Backtesting should include the drawdowns a trader can realistically follow, not only the returns a strategy produced on paper. If three consecutive losses make you abandon the rules, that breaking point belongs in the test before live capital is involved.
At 10:42 on a Wednesday morning, Marcus sat in a coffee shop near Chicago’s Loop with his phone face down beside a cooling espresso. His strategy had taken its third stopped-out trade in four sessions. The chart showed the same setup he had tested hundreds of times, but the last three red entries had made every signal look careless.
He had promised himself he would take the next qualified trade. Now he was about to skip it.
The cost was larger than one missed position. If Marcus changed the rules whenever a losing streak felt unbearable, his live results would no longer represent the strategy he tested. He would be trading a new, improvised system at the exact point when confidence was lowest.
That is the psychological gap many backtests leave open. A test can show a maximum drawdown, consecutive losses, and a return curve. It cannot assume that the person reading those numbers will keep following the plan once the losses arrive in real time.
A tested drawdown is only useful if you can remain inside it
A strategy’s historical drawdown is a record of what happened under its stated rules. It gives you a range to examine, not permission to expect the same path in the future.
Suppose a backtest shows a sequence of six losing trades during its deepest drawdown. That does not mean the next live drawdown will stop at six losses. Market conditions change, fills differ, and a historical result has limits. Still, the sequence gives the trader a serious question to answer before placing an order: could I take six planned losses at this size without changing my entry, exit, or position-sizing rules?
Marcus had looked at the drawdown figure. He had not translated it into his own tolerance. The number sat in a spreadsheet beside other metrics, detached from the moment when three losses made him want to intervene.
That translation matters. A 12% drawdown means something different to a trader risking $20 per position than to one risking $500. The percentage may be identical. The lived pressure is not.
Backtesting explained in practical terms means testing the strategy and testing the conditions required to follow it. If the position size creates a level of discomfort that causes rule changes, the position size is part of the strategy problem.
The third loss changes the decision environment
The first losing trade often feels normal. The second invites review. By the third, the trader may begin searching for evidence that the system has failed.
That search can become selective. A weak signal suddenly looks obvious. A skipped entry feels prudent. A valid stop-loss starts to look too tight. The written plan has not changed, but the trader’s relationship to it has.
This is why psychological tolerance belongs beside win rate, average loss, maximum drawdown, and trade frequency. A strategy with a tolerable historical drawdown but an intolerable live position size can produce the same result as a strategy with poor rules: the trader abandons it.
The answer is not to blindly accept every queued trade during a drawdown. Approval should remain a real decision. The useful question is narrower: am I rejecting this trade because it violates a written risk rule, or because the last three trades hurt?
An approval gate can make that distinction visible. Before approving or rejecting, compare the order with the planned entry, stop, size, and total risk limit. Write down the reason for a rejection. Over time, a trading journal can reveal whether “market conditions changed” means a defined condition or a reaction to recent losses.
For a related example of how consecutive losses and sizing interact, see [Max drawdown explained: how position size, consecutive losses, and stop-loss discipline interact]( /blog/max-drawdown-explained-how-position-size-consecutive-losses-and-stop-loss-discipline-interact-with-worked-examples-for-retail-traders-and-no-promised-outcomes-educational-content-n-315be2ad/ ).
Build a tolerance test into the backtest review
A tolerance test starts after the performance metrics. Review the worst historical sequences, then make them concrete.
Record the deepest drawdown, longest losing streak, and the time it took the strategy to recover in the tested period. Then reduce the position size until you could plausibly follow the rules through that sequence without needing the next trade to “make it back.”
There is no universal acceptable drawdown. A trader with a small account may need smaller risk per position. A trader managing a larger account may need a lower percentage risk because the cash swings still affect judgment. The correct level is the one that lets the written rules remain usable when the screen is red.
It also helps to define a review rule before trouble starts. For example, pause new entries only after a prewritten threshold, then review the strategy, market regime, execution quality, and position size. Do not invent the threshold after a bad morning. A rule created during distress usually serves relief before discipline.
The next signal should be evaluated against the plan
Marcus did not take the next trade automatically. He opened his journal and checked the signal against the four conditions he had written before the week began. The setup qualified. His planned risk remained within his limit. Nothing in his rules called for a pause.
So he approved the trade at the smaller size he had already decided he could tolerate.
The trade itself is not the lesson. It could have won or lost. The meaningful change was that Marcus had made room for the third loss before it happened. His next decision came from a documented process, not from the urge to stop feeling exposed.
Educational content, not financial advice.
Comments
No comments yet.