Quantum BiasHindsight Bias in Trading: Why Replays Make You Overconfident
The most comfortable bias in trading, and the one that quietly does the most damage.
In this article
Hindsight bias is the tendency, once you know how something turned out, to believe you would have predicted it. In trading, it turns every finished chart into an obvious setup. Research shows the bias is strong, automatic and costly: in a study of 85 investment bankers, the ones with the strongest hindsight bias performed the worst.
Every trader reviews charts after the session. Few realize that the review itself changes what they remember believing beforehand. This article goes through the research, shows how the bias shows up in trading, and gives you four habits to keep it in check.
Where the idea comes from
In 1975, the psychologist Baruch Fischhoff published two studies that defined the field.
In the first, Hindsight ≠ foresight: The effect of outcome knowledge on judgment under uncertainty, published in the Journal of Experimental Psychology: Human Perception and Performance, participants read a short historical account with four possible outcomes. Some were told which outcome actually happened. Those participants rated that outcome as more likely, and believed they would have seen it coming. Fischhoff called this "creeping determinism": knowing the result makes it feel inevitable.
In the second, written with Ruth Beyth, participants estimated the odds of different outcomes of President Nixon's 1972 trips to Beijing and Moscow before they happened. Afterward, they were asked to recall their original estimates. For events that had happened, they remembered giving higher odds than they really had. The paper's title says it all: I knew it would happen.
The finance study: bankers in London and Frankfurt
Bruno Biais and Martin Weber brought the idea into finance with Hindsight Bias, Risk Perception, and Investment Performance, published in Management Science in 2009 (vol. 55, no. 6).
Their argument is precise. To judge how risky a market is, you need to remember how uncertain you were before prices moved. Forget that uncertainty, and every move looks like it was predictable, so you underestimate volatility.
They tested it twice:
- With 66 students at the University of Mannheim, hindsight bias lowered their estimates of volatility.
- With 85 investment bankers in London and Frankfurt, the more biased bankers performed worse. The authors report that the result held after accounting for location, information, overconfidence and experience.
- 011
What did I write before the open? · Anchors the review to what you actually knew
- 022
What information arrived during the session that I could not have known? · Separates surprises from mistakes
- 033
Did each trade follow the written plan? · Grades the decision, not the outcome
- 044
Where would I have been wrong, using only the bars to the left? · Forces a forward-looking view
- 055
What, if anything, do I change in the plan? · Turns the review into a rule rather than a feeling
A visual summary of the article's table — not a performance chart or a new backtest.
How hindsight bias shows up in trading
1. The replay that looks easy. You scroll back through this morning's session. The reversal at 10:15 ET is obvious, and the entry would have been clean. But at 10:14, the chart showed the same price action in the middle of a move, not at the end of one.
2. The bias that was "always" bearish. At noon, after a drop, it's easy to remember leaning bearish at 9:00. Without something in writing, there's no way to check.
3. The backtest that feels too slow. Going through a strategy on old charts, you skip the trades that "obviously" wouldn't have triggered, because your eye already knows how it ends.
4. Underestimated risk. As Biais and Weber argue, if past moves look predictable, you size positions as if future moves will be too, and the next surprise is bigger than your plan allowed for.
Why it matters more if you trade intraday
As an intraday trader, you make a lot of decisions and review a lot of sessions, and each review is another chance for the bias to build up a false sense of skill. Over months, that can give you confidence no record supports.
It also messes with how you learn. If every loss looks, in hindsight, like an obvious mistake, the lesson becomes "be more careful". The real lesson might be that the trade was reasonable and the outcome was just noise.
Four habits that limit hindsight bias
1. Write before you look. Write down your read before the open, with the time: direction, levels, and what would prove you wrong. That's why every daily bias worth following has a time stamp.
2. Replay bar by bar. When you review a session, hide the right side of the chart and move forward one bar at a time, deciding before each new bar shows up.
3. Grade decisions, not outcomes. For each trade, ask whether the decision followed your plan based on what you knew at the time. A good decision can lose, and a bad one can win.
4. Keep the losers in the log. A journal that only remembers the trades you "should have taken" is a hindsight machine. Count every trade.
