Evaluations are rarely lost by people who cannot read a chart. They are lost by people who can, in five patterns that repeat with unusual consistency. Every one of them is recognisable from the inside if you know what to look for.
One: the recovery trade
An early loss, then a position sized to undo it rather than sized by the rules. This is the single most common cause and it does not feel like recklessness at the time — it feels like conviction, and like correcting an obvious mistake.
The tell is arithmetic: you are working out what size would get the account back to flat. That calculation has no place in position sizing, and noticing yourself doing it is the intervention.
Two: the rollover
A trade held across the daily reset, consuming part of two allowances, or opened just after a session that had already spent most of the current one. The rule was understood; the timezone arithmetic was done in a hurry.
Three: the trailing floor nobody was watching
The account is up, comfortably, and the trader is tracking profit rather than distance to the floor. Under a trailing drawdown those two numbers move together and the second one is the one that ends the account. A perfectly ordinary give-back becomes a breach because the floor had climbed and nobody had looked at it since.
Four: the size that grew
Nobody decides to double their position size. It happens by increments, over a good fortnight, each step justified by the last one working. Then the size that quietly tripled meets the losing streak it was never calculated to survive.
The countermeasure is to write down the position size at the start and audit it weekly against what you actually traded. Drift is invisible in real time and obvious in a list.
Nobody decides to double their size. It happens by increments, each one justified by the last one working.
Five: the news release
A position open through a scheduled release, with a stop that the gap jumps straight past. The loss is not the one that was risked, because in a fast market a stop is an instruction rather than a guarantee.
Some plans permit news trading and some do not, and the rule frequently differs between the evaluation and the funded account on the same plan. Read both rows before you assume the permission carries over.
What they have in common
Every one of the five is a risk failure rather than an analytical one, and four of the five happen after something has already gone wrong. The trades that end evaluations are overwhelmingly the ones taken in response to other trades.
Which suggests the highest-value rule you can write for yourself is not about entries at all. It is about what you are permitted to do in the hour after a loss.