We Need to Talk About Clutch Hitting

For the bulk of this season, all we seemed to talk about was how the Twins’ Achilles heel this season was the pitching staff. Through the August 2 trading deadline, Twins pitchers had a collective 4.04 ERA / 4.21 FIP and ranked 27th in pitching fWAR.

If only they had better pitching, they could run away with a wide-open AL Central, we said.

Part of that line of thinking was predicated on the idea that the offensive lineup was strong. In fact, it was. Prior to the trade deadline, Twins hitters had combined for a .251/.321/.419 line that worked out to 114 wRC+, a mark that ranked as baseball’s 4th-best offensive output.

Then the trade deadline came and the Twins were buyers of pitching, addressing their most obvious need. And it made a difference!

Following the deadline, Twins pitchers have combined for 3.36 ERA / 3.40 FIP marks and are tied for 9th in pitching fWAR.

Yet, the team has just an 11-12 record over that span. That tough stretch turned a one-game division standing lead into a two-game deficit following the weekend series with San Francisco.

That record was made possible by a prolonged and nearly team-wide offensive slump. In the last 23 games, their production at the plate has plummeted to .238/.307/.374 and 96 wRC+, which is ranked 16th. Not only has the overall production with the bats stalled, but the team has consistently failed to cash in the few run-scoring opportunities they have created. Minnesota averaged 3.91 runs scored per game over that stretch, significantly less than the 4.52 they averaged before the trade deadline.

Many have correctly pointed out that the team has had trouble hitting with runners in scoring position (RISP). That’s been a bit of an issue most of the season, as you can see in the numbers below:

Data from FanGraphs

That they’ve hit slightly worse with runners in scoring position than they have overall helps explain how a team with baseball’s 4th-best wRC+ and 6th-best on-base percentage could rank 13th in runs scored heading into the trade deadline.

During their post-trade deadline slump, that problem has gotten much worse:

Data from FanGraphs

As a group, they slumped to worse than the league average overall and fell off even more dramatically with runners on base and runners in scoring position. And, as we’ve all noticed when taking in these games, their production in high-leverage spots with runners in scoring position has been downright abysmal. That 8 wRC+ mark ranks last in the majors over that span, 40 points worse than Kansas City’s 48.

All together for the season, the Twins are tied for 17th in turning their baserunners into runs. They’ve done that at a 30% rate, lower than the 31% league average and closer to the last-place Nationals’ 27% than the league-leading Dodgers’ 35%.

FanGraphs maintains a statistic called Clutch that evaluates how well a player (or team) performs in high leverage spots relative to how they otherwise perform, using win probability and leverage. Clutch is a purely descriptive stat. It’s useful for understanding what has happened but not necessarily for predicting the future.

Prior to the trade deadline, Twins batters ranked 20th in clutch with -0.77. Said differently, the team had lost about three-quarters of a win as a result of their batting performance in high-leverage spots. Since the deadline, they rank 29th with -1.76. That’s actually improved thanks to the weekend sweep of the Giants. From August 3 through August 25, they ranked last in baseball with a -2.50 clutch score, more than a full win worse than the next closest team (Angels, -1.47).

Naturally, losing two and a half wins solely from performance in important situations leads to a lot of frustration. In the fanbase, that frustration reveals itself in commentary questioning the team’s approach, their experience (or lack thereof), their mental toughness, and their “will to win,” among numerous other things. It also brings about calls for the manager and front office to be fired or for certain players to never be given a uniform again. All of those kinds of sentiments might be grounded in thinking that performance in the biggest spots is something over which the players and coaches have control.

But here’s the thing: they really don’t. No evidence has been found that clutch hitting (or clutch anything) is a talent or repeatable skill. There are clutch hits and players that come through in clutch moments, obviously. There are teams that experience great timing and come up with lots of clutch hits over certain stretches of time. And clutch plays have an important role in determining wins and losses.

But, the ideas that some players can consistently elevate their abilities in the biggest moments or that some teams have a purposeful knack for stringing together clutch hits are myths that perpetuate, probably because they make for good stories.

Much like has happened with the hot hand fallacy, there have been plenty of analyses completed over the past several decades that have concluded that the tales of the clutch hitter or pitcher do not hold up to fact-based scrutiny.

Two of the players most often held up as paragons of “clutch” in my baseball-consuming lifetime have been Derek Jeter and David Ortiz. Their larger-than-life reputations for heroics in the biggest moments come mostly from memorable high-leverage October playoff situations in which they did come through. In his book Smart Baseball, Keith Law examined these two legends’ numbers in various clutch scenarios and found the numbers don’t support the narrative.

