Most traders blow up at the peak of their equity curve, not the trough. The winning streak that preceded the blow-up was the actual cause. The losing trade that emptied the account was the visible symptom.
This is a counterintuitive claim, and most retail traders resist it. Wins are supposed to be the good part. Losing streaks are the dangerous part. Surely the time to worry is not when the system is working but when it isn’t.
The data says otherwise. Most account-destroying losses come from positions taken at the high-water mark, sized larger than usual, on setups slightly worse than the trader’s normal criteria, after a stretch of wins had quietly eroded their discipline. The winning streak was the conditions under which the blow-up trade became takeable. Without the streak, the trader would not have taken it.
This article is about how winning streaks corrupt discipline, the four manifestations to watch for, the math of post-streak losses, and the reset protocol that prevents the streak from becoming the cause of the next disaster.
Four mechanisms, all operating below conscious awareness.
Dopamine creates the illusion of skill. A string of wins produces a neurochemical reward signal. Your nervous system processes the wins as feedback that you are doing something right. This feels like confidence. It is partly real — you are executing your edge — and partly a chemistry-driven over-reading of recent variance as skill. The over-read is what then sponsors marginal decisions.
Recent success reinforces marginal habits. If you took a B-grade trade and it won, your brain stored the B-grade behaviour and the win together. The next time the urge to take a B-grade trade arises, the precedent supports it. Across a streak, this accumulates — each marginal trade that wins makes the next marginal trade more likely to be taken. The criteria do not officially change; the felt threshold for taking trades quietly drops.
Position size creep. Confidence translates into size. Each subsequent win makes the next trade feel takeable at slightly larger size. The increase is gradual and often invisible to the trader. By the time they catch themselves trading at 1.5x or 2x their planned size, the streak has produced a position-size regime that is structurally riskier than their pre-streak baseline.
Lower scrutiny on entries. When the system is “hot,” entries get less analytical attention. The trader trusts their read more. They skip steps in the pre-trade checklist. The setups still look like valid setups to them — but the verification process that catches marginal ones has been bypassed by the feeling that they are reading the market well.
Each mechanism alone would be manageable. The four operating together, during a streak, produce a trader who is taking larger positions on slightly worse setups with less scrutiny — while feeling, internally, like they are at their best.
The most quantifiable and easiest to catch. Look at your last 10 trades. Is the position size on the most recent ones larger than the size on the first ones? If yes, and if your account has not grown proportionally, you have size creep. The size that the system was tested on is no longer the size you are trading. The math no longer applies the way you assume.
Subtler. Look at the trades in your last week and ask: would my best version of me have taken each of them? Would the trader I was a month ago, before the streak, have taken them? If the answer to either is “not all of them,” the criteria have slipped. The slippage is not a rule change — it is a felt threshold change. Both produce the same effect on expectancy.
Specific instance of criteria slippage. You see a setup that is close to your A-grade but not quite. Pre-streak, you would have skipped it. Mid-streak, you feel comfortable taking it because “you’re reading this market well.” The setup is the same. Your tolerance for taking it has changed. The market has not.
Mid-streak, the trader starts holding winners past the planned exit because “the move is strong.” This is the fear of leaving money on the table, supercharged by recent success. The exit signal fires. The trader overrides it. The trade gives back. Across multiple trades, the give-back accumulates and a streak of theoretical wins becomes a streak of mediocre realised wins — or worse.
Worth a concrete example.
You have a 55% edge with 1R losses and 1.5R wins. Expectancy per trade: +0.275R. Across 100 trades, +27.5R.
You hit a streak: 8 wins in a row, all at planned size. That is 12R captured — an excellent run.
The streak ends. Size creep has taken you to 1.7x your normal size on the next trades. Criteria have slipped — you are now taking some setups that pre-streak you would have skipped. Suppose the loosened criteria reduce your win rate to 48% temporarily, and the increased size means each loss is 1.7R instead of 1R.
