In short
Optimism bias is the tendency to rate your own chances of good outcomes as better than other people's and your chances of bad outcomes as worse than theirs, by more than the facts about you can justify. It shows up in how long people expect projects to take, how much they save, whether they buy insurance or wear a helmet, how they read their own health risks, and in the way most people in most surveys place themselves on the right side of average.
How optimism bias works
Nobody has the figures for their own future, so a question about your chances of a heart attack or a happy marriage is answered from what is at hand: your plans, your habits, the people you know, and a sense of yourself as careful or lucky. Several of those sources tilt the answer. People compare themselves with a vague average person rather than a real one, and the average person has no plans and takes no precautions, so almost anyone seems better placed. Events that feel controllable invite the thought that you will control them, and Weinstein found that the more controllable an event seemed, the larger the optimism. And a good outcome is easier to picture than a bad one, because you have rehearsed it.
Tali Sharot's 2011 review in Current Biology summarized the findings across methods and domains: people underestimate their chances of divorce, illness and job loss, overestimate their chances of success and longevity, and expect more pleasant experiences in the coming month than they go on to have. By most estimates, she wrote, about 80 percent of the population displays the bias. Her own experiments found an asymmetry in how people update: told that a risk was lower than they had thought, they moved their estimate down; told it was higher, they moved it less.
The bias is not the same as a bright outlook, and not always a fault. A person who expects to succeed may work harder and persist longer; Sharot argued that the bias is part of healthy functioning and is reduced or absent in depression. The costs come when the inflated estimate feeds a decision: the retirement fund not started, the helmet left at home, the project schedule no project has ever met.
There is also a serious argument that comparative measurements overstate the bias, and the demo on this page uses exactly that kind of measurement. Harris and Hahn showed in 2011 that when a bad event is rare, most people really are at below-average risk of it, because the average is pulled up by a minority at high risk, so a group whose members all report accurately will still look optimistic on a comparative scale. That does not show the bias is absent, but it makes a group placing itself below average on rare events weaker evidence than it looks, and an individual's answer proves nothing at all. The result screen above reports which side of average you landed on and nothing more.
The original study
Neil Weinstein published "Unrealistic optimism about future life events" in the Journal of Personality and Social Psychology in 1980. In the first study, 258 college students were given a list of 42 events, some desirable and some undesirable, and asked to estimate how much their own chances of each differed from the chances of their classmates of the same sex.
On average, the students rated their chances of the positive events as above their classmates' and their chances of the negative events as below. Weinstein's argument was that, since the students were being compared with each other, the average rating for any event should have been zero if the group judged accurately, so the consistent lean was the bias; Harris and Hahn's later objection to that inference is set out above. Weinstein also tested what moved it. Optimism was greater for events that people saw as controllable and for events associated with a stereotype of the kind of person it happens to; the students seemed to be comparing themselves with that stereotype rather than with an actual classmate. Personal experience cut the other way: the more experience students had with a bad event, the less optimistic they were about avoiding it; for positive events the association was not significant.
The second study, with 120 female undergraduates, tested the comparison explanation directly. One group listed the factors they thought affected their own chances of eight events. A second group read those lists before making their own comparative judgments, so that they had in front of them the circumstances of real classmates rather than an imagined average. Their optimism was significantly reduced, though not eliminated. Weinstein's conclusion was that people are unrealistically optimistic partly because they fail to consider that others have the same plans and advantages they do.
The paper set the pattern for a large literature using the same comparative method, the one Harris and Hahn's 2011 critique addresses. Later work, including Sharot's updating experiments, asked people for their own probability of an event and then gave them the base rate, which avoids that problem.
Everyday examples
The rows below are illustrations, not findings from the studies cited on this page.
| Situation | What the bias does | What helps |
|---|---|---|
| A couple decides whether to write a prenuptial agreement or a will | Divorce and early death are things that happen to other couples, so neither document gets written | Start from the base rate for people like you and ask what would make your case different |
| A smoker reads the warning on the pack | The risk is accepted as true of smokers in general and discounted for this smoker in particular | Ask what specific fact about you lowers the risk; if there is none, you are at the average |
| A project team estimates a delivery date | The estimate assumes nothing goes wrong, as every previous estimate did | Look up how long similar projects actually took and start from that |
| A thirty-year-old postpones retirement saving | Future income is expected to be higher and future health better than the figures suggest | Automate the contribution now, at the rate the base rates recommend, rather than the rate the forecast does |
| A cyclist skips the helmet for a short trip | Accidents are what happen to careless riders, and this ride is short | Decide the rule once, for all rides, rather than estimating each ride |
How to counter it
Optimism about your own chances cannot be argued away from the inside, so the counter-measures bring in information from outside.
