The overconfidence effect or bias is what makes a person believe their ability is greater than the evidence supports. It is exaggerated in confident people, resulting in worse decisions.
Overconfidence makes us think our decision-making is more accurate and our judgements more reliable than measured, empirical evidence would prove.
It is an individual’s subjective opinion, not tallying with the objective measurement of their ability or knowledge.
Personal and professional
Take any of the many and often repeated surveys of perceived driving ability and you will find that the majority of car drivers report that they are better than average behind the wheel (70% to 98% depending on the survey). Many admit that people who have driven with them might not agree.
That is the overconfidence bias. It is not possible for the majority to be above average.
Commercially the overconfidence bias is blamed for the Planning Fallacy (Buehler, Griffin, & Ross, 1994). This leads to the construction industry persistently having building projects that overrun in terms of cost and time.
Overconfidence in their own ability leads to underestimating both the time a project will take and the cost involved, despite past evidence of performance. The same happens with major IT projects.
Percentage success rates
The overconfidence effect was identified by psychologists Marc Alpert and Howard Raiffa in 1982.
They asked students at Harvard to guess at facts, such as the number of eggs produced daily in the US. Crucially, they said the students could provide an answer range as wide as they liked to make sure they were 98% certain they had the correct answer covered.
Despite being allowed to say as big a range as they liked, giving them the confidence they were wrong just 2% of the time, the students were actually wrong in 40% of their answers. They could have given larger spreads but were overconfident.
Forecasts and predictions
The sad fact is that the overconfidence effect means many forecasts – and that’s economic predictions, stock market expectations, share tips, company annual reports, chief executives’ boasts and analysts’ statements – are wrong.
The overconfidence effect means there is a difference – a gap – between what people think they know and what they really know. Their prediction might be right, but not because of what they know. Predictions are not reliable.
And because its impact is exaggerated by confident people, the louder someone shouts their prediction, the more publicity they receive, the greater following they attract, the more likely it is that the prediction is further away from fact.