Peter Dizikes of the MIT News Office highlights one of the finalists in the MIT Sloan Sports Analytics Conference, currently being held in Boston. The paper, “A Data-driven Method for In-game Decision Making in MLB”, was developed by John Guttag and Ganeshapillai Gartheeban.
Guttag and Gartheeban’s developed a model that suggested when managers should take their starting pitchers out of the game by using data from the first 80 percent of the 2006-10 seasons to develop a model. Then they tested their model against the results in the final 20 percent of the season.
Dizikes summarizes some of the important findings:
The study finds that from the fifth inning on, in close games, pitchers who were left in games when the model recommended replacing them allowed runs 60 percent of the time, compared to 43 percent of the time overall.
Over 21,538 innings, the Guttag-Gartheeban model disagreed with the manager’s decision regarding his starting pitcher 48 percent of the time. About 43 percent of the time, the manager left the starting pitcher in when the model indicated he should be replaced. In just 5 percent of the cases did managers pull starting pitchers when the model suggested they should stay in the game
Guttag and Gartheeban do note that their methodology doesn’t consider every factor and recognize that the manager is considering a lot more than just the current game situation. Hypothetically, leaving Justin Verlander in with a 110 pitch count in a 2-1 game in the bottom of the eighth in late September when the Tigers are only in the lead by a game could be justifiable when most objective models would suggest taking him out.
That being said, many managers can improve simply by not having their pitchers throw unnecessary innings, like the seventh inning after their team has taken a 10-1 lead.
The Rays started Sergio Romo on back-to-back days and if that sounds weird to you, you’re not alone. Romo, of course, was the star closer for the Giants for a while, helping them win the World Series in 2012 and ’14. He’s been a full-time reliever dating back to 2006, when he was at Single-A.
In an effort to prevent lefty Ryan Yarbrough from facing the righty-heavy top of the Angels’ lineup (Zack Cozart, Mike Trout, Justin Upton), Romo started Saturday’s game, pitching the first inning before giving way to Yarbrough in the second. Romo struck out the side, in fact. The Rays went on to win 5-3.
The Rays did it again on Sunday afternoon, starting Romo. This time, he got four outs before giving way to Matt Andriese. Romo walked two without giving up a hit while striking out three. The Angels managed to win 5-2 however.
Despite Sunday’s win, Cozart wasn’t a happy camper with the way the Rays used Romo. Via Fabian Ardaya of The Athletic, Cozart said, “It was weird … It’s bad for baseball, in my opinion … It’s spring training. That’s the best way to explain it.”
It’s difficult to see merit in Cozart’s argument. It’s not like the Rays were making excessive amounts of pitching changes; they used five on Saturday and four on Sunday. The games lasted three hours and three hours, 15 minutes, respectively. The average game time is exactly three hours so far this season. I’m having trouble wondering how else Cozart might mean the strategy is bad for baseball.
It seems like the real issue is that Cozart is afraid of the sport changing around him. The Rays, like most small market teams, have to find their edges in slight ways. The Rays aren’t doing this blindly; the strategy makes sense based on their opponents’ starting lineup. The idea of valuing on-base percentage was scoffed at. Shifting was scoffed at and now every team employs them to some degree. Who knows if starting a reliever for the first three or four outs will become a trend, but it’s shortsighted to write it off at first glance.