The Outside View and Reference Class Forecasting

Key Insight

The 'outside view' is a powerful method to mitigate the planning fallacy, contrasting with the 'inside view' which focuses solely on the specific details of a unique project. Instead of relying on individual circumstances and personal experiences, the outside view involves identifying a reference class of similar cases and consulting their statistical outcomes to generate a baseline prediction. This approach often reveals that inside-view forecasts are unrealistically optimistic, as it accounts for 'unknown unknowns' that are difficult to foresee in a specific project.

Implementing reference class forecasting involves three steps: first, identifying an appropriate reference class (e.g., kitchen renovations, large railway projects); second, obtaining statistics from this reference class (e.g., average cost overruns, success rates) to generate a baseline prediction; and third, adjusting this baseline with specific information about the current case only if there are particular reasons to expect a deviation from the average. This structured methodology significantly improves forecasting accuracy for complex endeavors.

Despite its demonstrable benefits, the outside view is frequently ignored. 'Pallid' statistical information is often discarded when it conflicts with personal impressions or the compelling coherence of an inside-view narrative. Professionals, such as lawyers or doctors, often emphasize the uniqueness of individual cases, resisting general statistics. This preference for the inside view leads to irrational perseverance in projects, even when the outside view indicates a high probability of failure, as demonstrated by the curriculum project that took eight years to complete despite an initial two-year estimate and a 40% chance of failure from the outside view.