The Dodgers’ Biggest Problem
Almost two months ago, I talked about how the Mariners were on a record pace for team Clutch. This is never a fun statistic to explain, since it’s rooted in win probability, which is already complicated enough, but in short, Clutch measures whether a player or team has done better or worse than expected in higher-leverage situations. A player who knocks in the game-winning run will have a high Clutch score for the day. The opposite would be true of the pitcher. The stat is hard to explain in a paragraph, but it still manages to be intuitive, if that makes any sense.
Since that post was written in early July, the Mariners have slumped and fallen well out of playoff position. Nevertheless, they’re still on pace to finish with the highest team Clutch score since 1974, which is as far back as our data goes. If you want to understand how exceptional the Mariners have been, you might consider this plot of all 30 team Clutch scores:
The Mariners are way out in front, with five extra wins even just compared to the next-most clutch team. Clutch performance explains why the Mariners have been able to overachieve their underlying numbers. But, you know, let’s look at that same plot again. Let’s just change what we highlight.
We can use this to talk about the Dodgers, too. Like the Mariners, the Dodgers presently find themselves several games removed from a playoff spot. Unlike the Mariners, the Dodgers were supposed to be good.
The Dodgers were recently swept at home by the Cardinals, and you could call it a representative series. While all three games were competitive enough, the Dodgers were undone by poor clutch performance. In the first game of the series, which was Kenley Jansen’s return from the disabled list, Jansen allowed a pair of ninth-inning home runs. Just before, when the game was tied in the bottom of the eighth, Justin Turner struck out with one out and a runner on third. Cody Bellinger lined out with two out and runners on second and third. In the last game of the series, Jansen allowed a two-run ninth-inning homer. Just before, when the game was tied in the bottom of the eighth, Manny Machado grounded out with two out and the bases loaded.
On the one hand, it’s not entirely fair to focus on individual half-innings, since the season is made up of so many of them. But as we know from leverage numbers, not every plate appearance is created the same. Some plate appearances mean far, far more than plate appearances earlier on, and when the stakes have been high this season, the Dodgers just haven’t looked like themselves. At least, they haven’t played like themselves. I mentioned earlier that the Mariners are on track for the highest team Clutch score since 1974. Here’s a table of the *bottom* team Clutch scores:
Team | Season | Batting | Pitching | Total | Games | Total/162 |
---|---|---|---|---|---|---|
Phillies | 1984 | -5.4 | -7.3 | -12.7 | 162 | -12.7 |
Rockies | 2001 | -5.7 | -7.0 | -12.7 | 162 | -12.7 |
Yankees | 2017 | -7.1 | -4.9 | -11.9 | 162 | -11.9 |
Dodgers | 2018 | -6.9 | -2.2 | -9.1 | 128 | -11.5 |
Padres | 1994 | -6.3 | -1.8 | -8.2 | 117 | -11.3 |
Mets | 1993 | -2.4 | -8.7 | -11.2 | 162 | -11.2 |
Braves | 1989 | -4.7 | -6.4 | -11.1 | 161 | -11.1 |
Dodgers | 1982 | -6.1 | -4.9 | -11.0 | 162 | -11.0 |
Orioles | 1988 | -2.6 | -8.4 | -11.0 | 161 | -11.0 |
Orioles | 1995 | -4.7 | -5.1 | -9.8 | 144 | -11.0 |
If the Dodgers were to keep this up, they’d finish with the fourth-worst Clutch score on record. Of course, given the extreme nature of where the Dodgers are, we wouldn’t expect them to remain quite so bad, but much of the season is already over. The Dodgers already have a team Clutch score of -9.1, even without extrapolating over 162 games. Only 26 teams have finished worse, over four and a half decades. The Dodgers and the Mariners have had kind of opposite seasons.
This is why the Dodgers are where they are in the standings. They’re on the outside of the playoff picture, looking in. The Dodgers have baseball’s 14th-best winning percentage. Yet, they have baseball’s second-best rest-of-season team projection. They have baseball’s fourth-best current BaseRuns winning percentage, and they have baseball’s fifth-best current run differential. By almost every indicator, the Dodgers look like the super-team they were expected to be. They’re just not there in wins, because they haven’t shown up with the game on the line. Clutch hitting and clutch pitching — or the lack thereof — might well send the Dodgers home at the end of the regular season.