A post-session review that resists hindsight
Most reviews start with the chart. A review built to resist hindsight starts with what you wrote before the session, and only then opens the chart. Here's a five-minute template:
| Step | Question | Why it helps |
|---|---|---|
| 1 | What did I write before the open? | Anchors the review to what you actually knew |
| 2 | What information arrived during the session that I could not have known? | Separates surprises from mistakes |
| 3 | Did each trade follow the written plan? | Grades the decision, not the outcome |
| 4 | Where would I have been wrong, using only the bars to the left? | Forces a forward-looking view |
| 5 | What, if anything, do I change in the plan? | Turns the review into a rule rather than a feeling |
Step 2 is the one most traders skip. A CPI surprise at 8:30 ET, a Fed speaker at 10:00 or news on one stock can explain a move no pre-market read could have seen coming. Calling those losses "mistakes" teaches you the wrong lesson.
Hindsight and recaps
Recaps, ours included, show which reference levels were reached and when. That's useful, and it's also exactly the format hindsight bias loves: a finished chart with the answers marked on it.
The fix is the order you read things in. Read the pre-session note first, then the recap. A level reached at, say, 10:43 ET only means something if you know it was written down before the open, before anyone knew it would get hit. And a level reached isn't a trade taken: it says nothing about the price you would have paid or the stop you would have used.
What this means for reading any research
Hindsight bias also shapes how traders read other people's work. A chart posted after the session, with arrows on the turning points, is a replay, not a forecast. It shows what happened, with the benefit of knowing it happened.
That's one reason we separate five kinds of evidence: a report published before the session, a replay, an illustration, a backtest and real performance. Only the first shows what was known ahead of time.
Limits
- The bankers in Biais and Weber's study weren't futures day traders. The mechanism is general, but nobody measured how big the effect is for intraday traders.
- Lab studies measure the bias on specific tasks, and how strong it is in real trading varies from person to person.
- Knowing about it doesn't make it go away. As early as 1977, Fischhoff found that just warning people about the bias barely reduced it, which is why the habits above rely on records, not willpower.
Sources
- Fischhoff, B. (1975). Hindsight ≠ foresight: The effect of outcome knowledge on judgment under uncertainty. Journal of Experimental Psychology: Human Perception and Performance, 1(3), 288–299. https://doi.org/10.1037/0096-1523.1.3.288
- Fischhoff, B., & Beyth, R. (1975). "I knew it would happen": Remembered probabilities of once-future things. Organizational Behavior and Human Performance, 13(1), 1–16. https://doi.org/10.1016/0030-5073(75)90002-1
- Fischhoff, B. (1977). Perceived informativeness of facts. Journal of Experimental Psychology: Human Perception and Performance, 3(2), 349–358.
- Biais, B., & Weber, M. (2009). Hindsight bias, risk perception, and investment performance. Management Science, 55(6), 1018–1029. IDEAS/RePEc
FAQ
What is hindsight bias? The tendency to believe, after learning how something turned out, that you would have predicted it.
How does hindsight bias affect traders? It makes past moves look predictable, which can pump up confidence and lead traders to underestimate risk, as Biais and Weber (2009) argued.
Does hindsight bias hurt trading performance? In a study of 85 investment bankers, the ones with stronger hindsight bias performed worse.
How do I avoid hindsight bias when backtesting? Lock in your rules before looking at results, replay bar by bar with the future hidden, and record every trade, not just the ones you would have taken.
Is chart replay useful? Yes, if you decide before each bar shows up. Scrolling through a finished chart mostly trains hindsight.
Is hindsight bias the same as overconfidence? They're related but different. Hindsight bias distorts your memory of what you knew. Overconfidence inflates your sense of what you know now. The first can feed the second, because past "predictions" look better than they were.
Do professional traders suffer from hindsight bias? Yes. The Biais and Weber study used investment bankers, not students, and still found both the bias and its link to worse performance.
Research and education only. Not financial advice.
How this article was made: drafted with AI assistance from the research cited above; facts and figures checked by Gabriel Décarie and Karel Panneton.
The week just closed, summarized, and the economic calendar for the week ahead.