Beyond the Numbers

Chapter 1: The Data Revolution

In 1977, a CBS statistician named Seymour Siwoff introduced something called the “Ultimate Rating System.” It was a simple formula that combined a few counting stats—passing yards, rushing yards, turnovers—and spat out a single number meant to summarize how good a team was. He printed it in a media guide, and then something remarkable happened: nobody cared.

That’s the honest origin story of football analytics. Not a lightning bolt of insight, but a spreadsheet nobody read. For decades, the sport’s relationship with data was essentially ceremonial. Coaches kept paper charts of down-and-distance tendencies. Scouts timed forty-yard dashes with stopwatches. Beat writers tallied “big plays” by hand. The numbers existed, but they were descriptive, not predictive—a way to describe what had already happened, never to anticipate what would happen next.

Then the internet happened, and with it, a small tribe of obsessive fans with too much time and a pathological need to know why their team kept losing on third-and-long. They weren’t professionals. They were message-board denizens who discovered that publicly available play-by-play logs—once the lonely province of stats nerds and gambling degenerates—contained more information than anyone had ever bothered to extract. They began asking questions that the box score couldn’t answer. Does it matter if you gain four yards on first down? Does a sack on third down hurt more than a sack on first? Is a turnover in your own red zone worth the same as one in the opponent’s?

The answers required a