Traders Take More Risk After a Loss. Here's What the Code Does Instead.
A 2005 study of Chicago pit traders found professionals took above-average risk after a losing morning. We ran the same test on 29 strategy versions: position size, win rate and time to the next entry, before and after a loss.
Transcript
In a study of proprietary traders at the Chicago Board of Trade, the ones who lost money in the morning took above-average risk in the afternoon far more often than the ones who won. Thirty-one percent versus twenty-seven. Journal of Finance, two thousand five. That's what a person does after a loss. We asked what the code does.
The test is simple. For every trade, we look at the trade before it — same strategy, already closed. Was it a loss or a win? Then we compare what came next: the size of the position, the win rate, and how long the system waited before entering again.
Position size: identical. Twenty-five of twenty-nine versions have no mechanism for adding to a position in the code at all. After a loss the system cannot size up, because there is nothing to size up with. The four that can add, add on price structure — the odds of adding after a loss and after a win are the same.
Win rate after a loss: sixty-three point six percent. After a win: sixty-four point five. Nine tenths of a point. Then we shuffled the order of every trade, two thousand times, and ran it again. The real sequence was indistinguishable from a random one. P of zero point three nine.
And timing. Across the catalog, ten strategies enter sooner after a loss, fifteen enter later. That is a coin flip, not a pattern. There is no rush after a loss, because there is nothing in the entry logic that knows a loss happened.
The code doesn't remember it lost. There is nothing in it that could. Rules written before the loss can't be rewritten by it. Every number is modeled, every layer labeled, on the site.
Read the full write-up
All videos → · Watch on YouTube →
All figures are hypothetical or from live-tracked accounts as stated in the video. Past and simulated performance does not guarantee future results. This is educational content, not financial advice.