Why Is Hitting Worse in Extra Innings?

Watching the Yankees last month, the team was scuffling in a four game series against the Baltimore Orioles. They lost three out of four in that series, including a 14-inning loss and a 12-inning loss. During the second of those extra inning losses, Michael Kay on the broadcast asked Paul O’Neill why hitters seemingly get worse in extras. Specifically, he hypothesized that hitters went up to the plate and were thinking about the one big swing that could send everybody home right then. O’Neill agreed, and I couldn’t help but wonder about if it were true.

First things first, it’s not just an illusion that teams hit worse in extra innings. From the start of 2017 to now, the Yankees had a .340 team wOBA in innings one through eight. In the 9^th inning plus extra innings with the go-ahead run at the plate (in other words, a home run would end the game), their team wOBA dropped down to .250, which is a gigantic difference. It’s not the just the Yankees who see a drop off in extras. The league as a whole had a .322 wOBA in the first eight innings and a .312 wOBA in the 9^th inning onward with the winning run at bat. It isn’t nearly as drastic as the Yankees, but the difference is still there in a much larger sample of 30 teams than just one team.

So is the long ball mentality the source of this? Well, home runs aren’t being hit at a higher rate in the 9^th inning onward. But surely they can still be trying to hit home runs (or at least more so than usual). So, I check out the differences in some metrics that would factor in a homer-happy mind state: strikeout and walk percentages, products of how free-swinging and violent-swinging they may be, flyball and pull rates, and average launch angles.

Besides strikeout rates, none of these stats significantly increase. In fact, some of them even decrease. If hitters really are trying harder to go deep in extras, I’m certainly not seeing it. Unfortunately I couldn’t find data for some other discipline stats like swing rates and swing rates on pitches out of the zone, but there still is some value in that strikeouts themselves increase heavily. The higher strikeouts is probably going in hand with the overall worse hitting, and if it isn’t because their home run intentions, then we still don’t know why hitting gets worse. Are the pitchers pitching in extras just better?

Average runs per inning steadily drops off after the sixth inning. It’s clear why: teams take out the starter around this time and bring in extra effective relievers, usually with them increasingly getting better by the inning as their specialized inning approaches the ninth inning for closers. It’s tough for me to say that extra inning offense is worse because of effective relief pitching. Old school thinking had teams saving their best relievers only for when the team was winning in save situations, thus leaving them available for extras. In this scenario it makes sense, but I imagine nowadays most teams have already used their best relievers by the time it reaches extras, especially considering you know the game must have been pretty close if nine innings weren’t enough. Could it be that with deep bullpens, even if the relievers pitching in extras aren’t one of the elite set-up men or closers, that they’re still more effective most?

I’d have to imagine this is the case. Case study: New York Yankees at Chicago Cubs in 2017, a game decided by a long 18 innings. By the time the game went extras, Dellin Betances and Aroldis Chapman, the Yankees’ two best relievers at the time, were already used. The pitchers in the extra innings were Tyler Clippard, Adam Warren, Jonathan Holder, and Chasen Shreve. Their ERAs that season, respectively, were 4.77, 2.35, 3.89, and 3.77. League average ERA in 2017 was 4.36. Clippard of course has the great track record anyways, but the other three guys will likely never be a glorious set-up man or closer. Yet compared to league average, they were still above average. In a 19-inning game last season, the Red Sox had already used Joe Kelly, Addison Reed, and Craig Kimbrel, but got the job done with guys like Brandon Workman (3.18 ERA), Austin Maddox (0.52), Heath Hembree (3.63), Robby Scott (3.79), Blaine Boyer (4.35), and Fernando Abad (3.30). Mediocre relievers are often pitching in extra innings, but hey, they’re still good pitchers. I still think they’re good enough to be the reason why offense slows down in extras.