How does people’s confidence in a decision align with reality?
From an article by the Behavioural Insights Team
What if you are relying on instinct over analysis? Your gut overriding your brain. But what if your gut is systematically lying to you?
Dr Mark Egan, PhD in Behavioural Science has spent years testing exactly that question – measuring how well people’s confidence aligns with reality. Last year, he found 8 in 10 UK adults were overconfident – more sure their judgements would be correct than they actually were.
Expanding that research to 4,000 people across USA, Singapore, UAE, Nigeria & Saudi Arabia, he found that, on average, respondents answered a series of factual questions with 87% confidence but were correct only 68% of the time – a 19 percentage point difference representing major overconfidence.
Even his colleagues at BIT, a place filled with PhDs and experts in decision-making, still fall prey to overconfidence. According to almost a decade of internal data, when they say they’re 100% sure about something, they still end up being wrong 7% of the time.
Perhaps it shouldn’t be surprising that overconfidence is so pervasive, given there is now 50 years of research showing how cognitive biases can lead to distorted judgement – like how confirmation bias makes us seek information that supports existing beliefs, or how availability bias makes dramatic events like plane crashes feel more probable than they are. Bias-distorted beliefs leading to misplaced certainty.
So if overconfidence is biased judgement, what does unbiased judgement look like? It's a sweet spot where being well-calibrated means alignment between confidence and accuracy. When a well-calibrated person is sure about something, they are rarely wrong – they can actually trust their gut.
If that sounds like a low bar for good decision-making, it’s one many groups fail to clear. One study found that when doctors were 80% sure that a patient had pneumonia, this turned out to be true only 20% of the time. But when weather-forecasters predicted an 80% chance of rain, it really did rain 80% of the time.
Being well-calibrated doesn't mean you getting decisions right 100% of the time. Given your track record, the information and experience you have, it's knowing what percentage of confidence you have in your decision.
Therefore you can be well-calibrated whether you’re getting decisions right 60% or 90% of the time – what matters is that your confidence matches your track record. Some domains naturally build this skill – like weather forecasting, where thousands of predictions get rapid feedback. Others, like medical diagnosis or political strategy, offer fewer learning opportunities. Not everyone can become a forecasting superstar but everyone can master the basics of well-calibrated judgement.
Most business and political leaders operate in feedback-poor environments, even if they’d prefer not to admit it. Fortunately, smart organisations can redesign their decision-making to encourage better judgment.
This starts with leaders modelling intellectual humility. Instead of asking “Are you sure?” – which when combined with social desirability bias will nudge people towards projecting false confidence – ask “How sure are you?” This opens space for graded responses and honest conversations about uncertainty.
Second, separate judgement from hierarchy. Run anonymous idea sessions where junior analysts’ insights compete on merit with senior executives’ hunches. Pre-register forecasts before outcomes are known to prevent hindsight bias. This will be both surprising and a great leveller.
Third, institutionalise pre-mortems. Before major decisions, assume the project has failed and work backwards to identify likely causes. This surfaces risks that overconfident teams would otherwise ignore.
Finally, track calibration systematically. Just as companies monitor financial metrics, they should monitor the accuracy of their predictions. Teams that consistently overestimate project success rates need recalibration, not just new strategies.
The goal isn’t perfect prediction – that’s impossible. It’s a mature acceptance of uncertainty. Well-calibrated leaders still make mistakes, but they’re rarely caught completely off guard. They plan for contingencies, build in buffers, and update their views when evidence shifts.
The next time you’re making a big decision, pause and ask yourself: ‘How sure am I about this?’ In a world that incentivises confident-sounding answers, that moment of intellectual honesty might be your competitive advantage.
Read the full article here.
Retweet about this article:
From an article by the Behavioural Insights Team, 04/02/2026