Texas sharpshooter

The Texas sharpshooter fallacy is when you pick a pattern or cluster in the data after you've already seen it (like drawing the target around where the bullet holes landed), then treat it as if it were a real target you aimed at. You're reasoning from noise or coincidence as if it were proof. The same idea appears as the cognitive bias known as clustering illusion: we see patterns in random data when we look after the fact. To avoid the fallacy, define what you're testing before you look at the data. Related: Clustering illusion.

Examples

• We looked at where crimes happened and found this street has more incidents—so we'll focus there.

• We checked which month had the most sales and it was March—so March is our best month.

• We found a cluster of illness in this building, so something here must be causing it.