Confirmation bias

A documented feature of ordinary reasoning rather than a failure of discipline: information consistent with an existing belief is noticed faster, weighted more heavily and remembered longer than information that contradicts it. It operates below awareness, which is what separates it from stubbornness, and it is strongest where the belief is costly to abandon.

In market work it takes recognisable forms. A level is redrawn slightly so that it holds. A time frame is changed until the picture agrees with the position. Commentary agreeing with the position is read to the end, and commentary against it is dismissed at the headline. The tell is asymmetry of effort: the supporting case is accepted as read and the opposing case is put on trial.

The conventional counterweight is a record written before the outcome is known, which is the whole reason a trading journal is kept at entry rather than at exit: a reason recorded in advance, together with the condition that would show it to be wrong, cannot be quietly revised afterwards. Whether that works is genuinely disputed. The psychological literature finds bias resistant to being told about it, and the strongest evidence is for structural fixes such as pre-committed exit levels and written falsification conditions, rather than for awareness on its own.