It’s not like this is something that’s just in the Dodgers’ DNA. Just last season, the Dodgers finished with the seventh-highest team Clutch score in baseball. But every month this season has been a bad one, in Clutch terms. Here’s where the Dodgers’ monthly Clutch scores have ranked, in batting, pitching, and overall:
Month | Batting | Pitching | Total |
---|---|---|---|
April | 25 | 13 | 22 |
May | 23 | 24 | 24 |
June | 25 | 7 | 20 |
July | 21 | 17 | 20 |
August | 29 | 30 | 30 |
This has been the most agonizing month yet — first, when Jansen was sidelined, and then when Jansen came back. This month in particular, the bullpen has been a liability, but what’s been less obvious is that the supposedly stacked lineup, too, has fallen short. The Dodgers’ lineup hasn’t been clutch in any single month. This year, the pitchers deserve some of the blame, but the lineup has been a nightmare when the stress level has risen.
According to Baseball Reference, the Dodgers have a .768 OPS in low-leverage situations. In medium-leverage situations, that’s gone up to .803. But in high-leverage situations, the Dodgers have a combined OPS of .635. There’s a stat called tOPS+, that compares a player or team’s split performance to the player or team’s overall performance. The Dodgers have a high-leverage tOPS+ of 71, where 100 would be average. That’s the worst such tOPS+ in the majors. It doesn’t get any better when you examine all-time recorded history for context.
Team | Year | PA | tOPS+ |
---|---|---|---|
Giants | 1953 | 1005 | 63 |
Phillies | 1934 | 862 | 70 |
Dodgers | 2018 | 1064 | 71 |
Tigers | 1929 | 752 | 72 |
Athletics | 1994 | 804 | 74 |
Cubs | 1954 | 1067 | 75 |
Senators | 1939 | 564 | 76 |
Cubs | 1933 | 1041 | 77 |
Tigers | 1940 | 570 | 77 |
Red Sox | 1945 | 828 | 77 |
This clutch hitting has been historically bad. The worst baseball’s seen since 1953, by this measure. The Dodgers’ high-leverage tOPS+ last year was a perfectly normal 100. It’s not an easy thing to explain, but I guess you don’t really have to explain it. It just is — what’s already happened has already happened. Blame isn’t shared by everyone equally. Matt Kemp, for example, has a positive Clutch score. So does Enrique Hernandez. So did Logan Forsythe. But Joc Pederson‘s Clutch score is strongly negative. Cody Bellinger’s is the sixth-worst in baseball. Max Muncy‘s is the fifth-worst. Yasmani Grandal’s is the fourth-worst. It’s never simple to tell what might be signal and what might be noise in this area, but for the Dodgers, the damage has clearly already been done. With just a few more clutch hits at the right times, the current Dodgers would probably be in first place.
It’s some consolation, and it isn’t. The silver lining is that it seems like the Dodgers should have a better record. It seems like the Dodgers haven’t actually come apart; the parts are still there of a National League super-roster. This doesn’t seem like a failure of ownership, or the front office, or the coaching staff. It’s a weird and almost inexplicable problem of timing. But it’s a problem, and a major problem, and it can’t be erased or undone. The Dodgers have been historically unclutch over the course of 128 games. There are only 34 more games to go, and they can’t afford to give up more ground. Because the Dodgers haven’t performed with the game on the line, it’s looking less and less likely they’ll play in the playoffs. Theirs are the players who haven’t hit well enough. Theirs are the players who haven’t pitched well enough. They’ve done well enough to look plenty strong overall, but they haven’t done well enough when it’s mattered. That partially helps to explain what sank a super-team in DC. It explains even better what’s been sinking a super-team in LA.
It’s not too late, but it’s getting there fast. For the Dodgers, right now, the pressure is on. And they haven’t done well under pressure so far.
Jeff made Lookout Landing a thing, but he does not still write there about the Mariners. He does write here, sometimes about the Mariners, but usually not.
I’d be curious if there any correlation with a player’s clutch score and having a big platoon split. Opposing managers will take advantage of that when leverage is at it’s highest.
Agreed. Opposing managers exploiting matchups + defensive shifts = lower performance. Obviously a team with less platoon/pull-heavy types is liable to have a higher clutch score.
And to add to that, on the other side teams with shutdown lefty specialists, etc. probably can use that in high leverage situations as well to improve clutch score.
