What is the Texas sharpshooter fallacy?

Cherry-picking a cluster in the data and drawing a target around it afterward — treating coincidental patterns as if they were meaningful.

Named for a shooter who fires randomly at a barn, then paints a bullseye around the tightest cluster of holes to look like a crack shot. The fallacy is ignoring the misses and the randomness, focusing only on a chance grouping, and then claiming it proves a pattern.

"This town has a higher cancer rate than its neighbours, so the local factory must be causing it."

If you scan enough towns, some will show clusters by pure chance — singling out the high one and inferring a cause ignores all the towns where nothing unusual happened. The "pattern" was selected after the fact to fit a conclusion. (LaBossiere's related spotlight fallacy is its media cousin: assuming the cases that get the most coverage represent the whole group.)

Tip: The giveaway is that the hypothesis is drawn after spotting the cluster, with the non-matching data quietly discarded.