Why Winning Streaks Make Traders Reckless: The Overconfidence Feedback Loop

By Sofia Harchich | Trading Psychologist & Behavioural Finance Writer

Overconfidence after a winning streak is the one trading-psychology problem that arrives disguised as evidence. The account is up, the last four calls were right, and the conclusion — that the read has improved — is the most reasonable thing available. It is also the mechanism that gives most of the streak back.

The damage is rarely done by a single reckless trade. It is done by a gradual loosening that no individual decision would have survived if it had been written down: slightly larger, slightly earlier, slightly looser on the criteria, one more position than the plan allowed.

What follows is the loop, stage by stage, and the two numbers that make it visible in a record before the drawdown does.

What overconfidence after a winning streak leaves in the account

Overconfidence after a winning streak shows up in the five trades that followed your three best weeks. The pattern, when it is present, is consistent enough to be diagnostic.

Position size rises, usually by a third or more, and usually without a decision that felt like a decision. The setup quality that qualifies drops — trades appear that would not have been taken a fortnight earlier. Trade frequency increases. And the interval between an idea and its execution shortens, because checking feels less necessary when checking has recently been rewarded.

None of that is visible as recklessness from the inside. From the inside it reads as conviction, and conviction is what every trading book told you to develop.

Why overconfidence after a winning streak is not arrogance

Treating this as a character flaw is the reason the standard advice fails. Overconfidence after a winning streak is an updating error, and it runs in people who are not otherwise overconfident about anything.

Self-attribution bias is the engine: outcomes that go your way are attributed to skill, outcomes that go against you to conditions. Four winning trades therefore do not update your model by four data points — they update your estimate of yourself. The behavioural finance literature documents this across investor populations, and the effect is strongest exactly where feedback is noisy, which describes every market.

Recency compounds it. The most available evidence is the most recent evidence, and the most recent evidence is a run of wins. A trader inside overconfidence after a winning streak is not ignoring their record. They are weighting the last fortnight of it as though it were the whole thing.

The uncomfortable part is that the streak may genuinely contain skill. The loop does not require the wins to be luck. It requires only that the size of the update exceed the size of the evidence.

The four stages of overconfidence after a winning streak

Named separately, the loop becomes something you can interrupt at a specific point rather than resolving to be more humble.

Stage one: the result arrives. A sequence of wins, close enough together that they read as a run rather than as separate events.

Stage two: attribution. The run is assigned to judgement rather than to conditions. This is the stage where intervention is cheapest and almost never happens, because nothing has gone wrong yet.

Stage three: the threshold moves. Criteria that were written as requirements become preferences. The trade that “nearly” qualifies now qualifies, and the reasoning is sound in each individual case.

Stage four: exposure rises to match the new confidence. Size increases, or frequency does, or both. The account is now carrying more risk on looser criteria than it was a fortnight ago, and nothing in the process announced the change.

The give-back trade is usually blamed on stage four. It was decided at stage two.

A winning streak feeds the instability loop as surely as a losing one; it simply feels like competence while it does. Each win makes the next trigger look safer, and the looser criteria that follow are the behaviour the loop was always going to produce.

The two numbers that expose overconfidence after a winning streak

Both measure overconfidence after a winning streak from a record you already keep, and neither requires you to change how you trade.

The first is planned risk against actual risk, trade by trade, across the twenty entries following your best week. If actual risk drifts upward while planned risk stays flat, size is moving without a decision.

The second is the proportion of entries that met every written criterion, compared between a flat month and the fortnight after a good run. A drop of ten points or more is the threshold moving, and it is the earlier of the two signals.

Traders who run only one of these should run the second. Size creep is easier to see and later in the sequence; criteria drift is the actual point of failure.

What to run after your next good week

One rule, one column, six weeks.

The rule: size does not change inside a winning streak. Not upward during it, not downward after it. Any change to the size band happens on a scheduled review date, from the full record, when nothing in particular is happening — which is the only condition under which the decision is being made by the part of you that wrote the plan.

The column: beside every entry, a single mark for whether every written criterion was met. Not a rating. A yes or a no.

At six weeks, put the yes-no column beside the equity curve. If the no marks cluster in the fortnight after your best runs, you have found overconfidence after a winning streak in your own data, and you now know which week to be careful in — which is not the week you expected.

For the opposite side of the same mechanism, read fear and greed in trading; for what happens when a position is running and the plan starts moving, why traders move their stop loss; for the version that fires after a loss rather than a win, how to stop revenge trading.

The pattern quiz identifies which behaviour is running your trading under pressure, and the free sample it sends maps that result onto the four surface patterns — reactive, controlled, avoidant and impulsive.

---

About the Author

Sofia Harchich is a Trading Psychologist and Behavioural Finance Writer with a Master’s in Psychology. She works at the intersection of Jungian shadow work, neuroscience, and market behaviour — helping traders understand the psychology driving their decisions, not just the strategy.

Read more at thewealthmirror.com/about