Statistics is not Science

One of the worst things about the fake nerd movement is the belief that statistics are the same as science. Science certainly uses statistics for all sorts of things. Correlations can narrow the search for causal relationships. But, you have to use other tools to reveal those links. That’s no more obvious than in how the sabermetrics crowd completely missed the steroid era. Bill James, the godfather of baseball stat-nerds, was silent on the steroid era. You would think his spreadsheets would have revealed to him what everyone noticed from the stands.

This story on Grantland is another fin example of missing the forest for the trees.

One of the things that makes it such a joy to watch the Chicago Cubs’ rebuilding plan unfold is that the team’s approach is completely transparent. There’s no trickery here, no deceit, no super-secret process that’s inscrutable to everyone outside of the front office.

I don’t simply mean that the Cubs are rebuilding with complete conviction; under the terms of MLB’s collective bargaining agreement, that’s really the only way to go.1 Nor do I mean that the Cubs are nearly the extremists that the Houston Astros are. I’m referring instead to the core principle with which the Cubs have been trying to build a championship roster since team president Theo Epstein and general manager Jed Hoyer were hired after the 2011 season, a principle that distinguishes this rebuilding project from almost every other one in baseball history: They’re building an offense from within and a pitching staff from spare parts.

This flies in the face of more than a century of conventional baseball wisdom, which states that (1) pitching wins championships, and (2) a team can never have too much pitching. The Cubs’ approach is completely counterintuitive. It’s also completely right.

Again, no mention of steroids. From World War II through the eighties, you followed a well known template to build your team. Power at the corners, defense up the middle, speed in the outfield. Mark Belanger could start at short on a title team with a .228 batting average. Elrod Hendricks could make a career as a catcher, despite a .220 career average. Pitching was a given. It was starts and innings you wanted from the rotation. That 1971 Oriole team had four starters account for 1080 innings.

Then the steroid era happened. Suddenly everyone in the lineup was a fearsome slugger. Pitchers were getting killed. That changed how teams looked at pitching. Getting hitters was easy. Getting pitchers that could give you 30 starts was rare. Every team shifted resources into getting and developing pitchers. Teams would draft nothing but pitchers some years. Technology was brought to bear to help pitchers compete with hitters who were jacked up on steroids.

Now, the steroids have gone away. The stat guys have not noticed, but front offices have noticed. The Red Sox traded four pitchers for hitters at the deadline. They just signed a Cuban slugger. The Cubs are doing the same thing. They traded their two best pitchers for hitters. Those teams that acquired the pitchers, by the way, are now struggling. Oakland has slumped and Detroit is fading. There’s plenty of pitching to be had these days. It is hitting that is rare.

3 thoughts on “Statistics is not Science

  1. Steroids, shmerodis! The problem is the umpires, who have shrunk the strike zone to favor the batters over the pitchers. Open up the strike zone, restore it to what it used to be, and the pendulum would swing the other way from hitting back to pitching.

  2. Thomas Sowell pointed out that facts are not truths, they are only arguments in a search for truths.

Comments are closed.