Advertisement IC Markets

Recency bias: judging a system by its last ten trades

Master exponential money management, position sizing calculators, strict daily stop-loss limits, and overcoming FOMO on micro-timeframes.
Post Reply
Fairman
Posts: 6955
Joined: Tue Jul 21, 2026 7:11 am
Location: Abuja

Recency bias: judging a system by its last ten trades

Post by Fairman »

Recency bias: judging a system by its last ten trades

Ten trades is nothing for a statistical judgment, yet it dominates emotional perception. A cluster of losses feels like the system has stopped working. A cluster of wins feels like mastery.

For scalpers who see many trades a day, the recent window shrinks even further. Two hours of poor results can trigger parameter changes, while two hours of good results can trigger overconfidence.

The remedy is a predefined evaluation window. Decide in advance how many trades or how many days are needed before any change is considered. For many systems, that means at least a few hundred trades.

Track confidence intervals. With a win rate near 55 percent, the standard error over 100 trades is about 5 percentage points, so results between 45 and 65 are statistically indistinguishable.

Discipline here protects against two errors: abandoning a good system too early and clinging to a bad one too long.

What sample size must a change in performance reach before any adjustments are made?
It’s Fairman :geek:
Recommended broker for automated trading & scalping IC Markets
Amanhasnoname
Posts: 2060
Joined: Fri Oct 30, 2026 9:45 am

Re: Recency bias: judging a system by its last ten trades

Post by Amanhasnoname »

A bad morning used to be enough to make me tinker, which is how I chopped up a system that was fine. Twelve or fifteen losing trades would arrive in a couple of hours and I would start moving a filter. I now refuse to change anything until I have two hundred and eighty live trades or twelve full sessions, whichever comes later. The last genuine change came after three hundred and twelve trades, when the win rate had slipped from fifty-seven percent to forty-nine percent and one filter's lookback turned out to be the cause.

Ten trades are not a sample. I shuffled my own history and watched ten-trade win rates swing between thirty-eight and seventy-two percent while the whole book sat at fifty-four. That range is wide enough to feel like a broken system or a new lease of life, and both feelings were noise.

The rule sounds rigid, but it cut the two mistakes the opening post named. I am slower to abandon something still inside its normal band, and slower to keep defending a change I made in a hurry after a bad morning.
Post Reply