- Start from the base rate. Find the actual frequency of the event among people like you and ask what specific fact would move you off it; Weinstein's second study showed that seeing real classmates' circumstances reduced the bias.
- Compare with a real person, not an average one. The average person has no plans; the colleague at the next desk has the same plans you do.
- Treat controllability with suspicion. The events people feel most optimistic about are the ones they believe they control.
- Make the decision rule in advance. Insurance, savings rates and safety equipment are better decided once by rule than estimated case by case.
- Keep the bias where it helps. Expecting to succeed at a task can improve effort; expecting to escape a risk that cannot be worked at only removes the precaution.
What it is not
Optimism bias sits in a cluster of self-favoring judgments, and the names are often used for one another.
- It is not illusory superiority, the tendency to rate your own abilities and traits above average. That bias is about what you are like now; optimism bias is about what will happen to you, and a person can rate their driving as average while still expecting to avoid the crash.
- It is not the planning fallacy, the underestimation of how long a task will take and what it will cost. The planning fallacy is optimism bias applied to one kind of forecast, with its own mechanism: imagining the plan going right rather than consulting how similar plans went.
- It is not the illusion of control, the belief that you can influence outcomes that are in fact chance. The illusion of control feeds optimism bias, since events that feel controllable attract the most optimism, but it is a belief about influence rather than about outcomes.
- It is not simply the mirror image of pessimism bias. A pessimistic lean appears in some conditions and domains, and the two are studied separately.
Tests and tools that use this
Related biases
Optimism Bias FAQ
What is optimism bias?
It is the tendency to rate your own chances of good events as better than other people's and your chances of bad events as lower, beyond what the facts support. Weinstein demonstrated it in 1980 by asking 258 students to compare their chances of 42 life events with their classmates': on average they placed themselves above average for the good events and below for the bad, which he argued could not hold for the group as a whole. The pattern has since been found across ages, nationalities and domains.
How does the demo on this page measure optimism bias?
You set your chance of eight events against the average for people of your age and sex, with every slider starting at the average. An answer on the lucky side, above average for the four good events or below it for the four bad ones, counts in the direction of the bias, and the result reports whether most of your eight answers fell that way. It cannot tell whether any answer was wrong, since the page does not know your real risks. Only the outcome label leaves your browser.
Is optimism bias real, or a statistical artifact?
Both answers have evidence. Harris and Hahn showed in 2011 that for rare bad events most people really are at below-average risk, because a small high-risk minority pulls the average up, so a group that answers accurately will still look optimistic on a comparative scale. Their argument raises doubts about how large the bias is and when it appears, rather than showing that it does not exist. Studies that ask for absolute probabilities and give feedback, such as Sharot's, approach it differently.
Is optimism bias a bad thing?
Not always. Sharot's 2011 review argued that moderate optimism about one's own future is part of ordinary mental health, is reduced in depression, and can improve effort and persistence at tasks where effort matters. The costs come when an inflated estimate feeds a decision that effort cannot rescue: skipping insurance, underfunding retirement, scheduling a project as if nothing will go wrong, or ignoring a health warning on the ground that it applies to other people.
Why do people think bad things will not happen to them?
Partly because they compare themselves with an imagined average person who has no plans and takes no precautions, so nearly anyone looks better placed. Weinstein found that optimism was largest for events people believed they could control, and that showing people the actual plans and circumstances of their classmates reduced it. Sharot's experiments add an updating asymmetry: people revise their estimates readily when told a risk is lower than they thought, and less when told it is higher.
How can I correct for optimism bias in my own decisions?
Begin from the base rate for people like you and require a specific reason before moving off it. Compare yourself with a particular person rather than with an average, since the particular person has the same advantages you were counting on. Be most careful about events that feel within your control, which attract the most optimism. And decide rules in advance for insurance, saving and safety, so that each case is not estimated fresh by a forecaster who expects to be lucky.
References
- Weinstein, N. D. (1980). Unrealistic optimism about future life events. Journal of Personality and Social Psychology, 39(5), 806-820. doi.org
- Sharot, T. (2011). The optimism bias. Current Biology, 21(23), R941-R945. doi.org
- Harris, A. J. L., & Hahn, U. (2011). Unrealistic optimism about future life events: A cautionary note. Psychological Review, 118(1), 135-154. doi.org
Cite this source
Psychology.com. (2026, October 21). Optimism Bias. Psychology.com. https://psychology.com/biases/optimism-bias