Forecast Illusion

Dobelli's term for the persistent belief that experts can predict the future—despite overwhelming evidence that expert predictions are barely better than chance for most complex domains. Philip Tetlock's research showed that the average expert's predictions were roughly as accurate as 'a dart-throwing chimpanzee.' Stock market forecasters, political analysts, and economic prognosticators consistently fail to outperform simple baseline models. Yet people continue to consume and act on forecasts because: the forecasters are confident, occasional correct predictions get remembered while failures are forgotten, and the alternative (accepting fundamental uncertainty) is psychologically intolerable.

When to use it

When strategic decisions are based on expert forecasts without track record verification; when confidence in predictions is high despite prediction domains being fundamentally uncertain; when planning needs to shift from prediction to robustness; when evaluating whether forecasting services provide value above simple baselines.

How it can help

Evaluate forecasts by track record, not by confidence. For any forecast: (1) Does this forecaster have a documented track record of accuracy? (2) Is this a domain where prediction is possible (weather short-term, physical systems) or fundamentally uncertain (markets, geopolitics, technology adoption)? (3) How would a simple baseline model (last year's numbers, trend extrapolation) perform? If the expert can't beat the baseline, the forecast has no value above free information. For your own planning: build strategies that are robust across multiple scenarios rather than optimized for one predicted future.