I made this point to Jon Weisman yesterday on Twitter. Just anecdotally it seems like platoon players are more streaky as well, and I believe there’s some kind of pinch hitting penalty too. The Dodgers have used a ton of pinch hitters late in games to get the platoon advantage, but I’d imagine the success rate isn’t as high in those situations this year as it was last year. Probably some randomness involved there too.
Sort of. Certain types of players are more likely to get a new reliever to take advantage of a platoon split. But it’s the good players. This is why “Clutch” isn’t a useful statistic for hitters. Also, the statistic is calculated in a way to punish good hitters. It’s hard to separate out those two effects, but they’re both in there.
It’s not exactly surprising that the Grandal, Bellinger, and Muncy hitters are the three least clutch guys on the team. Clutch uses overall performance as the denominator, meaning that the best players have to be total bosses in higher leverage spots to make up ground. It’s a lot easier for guys having lousy hitting seasons (like Forsythe was) to gave good “Clutch” scores.
To give you an idea, here are the least Clutch players this year, and their wRC+’s:
1 Giancarlo Stanton -2.20 139
2 Mike Trout -2.05 192
3 Willson Contreras -2.03 111
4 Yasmani Grandal 1.99 129
5 Cody Bellinger 1.79 120
6 Trey Mancini -1.64 85
7 Mallex Smith -1.49 127
8 Marcus Semien -1.42 96
9 Jose Ramirez -1.34 169
10 Kyle Schwarber -1.29 111
What do we learn about this list? Three of the four best hitters in the league are in the Top 10 (the fourth, FWIW, is JD Martinez, who has been moderately “clutch”). And eight of the 10 are above the league line–hopefully I don’t need any fancy tests to show that there is a relationship between wRC+ and “Clutch”. Some of this, I strongly suspect, has to do with the fact when a big homer guy (and all these guys are big homer guys) comes up to bat in a high-leverage situation, they are far more likely to get a new pitcher to come who is fresh (and hopefully takes advantage of platoon splits).
Further give you an idea of how worthless Clutch is, let me show you how much of a difference there is between wRAA and RE24. RE24 is like Clutch, but it actually makes sense–it adjusts wRAA for performance depending on how many players are on-base and how many outs there are. The Dodgers have a wRAA of 37.5, good for 7th in the league. The Dodgers have an RE24 of 31.57, good for 5th in the league. That’s right–the Dodgers actually have gotten more runs out of their performance based out and runners on base than most teams do!
This is pretty alarming. If I was Dave Roberts, I would print out this article and post it all over the locker room.
I’m sure there’s nothing better for clutch performance than for all the players to start thinking about their clutch performance.
You could look at it one of two ways:
1. A simple call to action
2. Take a deeper dive into the numbers to reassure the team and tell them to keep their heads up. Positive regression is coming (like the Cards did to Matt Carpenter earlier this season). That said, it would be very interesting to really dig down and look at how BABIP and batted ball profiles change with runners on or in scoring position. Could do the same with pitchers…. That would tell us if it is more of a mental problem that requires addressing or just bad luck which you might as well sweep under the rug (a la Matt Carpenter).
This is unacceptable when you take into account the talent level. I mean seriously…. What do you have to lose?
I disagree with that because the more aware you are of the issue the better chance you have of correcting it in my opinion.
I also find it interesting that of the least clutch players Jeff mentioned, all seem to have another thing in common – all of them get shifted on as much as anybody in baseball. The shift must be tough enough in low-leverage situations, but when you are thinking about beating the shift AND getting the run in – that might help explain the low performance.
I don’t think this makes sense. High-leverage situations are more luje with men on base. There are fewer shifts with men on base. You would think think that if anything batters who get shifted on a lot would have above-average clutch scores.
That’ll turn it around for sure.
I would direct you to this comment when discussing whether just anyone is capable of managing a baseball team
Yeah because there is clearly only one way to manage a baseball team.
But how will it help? As far as I know Clutch is almost always entirely influenced by luck and has no predictive value.
It’ll be like, “Hey, look, you guys suck at this.”
“Damn, coach, how do we fix it?”
“I dunno, be more clutch obviously.”
Not a super productive conversation.
Mr. Burns: You, Strawberry, hit a home run.
Darryl: OK, skip.
[Strawberry hits a homerun]
Mr. Burns: I told him to do that.
Smithers: Brilliant strategy, sir.
Mattingly still likes him better than Steinbrenner.