Memory is not an average. A session with one terrifying drawdown and a flat finish is remembered as a disaster even if it ended green. A month that closed with three wins is remembered fondly even if the middle was a mess.
This matters because you set next week's behaviour from a remembered summary, not from the data. Traders often reduce size after a month that was statistically fine but ended badly, or raise size after one that ended well and was otherwise poor.
Let the records, not the feeling, set the parameters. A simple end-of-month sheet with expectancy, max-drawdown, and rule adherence overrides a distorted memory.
Related: availability-heuristic, outcome-bias