The reason you underestimate how long everything takes

The task was supposed to take two weeks. It took five. The previous, similar task also overran, and so did the one before that. Each time the estimate was made with full knowledge of the previous overrun, and each time it was made the same way.

What everyone sees

The overrun is treated as an anomaly each time it occurs: unexpected complications, bad luck, unusual circumstances. The pattern across projects is not noticed because each project is planned from its own interior, where the complications have not yet appeared and therefore do not exist.

What is actually happening

Kahneman and Tversky described the planning fallacy: estimates are generated by constructing a scenario of how the task will unfold rather than by consulting the base rate of similar tasks. Buehler, Griffin and Ross confirmed that people consistently underestimate completion times even when reminded of past overruns. Flyvbjerg showed the same pattern at the scale of major infrastructure, where cost and time overruns are the statistical norm, not the exception.

Why it stays hidden

The fallacy persists because each new plan feels different from its predecessors. The planner does not experience themselves as repeating the same error; they experience themselves as addressing a specific problem whose specifics will go as imagined. Base-rate data is available in principle but is never consulted, because the inside view feels like knowledge and the outside view feels like statistics about other people’s projects.