Matcha Time

← All 12 tools

Survivorship bias

You only see the ones that made it; the rest are missing from the picture.

What it is

Survivorship bias is drawing conclusions from the cases that passed some filter while overlooking those that did not. The missing cases often carry the real lesson.

An example

In the Second World War, bombers came back with bullet holes in some areas and few in others. The tempting reading was to armour the damaged areas. The statistician Abraham Wald started from the planes that had not come back: the armour belonged where the survivors were clean.

Where to use it

When advice comes from winners, or data comes only from what is still around.

The question to ask

Which cases am I not seeing, and why are they missing?