To illustrate: Jeter’s career triple slash line was .310/.377/.440 (.817 OPS). In situations with two outs and runners in scoring position (inarguably one of the most clutch situations the game has to offer) Jeter hit .298/.399/.418 (.817 OPS) – more or less the same level and shape of production as you would expect from his overall numbers. To that point, Jeter’s career total clutch score is 0.48. That’s a positive mark for sure, but if Jeter were really preternaturally clutch, wouldn’t his clutch performances have been worth more than half a win over 2,747 career games? In fact, in twelve of Jeter’s twenty career seasons, he had a negative seasonal total clutch score.

Ortiz’s career line was .286/.380/.552 (.932 OPS). With two outs and runners in scoring position, he hit .256/.417/.534 (.951 OPS), a mark that is slightly above his overall career OPS but also explainable by a higher on-base percentage that was helped by opponents frequently pitching around him in such spots. His career total clutch score was -1.16 and he had negative seasonal marks in thirteen of his twenty career seasons. Nine of those negative seasons came while playing for Boston, where his legendary clutch reputation was birthed and took hold.

None of that is meant to say that Jeter and Ortiz were not great hitters or that they didn’t come through in the clutch situations that we remember. They were and they did. But, they were just great hitters in any situation, not *even more* great in clutch spots.

Those two examples are consistent with the conclusions reached by other, much larger sample analyses. Dick Cramer wrote the seminal study back in 1977 and its main conclusion has been supported by many other attempts to find evidence of consistently clutch hitters, including work in The Book (Chapter 4) and an extensive study with data from 1960 to 2004 by Tom Ruane.

Now, the absence of evidence is not evidence of absence. But at some point, it's reasonable to accept that if a skill for being clutch exists, its effects are very weak or that truly identifying players that have the skill is like trying to find a needle in a haystack.

At a team level, Jeff Sullivan of FanGraphs has analyzed team clutch performance data from 1974 through 2017 and found no relationship (literally, an R^2 of 0.00) between one year’s clutch performance and the next:

Jeff Sullivan, FanGraphs

Sullivan wrote, “It stands to reason, I think, that if Clutch is a real skill, we should see it hold up over time.” Yet, he concludes from this data, “Clutch teams in one year haven’t gone on to remain clutch teams the next. Unclutch teams in one year haven’t gone on to remain unclutch teams the next.” He also looked within seasons, comparing first-half clutch to second-half clutch, and again found no relationship.

None of those findings mean clutch hitting is not important. Clutch does help to explain teams over- or under-performing expectations. One way to think about Clutch is that it measures how well a team (or player) has timed their performance. If a team has had great timing, (i.e., it has gotten hits in big spots) it will likely over-perform its expected records as estimated by things like the Pythagorean Theorem of Baseball or BaseRuns.

Sullivan found a fairly strong relationship (R^2 = 0.54) between team Clutch and the difference between its actual wins and BaseRuns estimated wins on a dataset from 2002 through 2014. Sullivan wrote, “[c]lutch can turn a mediocre team into a playoff team. Clutch can also turn a would-be playoff team into a mediocre team.”

The 2022 Twins, who, as of this writing, are underperforming their BaseRuns projected record by one game might be finding this out the hard way.

The good news, in light of the Twins' recent slump, is that no one has found evidence of past performance in clutch situations being predictive of future performance in similar spots.

That’s encouraging for someone like Carlos Correa. A lot has been made on social media about Correa’s clutch hitting this season. Before this season, he had hit .278/.365/.467 (122 wRC+) and 33 home runs in 984 career plate appearances with runners in scoring position. This season, in 94 plate appearances with runners in scoring position, Correa has a .256/.340/.321 (88 wRC+) line and no home runs.

Given the findings described above, we shouldn’t put too much stock in this depressed production. We should trust that Correa hasn’t suddenly lost his ability in pressure-packed situations and that things will eventually even out around his usual level of production.

Correa’s 2022 is an example of the random variation that occurs over the course of a career. It’s also an example of unfortunate timing, like this situation from Friday’s game against the Giants shows:

None of this is to suggest that hitting or pitching in clutch situations is not important. It’s very important! It can be the difference between making the playoffs or not. It’s also not to say the fans, players, and coaches should not be frustrated with the results. Watching the season potentially slip away from missed chances is excruciating.

It’s just that there is no historical evidence to suggest players and teams can do much of anything about it. There is no approach, practice drill, or heart of a champion supplement to make themselves more clutch. If there were, teams would build rosters full of clutch players and the plot from Sullivan above would look a lot different.

For the Twins this season, that’s both good and bad news. On one hand, their struggles in clutch spots are almost certainly not representative of a fatal flaw or lack of ability. Their big moment fortunes could turn around overnight. But on the other hand, there is not an obvious fix or adjustment they can employ to force the issue. In any event, they just need to play better and hope things quickly even out in big situations. The fate of their season might depend on it.

---

John is a writer for Twinkie Town and Pitcher List with an emphasis on analysis. He is a lifelong Twins fan and former college pitcher. You can follow him on Twitter @JohnFoley_21.