Winning is random

You aren’t what your record says you are

Put simply, 42% of an NFL team’s regular season record can be accounted for by randomness, otherwise known as sample error. The short 16-game season is too small of a sample to provide much confidence that team records accurately reflect their ‘true’ level of ability. The more games in a season, the smaller the sample error, and the more certain we could be that the teams with the better records are truly the better teams. (Please note I am not advocating a longer season. The purpose of the NFL is not a scientific experiment to clinically determine the best team.)

Arrow to top