How Baseball Became a Game of Numbers
Exploring baseball's unique statistical nature, plus a glimpse into the life of a night watchman at a Kansas factory.
5 minutes · No politics · Just things worth knowing
Transcript
It's Wednesday, October seventh. I grew up in Milwaukee, so October means Brewers baseball. They're up two games to one on the Padres in the playoffs, with Game 4 tonight. Watching these games, half of what the announcers say isn't about baseball at all. It's numbers. Exit velocity. Launch angle. Spin rate. A win-probability counter ticking up and down on the screen while a batter stands in the box and does nothing.
The fans are just as deep in it. I know people who keep score at games, who track pitchers in notebooks, who can tell you how a reliever has done against left-handed hitters in September. No other sport produces fans like that. Baseball is the one major sport that turned into a math problem. That didn't happen because baseball people love math. It happened because of the shape of the game.
Think about basketball. Ten players on the court at once, all moving, all affecting each other. A screen, a cut, a pass — you can't cleanly separate one from another. Soccer is worse. Twenty-two players flowing around a field with almost no stoppages. The action is continuous, and it's tangled. When a goal happens, you can't point to the one player who caused it.
Baseball is the opposite. It's a series of isolated duels. One pitcher, one batter, everyone else standing still and watching. Then the pitch. Then the ball is hit, or missed, or taken, and the play is over. Then it starts again.
That structure is why baseball produces clean data. Every pitch can be labelled — fastball or curveball, in the zone or out of it, swung at or not. Every plate appearance ends in one of a small number of outcomes you can count: out, hit, walk, strikeout. Every player's contribution is isolated in a way that a midfielder's never is. The sport measures itself, whether anyone asks it to or not.
In the 1970s, a night watchman at a pork and beans factory in Kansas spent his shifts thinking about baseball. His name was Bill James. He wasn't a player, wasn't a coach, had never set foot in a major league front office. He just kept asking one question: which of the numbers baseball already tracks actually tell you who's good?
Baseball at the time judged hitters mostly by batting average and runs batted in. James looked at the data and found both were misleading. Batting average counts a walk the same as a strikeout, as though both were equally worthless — when one puts you on base and the other is just an out. Runs batted in mostly measure how many runners your teammates got on base ahead of you. Neither number was measuring the thing it claimed to measure.
So he started writing it all down. Every year he self-published a book of his findings called the Baseball Abstract. It sold a few hundred copies at first. He kept at it for years, and his message stayed the same: the statistics baseball trusted were measuring the wrong things, and whichever team figured that out first would find bargains — good players that everyone else was misjudging.
He even gave the approach a name, sabermetrics, after the Society for American Baseball Research. For about twenty years, almost nobody inside professional baseball paid attention.
Then one team did. In 2002, the Oakland Athletics had one of the smallest payrolls in baseball, about forty million dollars, and they were competing against clubs spending three times that. The New York Yankees that season spent around a hundred and twenty-five million. Oakland couldn't buy stars. So they went looking for players everyone else had undervalued.
The number they cared about most was one the rest of baseball ignored: on-base percentage — basically, how often a player avoids making an out, whether by getting a hit or by drawing a walk. Baseball had long treated a walk as a kind of accident, a thing that happened when the pitcher was wild. Oakland treated it as a skill, and went and found players who were good at it and cheap because of it.
That season the A's won a hundred and three games, the best record in the American League, on roughly a third of the Yankees' payroll. They also won twenty games in a row, an American League record that still stands. A book called Moneyball was written about that season, and a decade later it became a movie.
Within a few years, the approach was everywhere. Every team hired analysts. Front offices filled up with people who had never played the game. And the numbers started changing how the sport itself was played on the field.
What the numbers said, run honestly, was that the most efficient way to score runs was to stop caring about contact. Walks are good. Home runs are good. Strikeouts don't matter much, because an out is an out. So teams optimized for exactly that: work the count, swing for the fences, accept the strikeout. Baseball people called it the three true outcomes — walk, strikeout, home run. The three results where the defense never touches the ball.
It worked, and it was boring. Games filled up with strikeouts and walks and long at-bats where nothing happened. Fielders stood around, because the ball was either hit over their heads or not hit at all. By 2022, the average game had stretched past three hours, and the ball was being put in play less than at any point in the modern game.
So in 2023, Major League Baseball did something it almost never does. It changed the rules.
They added a pitch clock — fifteen seconds between pitches with the bases empty, twenty seconds with runners on. They banned the defensive shift, the trick where a team slides three or four fielders to one side of the infield to smother a left-handed pull hitter. They made the bases slightly bigger, which made stealing more tempting.
The effect was immediate. The average game dropped to about two hours and forty minutes, the shortest in decades. Stolen bases jumped roughly forty percent in a single season. More balls were hit, more runners moved, and the game started to look more like the version people remembered.
Tonight the Brewers play Game 4 with a chance to move on, and the broadcast will be full of numbers that didn't exist twenty years ago. Most of them trace back to a night watchman in Kansas with a notebook, who kept asking which ones actually mattered.
Stay informed, stay curious, and we'll see you tomorrow.
Prefer your podcast app?
Or wherever else you get your podcasts.
☕ Get today's briefing in your inbox
5 minutes every morning. Interesting things happening in the world — not politics. Unsubscribe any time.
Want streak tracking and saved preferences?