Trading Concept
Recency Bias
The tendency to overweight recent events when evaluating probability — making the last few trades feel more representative than they are.
A strategy that lost money last week feels broken. A strategy that made money last week feels invincible. Neither assessment is rational if the sample size is too small to draw conclusions. Recency bias is the cognitive shortcut that treats the most recent observations as the most representative — even when they are noise.
Consider a strategy with a documented 58% win rate over 500 trades. It just lost 7 of its last 10. The probability of this happening by chance, given a 58% true win rate, is approximately 7%. Unlikely but not rare — it will happen roughly once every 14 stretches of 10 trades. Recency bias makes this 10-trade window feel like proof of failure. The 500-trade track record fades into the background.
The antidote to recency bias is sample size awareness. Ten trades is not a sample — it is an anecdote. Fifty trades is the minimum for a preliminary signal. Two hundred trades begins to be statistically meaningful. The ledger exists partly to provide this perspective. When the last week feels terrible, the ledger shows the full history — every trade, every gate decision, every outcome. The numbers do not care about last week. They report the entire record.