Joachim Andersen

Accelerated Learning

Hindsight Bias: Why the Past Looks Obvious

A small antique brass telescope on a tripod and a closed leather notebook on a dark walnut desk at night, beside a brass standing mirror that reflects a glowing golden winding path
Looking back, the winding road always seems to lead straight to where we are.


Hindsight Bias is the tendency to see past events as more predictable than they really were. Once I know how something turned out, the outcome starts to feel obvious, and I find it hard to remember how uncertain everything looked before.

I notice it most after a decision has gone badly. Looking back, the warning signs seem clear and I wonder how I missed them. But at the time, those signs were mixed in with many others that pointed in different directions, and I only know which ones mattered because I now know the ending.

The essence of Hindsight Bias is to ask: “What did I actually know before the outcome, and how many other outcomes seemed possible at the time?”


Origin: Creeping Determinism

The bias was first studied systematically by the psychologist Baruch Fischhoff in the 1970s. As a psychology graduate student in 1973, he heard the psychologist Paul Meehl observe that clinicians often overestimate how well they could have foreseen the outcome of a case, and set out to test it. He called the effect creeping determinism: as soon as we learn an outcome, our minds quietly absorb it into what we already know, so the outcome begins to look as if it had been inevitable all along.

In one of his first experiments, Fischhoff and Ruth Beyth asked students to estimate the likelihood of various possible outcomes of the US president’s visits to Beijing and Moscow in 1972. After the trips, the students were asked to recall the probabilities they had given. They remembered assigning higher probabilities to the events that had actually happened, than they really had. The title of their paper, “I Knew It Would Happen”, gave the bias its popular name: the knew-it-all-along effect.


Key Concepts

  • Memory Distortion
    We mis-remember our earlier predictions, so they seem closer to what actually happened.

  • Inevitability
    The outcome seems as if it was bound to happen, because we can now see a clear path leading to it.

  • Foreseeability
    We believe that we, or someone else, could and should have seen it coming.

  • Unaware of the Effect
    Knowing the outcome changes our judgment without us noticing, so we cannot simply decide to ignore what we know.

  • Unfair Judgments
    Because the outcome seems obvious, we may blame people for decisions that were reasonable given what they knew at the time.


Some Examples

  • A War Nobody Could Call
    In a 1975 study, Fischhoff gave students a short historical passage about the 1814 war between the British and the Gurkhas of Nepal, and asked them to rate the probability of four possible outcomes. Some students were told nothing about how it ended, while others were told that one outcome had happened. Being told that the British won raised the average probability given to a British victory from 33.8 to 57.2%, and being told that the Gurkhas won raised a Gurkha victory from 21.3 to 38.4%. The students were largely unaware that the outcome had influenced them, and overestimated what they would have known without it.

  • The Diagnosis That Seemed Obvious
    In a 1981 study by Hal Arkes and colleagues, 75 physicians read the same case history of a young man with a painful, swollen knee and estimated the probability of four possible diagnoses. One group was given no diagnosis, while the others were told which diagnosis was correct and asked to judge the case independently. For the two diagnoses the uninformed group thought least likely, physicians who were told that diagnosis was correct rated it 2-3 times more likely.

  • The Flood That Should Have Been Foreseen
    In a 1995 study, Kim Kamin and Jeffrey Rachlinski asked participants whether a city should pay for a bridge operator during the winter to prevent flooding. When asked in advance, as a planning committee would, 24% said the city should hire one. When participants were told that a flood had occurred and asked to judge the city as jurors, about 57% found it negligent for not hiring one. Even a judge’s explicit warning about hindsight bias made almost no difference, with 56% still finding the city liable.

  • Reading the Past Through the Present
    When I look back at a career choice, an investment or a project that went wrong, I often find myself thinking that I should have known better. When I write down what I actually knew at the time, I usually find that the outcome was one of several reasonable possibilities, not the obvious one it seems today.


Applying Hindsight Bias

Because the bias works without my awareness, simply telling myself to ignore the outcome does not help, as the flood study shows. What does help is to bring back the uncertainty I faced at the time. In a 1988 study, Arkes and colleagues found that asking experts to give a reason why each alternative diagnosis might have been correct reduced hindsight bias. This is the simple process I follow:

  1. Write Down Predictions Before the Outcome
    I record what I expect to happen, how confident I am and why, so that I can later compare it with what really happened.

  2. Reconstruct What Was Known
    When I review a decision, I list only the information that was available at the time, not what I learned afterward.

  3. Explain the Other Outcomes
    I give at least one reason why each outcome that did not happen could have happened, which reminds me how open the future really was.

  4. Separate Decisions From Results
    I judge a decision by the information and reasoning behind it, not only by how it turned out, as I describe in Probabilistic Thinking.

  5. Be Careful When Judging Others
    Before I blame someone for not seeing something coming, I ask whether I would have seen it without knowing the ending.


Pitfalls and Limitations

  • It makes learning harder. If every outcome seems obvious in retrospect, I conclude that I understood more than I did, and I stop looking for what I really missed.

  • It can make me too harsh, or too forgiving. Hindsight can make a reasonable decision look careless, but it can also make a lucky decision look wise.

  • Not every warning sign was invisible. Sometimes a risk really was foreseeable and ignored, so the goal is not to excuse every mistake, but to judge it by what could have been known.

  • Warnings alone do not work. Knowing about the bias does not protect me from it, which is why I must rely on written records and structured reviews rather than memory.


Related Mental Models

Hindsight Bias reminds me that the past only looks obvious because I know how it ended. By writing down my predictions, reconstructing what was known at the time and giving the outcomes that did not happen a fair hearing, I can learn from experience without fooling myself that I knew it all along.



To learn more about mental models and philosophy,
check out: Wise: How to Attain a Deep Understanding of the World, History, Our Place in It, and How to Live Well



📚 Sources and References