Last weekend a couple of college roommates and I discussed clutch hitting, if it was real or not. One of my friends suggested that Jean Segura passed the eye test of being clutch. Here’s what I wrote, starting with a look at Paul Molitor:
Here is the link to Molitor’s career splits.
He hit .306 overall in 10835 at bats, .326 with men in scoring position in 2463 at bats. So we can ask the question, what is the probability of a .306 hitter producing at least 804 hits in 2463 at bats.
Here is the Excel formula: =1-BINOM.DIST(803,2463,0.306,TRUE)
So I figure the cumulative probability of getting no more than 803 hits, and subtract that from one. The answer is 0.015. I believe most people would find that significant.
Let’s look at Big Papi. Ortiz batted .286 in 8640 at bats. With runners in scoring position, he batted .296 with 708 hits in 2389 at bats. So his probability is 0.136, which most people would consider not significant. It’s possible to dismiss Ortiz’s clutch hitting as random luck.
Let’s look at a terrible hitter with a somewhat long career, Martin Maldonado. Maldonado hit .209 in 2952 at bats. With runners in scoring position, he hit .216, with 152 hits in 705 at bats. His probability is 0.347. In general, poor hitters are going to get fewer clutch opportunities because they hit lower in the order, behind poor hitters, and they get replaced by pinch hitters in big situations.
And this is the problem. People will look at this and say, “Well not all RISP situations are clutch!” That’s true. So you wind up defining clutch as RISP in a close game from the seventh inning on, and the number of opportunities are so low that it’s nearly impossible to be significant. The splits have a category called close and late, which is seventh inning on, score difference for the batting team from -2 to 1. Base runner situation doesn’t matter. All three have lower batting averages in this situation. The probabilities for Molitor, Papi, and Moldando are .60, .99, and .97.
Just for completeness, here is Jean Segura. He hit .285 over all for his career in 5195 at bats. He hit .279 with runners in scoring position in 1097 at bats. He hit .266 in Close and Late games in 775 at bats. He’s really good with less than 2 out and a man on third base, hitting .358 in 201 at bats. The probability of a .285 hitter doing that, however, is 0.015. So he’s very good in a very specific clutch situation. He also has 24 SF, but in general I would not call him a clutch hitter.
The best clutch hitters are good hitters is the upshot of most studies on the subject.
David Pinto, personal email
I suggested we keep track of clutch situations in the LCS games Sunday night, for us to get a feeling as to what we really consider a clutch situation. By writing down the value beforehand, we would not be influenced by the outcome. It would also force us to seriously think about the question, “What is a clutch situation?” Here is my part of the exercise. I rated every plate appearance from 0.0 to 1.0 by tenths of a point.
I will point out that I thought the biggest clutch moments for the Phillies came in the bottom of the seventh inning with two out. Kyle Schwarber came up with the Phillies trailing by one and a man on second with two out. I rated that a 0.9. The Padres intentionally walked Schwarber, putting Rhys Hoskins up with two on and two out, also a 0.9.
When Bryce Harper came up the next inning with a man on first and no one out, I rated that as a 0.7. It was late in the game, and the Phillies needed a run, but it really wasn’t a make or break moment, like when Hoskins batted in the seventh.
If we were able to get a large number of people to rate PA this way, we could start exploring the type of situations that made a PA clutch. We could average together the values for each PA and come up with a clutchiness index for the player. We could examine how a player hits based on the clutchiness of the situation, etc.
On Thursday, Jayson Stark at the Athletic explored these themes. It was nice to know I was on the same page as Tom Tango:
“All the proclamations about clutch — they’re always after the fact.”
TheAthletic.com
And:
In other words, great players aren’t “clutch.” They’re just “great.” I told him about Rob Thomson’s theory on Jeter. Then I asked: “Was Derek Jeter clutch?”
TANGO: “He’s as clutch in the postseason as he was in a regular season. Because he’s Derek Jeter — and from the time he was probably 12 years old to the time he retired, he was as perfect as he’s always been. And we don’t learn anything new about Derek Jeter just because it’s April or June or October.”
If you have a subscription, the whole Stark article is well worth the read.
I do ask you as you watch Game One of the World Series tonight, think about the clutchiness of the situation. Write down the number. You can even record how the fared, again 0 to 1. I was recording 0 for failure and 1 for success, but Trent Grisham‘s bunt in the ninth threw me. He made an out (failure) but he put two runners in scoring position (success). Maybe that was a 0.5 success. Maybe Schwarber’s IBB was a 0.1 success. An infield hit with a man on second and two out that only advances the runner to third? That’s not as good as a single to the outfield that scores the runner, maybe 0.6 to 1. Give it a try and ask yourself, “What is really clutch?”