The next 8 trades: 4 winners at 1.5R × 1.7 = 10.2R captured. 4 losers at 1R × 1.7 = 6.8R given back. Net 3.4R — positive, but a sharp reduction from the streak rate.
Now suppose one of the losses is a particularly bad one — a setup that did not have a defined exit because you ignored your time stop. That trade does not lose 1R; it loses 4R. Now the 8-trade post-streak run is 10.2R captured minus 6.8R from three normal losers minus 4R from the catastrophic one = -0.6R. The eight-trade streak that produced 12R has been almost entirely given back in the eight trades after.
Scale this up: a year of streaks and post-streak give-backs produces an equity curve that looks busy but goes nowhere. The trader works hard, takes many trades, feels emotionally invested — and ends the year flat or down, having converted real edge into break-even outcome through the post-streak corruption pattern.
Pre-committed steps after any defined run. The point is to interrupt the discipline-erosion sequence before it produces the post-streak give-back.
Step 1: Define the trigger. A run of +5R, +10R, or +15R captured — pick the number that fits your typical R distribution. The number is committed to in advance, not adjusted during the run.
Step 2: Drop size back to baseline. Whatever your normal 1% (or 0.5%, or whatever your defined size is) was at the start of the run, you return to it. Not at the end of the trading day — at the moment the run trigger fires.
Step 3: Re-read your rules. Yes, actually. Open the document where your rules are written. Read them. The point is to reset the felt threshold for taking trades against the written threshold. After a streak, the two have drifted.
Step 4: Review the streak for whether process grades were maintained. Were the trades that produced the run A and B grades? Or did C grades sneak in and happen to work? If the latter, the streak was not the validation it felt like.
Step 5: Treat the next trade as fresh. The streak is over. The next trade does not need to continue it. The next trade is a normal trade, executed by your normal rules, at your normal size, with no carry-forward of recent emotional charge.
The mechanical version of the protocol, suitable for traders who know their discretion will not catch the drift in real time.
Define a fixed number — 10R is a common choice; pick yours from your historical R distribution. Whenever a continuous run captures that much R, the reset protocol fires automatically. No discretion. No “just one more” trade at the elevated size. The reset is binary.
This sounds restrictive. It is. The restriction is the point. Discretion during a hot streak is the exact moment your discipline is most vulnerable; outsourcing the reset decision to a pre-committed rule means it does not have to survive your in-the-moment judgement.
Most traders resist this rule because it feels like artificially limiting their good periods. The math says good periods are exactly when you most need limits, because the post-streak give-back is what destroys the equity curve. The 10R rule does not limit upside; it preserves it from the predictable corruption that follows it.
The losing-streak protocol says: hold size, hold criteria, do not quit, do not try to make it back. The winning-streak protocol says the same things in different costume: hold size (do not let it creep up), hold criteria (do not let them slip down), do not skip the reset, do not try to extend the streak.
Both fears — the fear during the losing streak that the edge is broken, the overconfidence during the winning streak that you have figured the market out — are noise. The signal is your edge running its variance. The discipline is to operate the same way through both, with mechanical reset rules to catch the drift your discretion will miss.
The trader who treats winning streaks with the same suspicion they treat losing ones produces a much cleaner equity curve than the trader who treats wins as validation. The market does not care which streak you are in. Your discipline should not either.
Most blow-ups happen at the high-water mark. The winning streak that preceded the blow-up is the cause; the visible loss is the symptom. Size creep, criteria slippage, overconfidence in marginal setups, and ignored exit signals are the four manifestations. The reset protocol — ideally mechanical, triggered at a defined R level — interrupts the corruption sequence before it produces the give-back.
Treat your wins the way you treat your losses. Both are variance inside the edge. The discipline through both is what produces the year-on-year equity curve. The asymmetric reaction to them is what destroys it.
Next planned: Revenge Trading — the single quickest way to turn a manageable losing day into an account-destroying one, and the protocol that prevents it.