extra information, and the gap is widest exactly where they feel surest. If they had been sizing up on the 80% trades, they were paying more for the same odds. Once you see your own table, it’s hard to unsee.
A word of caution: 15 or 20 trades in a bucket is still a small sample, so treat the result as a warning light rather than proof. The habit matters more than any single table.
Signs the Bias Is Already Steering You
Overconfidence is easy to spot in other people and nearly invisible in yourself. A few practical tells:
- Your size changes with your mood, not your rules. If position size after a good week differs from position size after a flat week, and nothing in your written plan says it should, that’s the bias.
- You trade more after wins. Barber and Odean’s core finding was about turnover. Count your trades per week and see whether the number jumps after a profitable stretch.
- Stops get wider or vanish. Overprecision says “it won’t hit my stop”, so the stop moves further away or becomes a mental one.
- You stop journaling the losers. Skipped entries are a quiet way of overestimating your own record. The hindsight bias makes this worse, because the losers start to look like obvious mistakes you’d “never make again”.
- You move from cash to F&O because cash feels slow. Given SEBI’s loss figures for individual F&O traders, that switch deserves more evidence than a good month.
Practical Fixes That Actually Hold Up
Willpower is a poor defence against a bias you can’t feel. Rules that run whether you feel confident or not work better.
- Fix risk per trade in writing. Decide a percentage of capital, write it at the top of your plan, and calculate position size from the stop distance every single time, as in Example 1. Our pillar guide on risk management in trading covers the mechanics.
- Allow size increases only on a schedule. For instance, review size once a month or once a quarter against your journal, never in the middle of a streak. A rule like “no change in risk per trade until the next review date” removes the decision from the moment you’re most likely to get it wrong.
- Keep a trade-count limit. Since excess trading was where the research found the damage, cap trades per day or week. If you hit the cap, you’re done.
- Log confidence before the trade. Run the calibration table from Example 3 every few months.
- Track costs separately. Barber and Odean found the performance gap came after costs, and SEBI’s release puts individuals’ F&O transaction costs at about ₹50,000 crore over FY22 to FY24. Seeing brokerage, exchange fees and taxes as a separate line item makes the true cost of “one more trade” visible.
- Write the bear case. Before entry, write one or two sentences on what would prove you wrong. If you can’t, your certainty is probably overprecision.
Is Some Confidence Useful?
Yes. A trader with no confidence never pulls the trigger, or exits every position at the first wobble. The goal isn’t to feel less sure. It’s to make sure your sureness is earned by a large, honest sample and that it never gets to change your risk on the fly. Confidence can decide whether you take a trade that fits your plan. It shouldn’t decide how much you put on it.
What We Take Away
Overconfidence bias in trading is quiet, flattering and expensive. The research on real accounts points to the same pattern: the people who trade most do worst once costs are counted, and India’s F&O data shows most individual traders lose and many keep going regardless. Short streaks prove little, as the coin-flip maths shows, and a confidence journal will usually tell you your “sure things” aren’t much surer than anything else. Fix risk in writing, change size only on a schedule, cap your trade count, and let the journal, not the last five trades, tell you how good you are.
Disclaimer: This article is for educational purposes only and is not investment advice. Worked examples use illustrative figures and do not refer to any real account or security. Trading in equities and derivatives involves risk of loss. Please do your own research or consult a SEBI-registered investment adviser before making any investment decision.





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