Game Simulations Answer Baseball’s What-Ifs

Alan Schwarz –

I think the most integral bit of sabermetric revelation is Bill James’s Pythagorean Theorum, which converts team runs scored and team runs allowed into estimated won loss records. From this single metric one can seemingly assess how much success in the game is tactical and how much is brute force.

But it turns out that this inspired bit of arithmetic may conceal all the interesting parts of the game. The Theorum describes the average results of all the players, and how they convert to a won loss record. It’s all a little mechanical, even though the math is pretty irrefutable.

The interesting part arrives in somewhat recent James backtracking about Clutch hitting, and what Tom Tippett says in this story about baseball simulations. It’s possible for the numbers to add up to no effect, but that doesn’t mean there is no effect.

After a bad beat at the poker table, I used to go home and run a million iterations of the hand in Turbo Texas Hold ‘Em. What I found out, was that I usually had table stakes, but I no longer had the money.

For individual players, performance matters. It’s impossible for me not to imagine that some players are more clutch, some might be less so. But I once directed a commercial video that starred Michael Jordan, and the point he made was the clutch players made the big plays because they were the best players. They got more chances, and their successes were remembered.

I’ve always used that as the example of how a clutch player denigrates clutch performance, but it doesn’t necessarily mean that.

For managers, the sacrifice bunt, the stolen base, the batting order (as Schwarz explores here) are their tools. Do they matter? Absolutely, sometimes. Sometimes, not.

So, while it’s possible to flatten the landscape using the historical data, for someone like Tony LaRussa, who is in the game, the job is all about getting his players into a winning frame of mind, to make sure the pitcher bats eighth, and to suggest that the game is all human. For him, it’s all about the individuals, because we know the best team, as defined by the numbers, doesn’t always win.

The biggest point about Schwarz’s story is that Game Simulations Don’t Answer Baseball’s What-Ifs.