The Difference Between a Losing Streak and a Broken Strategy

Seven losing trades in a row feels like proof that something is wrong. Sometimes it is — and sometimes it is exactly what a profitable strategy looks like on a bad week. Both produce the same red stretch on the equity curve, and reacting to the wrong one is how a working edge gets abandoned right before it was due to pay off, or a broken one gets traded for another three months on the strength of “it'll come back.”

Why a losing streak alone tells you nothing

A strategy with a 40% win rate will produce a run of seven consecutive losses roughly once every 45 trades, just from variance — no change in edge required. A strategy with a 55% win rate still produces one roughly once every 200 trades. Neither number is rare enough to rule out on the basis of the streak's length by itself. If you only look at “how many losses in a row,” a normal cold stretch and an actual breakdown are indistinguishable, because both start the exact same way.

This is the reason streak length is a bad trigger for a strategy review on its own. It tells you something happened. It does not tell you what.

What actually changed vs. what is just variance

The question worth asking isn't “how many losses,” it's “did the inputs stay the same while the outcome got worse.” A few things worth checking against the trade record before and during the streak:

• Setup criteria — were the losing trades taken on the same entry conditions as the winning ones, or did the criteria quietly loosen once losses started piling up? A streak that starts after the rules already drifted is not evidence against the original strategy.
• Market regime — did volatility, trend strength, or session behavior shift under the strategy? A mean-reversion setup built for range-bound price action will lose consistently the moment the market starts trending, and that is a regime change, not a broken edge.
• Average loss size — did losses during the streak match the planned stop distance, or did they run larger than usual? Bigger-than-planned losses point at execution breaking down, which is a different problem from the strategy itself losing its edge.
• Position size— did risk per trade stay flat through the streak, or did it creep up mid-run trying to recover faster? That pattern turns an ordinary variance stretch into a much deeper drawdown, and it's a sizing failure, not a strategy failure.

If the setup criteria, regime, and per-trade risk all stayed constant and the losses were the size they were supposed to be, the streak is more likely variance. If any of those shifted, the streak is a symptom of something else that changed — and fixing that thing is a different task than abandoning the strategy.

The sample-size problem this runs into

None of the checks above resolve instantly, because distinguishing variance from a broken edge is fundamentally a sample-size question — a strategy's expectancy only becomes visible over enough trades to average out. Judging it from the seven trades inside the streak alone is the same mistake as judging it from seven winners: neither stretch is long enough to mean much by itself.

What does move the needle is watching whether the same warning signs — loosened setup criteria, larger-than-planned losses, creeping size — show up again the next time drawdown hits. A single instance of any of them could be noise. A pattern across multiple streaks is no longer a coincidence.

What to do while you wait for the answer

The honest position, most of the time, is that you cannot know in the moment which one you're in. What you can control is not making it worse while you find out: keep risk per trade at the size it was before the streak started, keep the setup criteria exactly as written, and let the sample grow rather than forcing a verdict out of seven data points. Cutting size during an uncertain stretch is reasonable. Changing the rules mid-streak removes the one piece of information — did the original rules keep working — that would have actually answered the question.

getALPHA logs setup, stop distance, and position size on every synced trade, so a drawdown stretch can be checked against what actually changed instead of against memory, and the AI coach flags when sizing or entry criteria drift during a losing run — the two failure modes most likely to turn ordinary variance into something worse.