Optimism Bias

Optimism bias is the tendency to believe that you are more likely than average to experience positive events and less likely than average to experience negative ones. It’s the quiet voice that says the deadline is achievable (it usually isn’t), the project will come in on budget (it rarely does), and the bad thing will happen to someone else (it might not).

Neuroscientist Tali Sharot, whose research at University College London has been central to understanding this bias, estimates that around 80% of people exhibit optimism bias to some degree. It’s one of the most consistent and widespread cognitive biases ever documented - found across cultures, age groups, and levels of intelligence.

What optimism bias looks like

Optimism bias isn’t about being a cheerful person. It’s a systematic error in how we estimate probabilities - specifically, probabilities that involve us.

Ask people to estimate their risk of divorce, and they’ll consistently underestimate it, even when they know the base rate is roughly 40-50%. Ask them to estimate how long a home renovation will take, and they’ll underestimate by an average of 30-40%. Ask them whether they’re more likely than their peers to live past 80, and a statistically impossible number will say yes.

The bias is selective. People are reasonably accurate at estimating risks for others. They can look at someone else’s situation and make a fair assessment. But when it comes to themselves, the calculation shifts. The same brain that correctly identifies risk in someone else’s life systematically downplays it in their own.

This connects to naive realism - the broader tendency to believe that your perception of reality is accurate while others’ perceptions are distorted. Optimism bias is naive realism applied to the future.

How optimism bias works in the brain

Sharot’s neuroimaging research revealed something fascinating. When people receive information that is better than expected (say, learning that their risk of cancer is lower than they estimated), their brains update efficiently. The new information is incorporated, and future estimates adjust.

But when people receive information that is worse than expected, the brain does something different. It partially ignores it. The update is smaller, slower, and less complete. The machinery for processing good news works well. The machinery for processing bad news is selectively impaired.

This asymmetry means that over time, our internal model of the future drifts steadily toward the optimistic end. Each piece of good news is fully absorbed. Each piece of bad news is partially discounted. The result is a picture of the future that is systematically rosier than reality warrants.

Optimism bias in everyday life

Optimism bias and time

The planning fallacy - the consistent tendency to underestimate how long tasks will take - is optimism bias applied to time. Students estimating how long an essay will take, contractors quoting project timelines, software developers forecasting delivery dates - all systematically err on the optimistic side.

Daniel Kahneman and Amos Tversky documented this extensively, showing that even people who have repeatedly experienced delays on previous projects will predict that the current one will go smoothly. The past doesn’t update the forecast because each new project feels different - and optimism bias ensures that “different” almost always means “better.”

This connects to hindsight bias. After a project overruns, people rewrite the narrative: “We couldn’t have known about that problem.” But often they could have - the base rate of delays was always high. They just didn’t expect this time to be typical.

Optimism bias and money

Personal financial planning is riddled with optimism bias. People overestimate their future earnings, underestimate their future expenses, and assume their investments will perform above average. The result is under-saving, over-borrowing, and a persistent gap between financial plans and financial outcomes.

The sunk cost fallacy often operates alongside optimism bias. An investor holds onto a declining stock not just because they’ve already invested (sunk cost) but because they genuinely believe it will recover (optimism bias). The two biases reinforce each other - sunk costs keep you committed, and optimism bias keeps you believing commitment will pay off.

At the organisational level, optimism bias explains why large infrastructure projects almost always exceed their budgets. A study of major projects found that nine out of ten came in over budget, with cost overruns averaging 28% for rail projects and 45% for IT projects. The planners weren’t dishonest - they were optimistic.

Optimism bias and health

“It won’t happen to me” is the health version of optimism bias. Smokers underestimate their personal risk of lung cancer. People who don’t exercise overestimate their future health. Those with a family history of heart disease assume they’ll be the exception.

This isn’t ignorance. These same people can accurately describe the risks in general terms. They know the statistics. They just don’t believe the statistics apply to them. Confirmation bias supports the delusion - they notice the heavy smoker who lived to 95 and overlook the thousands who didn’t.

Optimism bias at the societal level

When optimism bias operates at scale, the consequences compound.

Climate change planning is consistently undermined by collective optimism bias. Governments set targets based on best-case scenarios, assume technological solutions will arrive on schedule, and underestimate the severity of projected impacts. The result is systematic under-preparation for outcomes that the science says are highly probable.

Normalcy bias plays a closely related role - the assumption that because things have been stable, they will continue to be stable. Together, optimism bias and normalcy bias create a powerful cognitive cocktail that makes societies slow to respond to gradual threats.

Financial markets exhibit optimism bias in cycles. During booms, investors collectively underestimate risk, overvalue assets, and assume that growth will continue indefinitely. The crash, when it comes, always feels like a surprise - even though the historical pattern of boom and bust is one of the most reliable in economics.

The evolutionary purpose of optimism bias

If optimism bias is so consistently wrong, why has evolution preserved it? Because in many contexts, being slightly wrong in the optimistic direction is more useful than being accurately pessimistic.

An optimistically biased ancestor who attempted a difficult hunt, tried to build a shelter in uncertain terrain, or pursued a mate despite the odds was more likely to succeed than a perfectly calibrated realist who calculated the risk and stayed home. Optimism bias isn’t a bug in human cognition - it’s a feature that trades accuracy for motivation.

Sharot’s research suggests that optimism bias is also linked to better mental health. Mildly depressed people tend to be more realistic in their risk assessments - a phenomenon called “depressive realism.” The psychologically healthy brain is the one that distorts reality slightly in the positive direction.

This doesn’t mean optimism bias is harmless. It means the solution isn’t to eliminate it but to know when to compensate for it - particularly when the stakes are high and accurate forecasting matters more than motivational momentum.

How to work with optimism bias

You probably can’t eliminate optimism bias, and you probably shouldn’t try. But you can build systems that correct for it.

Use probabilistic thinking when making important decisions. Instead of asking “what will happen?”, ask “what’s the range of things that could happen, and how likely is each?” This forces you out of a single optimistic scenario and into a more honest assessment of possibilities.

Apply a “reference class forecast.” Instead of estimating a project from the inside (how long will this specific task take?), look at how long similar projects have taken in the past. The base rate is almost always a better predictor than your internal estimate.

Build in buffers explicitly. If you think something will take three months, budget for four. If you think a project will cost ten thousand pounds, plan for thirteen. You’re not being pessimistic - you’re compensating for a known, documented, and predictable bias.

And when someone in your team raises concerns about a plan, resist the urge to dismiss them as “negative.” They might be the one person in the room whose optimism bias isn’t doing the talking.