Skip to content
Research / Education / Cognitive bias

Trading Psychology · Part 5 of 20

Overconfidence: why winning streaks are dangerous

A run of wins feels like skill and is frequently luck. How outcome bias distorts self-assessment, and why the largest losses often follow the best weeks.

InnoMP Research Published 31 Aug 2026 · Updated 31 Aug 2026 6 min read
In short

Overconfidence in trading is overestimating the accuracy of your own judgement, usually after a run of wins. Because short-term results are dominated by variance, a winning streak is weak evidence of skill — yet it reliably produces larger position sizes and looser rules. This is why substantial losses often follow a trader's best period rather than their worst.

Overconfidence is overestimating the accuracy of your own judgement. In trading it arrives after wins, which makes it the mirror of the recency bias covered in Part 4 — the same distortion, running in the direction that feels good.

It is more expensive than pessimism, because pessimism makes you trade too little and overconfidence makes you trade too much at exactly the wrong moment.

Why a winning streak proves little

How often four consecutive wins occurs for a 45 percent win rate A row of twenty-five sequence markers, of which one is highlighted to show that four consecutive wins occurs roughly once in twenty-five sequences, labelled uncommon but unremarkable across a year. ← four wins in a row1 sequence in 25feels like the market became readable InnoMP Research
Uncommon, and entirely unremarkable across a year of trading. It feels categorically different from the inside.

Take a strategy that wins 45% of the time. Four consecutive wins happens roughly once in every twenty-five sequences.

That run feels categorically different from the inside. And the response is predictable: size increases, criteria loosen, trades that would have been skipped now look acceptable. The strategy has not changed. Only the confidence has — and it is now attached to larger positions.

This is why the largest single losses in many trading records arrive shortly after the best week rather than during the worst one.

Position size rising through a winning streak, then one oversized loss A bar chart of position outcomes across fifteen trades. The first bars are equal size with mixed results. After four consecutive wins the bars grow progressively larger, and the final oversized bar is a loss deeper than all the preceding gains combined. flatnormal sizefour wins → size growsone loss, oversized InnoMP Research
The strategy did not change. Only the size did — and it changed on the weakest evidence, in time for the trade that gave it all back.

Outcome bias

The mechanism that makes this durable is outcome bias: judging a decision by its result instead of by its quality at the time.

The four combinations of decision quality and outcome A two by two grid. Good decision with a win and bad decision with a loss are unshaded. Bad decision with a win is shaded and labelled rewarded for a mistake. Good decision with a loss is shaded and labelled punished for doing it right. good decisionbad decisionWonLostcorrect process,correct resultrewardedfor a mistakepunishedfor doing it rightcorrect process,correct resultgrade the left column, not the top row InnoMP Research
The two shaded cells are where the damage lives. Left unchecked, they teach the opposite of what you need to know.

A trade taken outside your rules that happens to profit gets filed as skill, and the behaviour is reinforced. A trade taken correctly that loses gets filed as a mistake, and correct behaviour is discouraged.

The tell is size, not speech

Overconfidence rarely announces itself. Very few traders think “I am invincible now.” What happens is quieter.

Four small individually defensible changes compounding into several times the usual risk Four stacked rows each describing a small change: risk raised from one to two percent, a marginal trade taken, the stop widened, and a second position added. A summary bar at the bottom shows the combined exposure as several times the normal level. 1% → 2%“this setup is clearly better”marginal trade”the read feels obvious”wider stop”a shame to be stopped out”second position”both look strong”usual riskcombined exposureno single step looked reckless InnoMP Research
Each is individually defensible. Together the account carries several times its usual risk at the moment the justification is thinnest.

Key takeaway Fix your risk percentage in advance and do not vary it with recent results. Overconfidence acts almost entirely through position size — so a size that does not respond to a winning streak removes most of the damage without requiring you to feel less confident.

Constant risk against risk that responds to recent results Two rows of position sizes across ten trades. The first row keeps every bar the same height regardless of wins and losses. The second row grows the bars after wins and shrinks them after losses, with the largest bar landing on a loss. Fixed riskthe streak changes nothingRisk by feelthe loss lands here InnoMP Research
Overconfidence acts almost entirely through size. A size that does not respond to a winning streak removes most of the damage.

Practical defences

Constant risk per trade. Whatever your percentage, it does not move with recent performance. Use the calculator every time rather than sizing by feel — the arithmetic does not get overconfident.

Grade the process, not the result. In your journal, record whether each trade met your written criteria as a separate field from whether it profited. A rules-compliant loss is a good trade; a rule-breaking win is a bad one that got lucky.

Notice the feeling as a signal. The sense that the market has become readable is itself information — reliably, that variance has been kind. Treat it as a prompt to check your recent position sizes rather than as a reason to increase them.

Set a size ceiling in advance. A maximum risk per trade and a maximum total exposure, written down while calm. Overconfidence negotiates well in the moment; it cannot negotiate with a number you wrote last month.

Next: Fear in trading — the emotion that keeps traders out of good trades and, occasionally, saves them.

Key facts
  • Short-run results are dominated by variance, so a winning streak is weak evidence of skill.
  • Outcome bias is judging a decision by its result rather than by the quality of the decision.
  • Overconfidence typically expresses itself as increased position size rather than as a stated belief.
  • A fixed risk percentage removes the main channel through which overconfidence causes damage.

Frequently asked questions

Why are winning streaks dangerous in trading?

Because they feel like evidence of skill when they are usually variance. The response — larger positions, looser rules, more trades — increases risk exactly when the evidence for doing so is weakest.

What is outcome bias?

Judging a decision by how it turned out rather than by whether it was sound when made. A reckless trade that happened to profit gets recorded as a good decision, which reinforces the behaviour that will eventually be costly.

How do I know if my results are skill or luck?

Sample size and process. Below roughly 100 trades, results are largely variance. Beyond count, ask whether each winning trade met your written criteria — a win from a setup you did not plan is not evidence of skill.

How do I protect against overconfidence?

Fix your risk percentage in advance and do not vary it with recent results. Overconfidence almost always acts through position size, so removing size as a discretionary variable removes most of the damage.

InnoMP Research

Market research, trading education and platform guides from the InnoMP research desk — covering forex, metals, indices and stock CFDs.

Published 31 Aug 2026 · Updated 31 Aug 2026 · Reviewed by InnoMP Compliance

Disclaimer: This content is provided for general informational purposes only and does not constitute investment advice, an offer, or a solicitation to buy or sell any financial instrument. It has been prepared without regard to your individual financial circumstances or objectives. Trading CFDs involves a high risk of loss.

Related notes