Why has Craig Kimbrel been slightly less dominant lately?

Note: This post was written before Tuesday’s game. Thus, the numbers below do not reflect the three-strikeout performance against the Rays.

The Red Sox bullpen has been phenomenal this season and is one of the big reasons they find themselves in first place despite inconsistent — to put it kindly — performance from the lineup all year. With a 2.94 ERA on the season, only the Dodgers have been better at preventing runs than Boston’s relief corps. When you add in the fact that Boston plays in the more difficult league in a more difficult home park, it’s probably fair to say they’ve had the best season of any bullpen.

Craig Kimbrel has, of course, been the biggest reason they’ve been such an effective unit. The Red Sox closer has been nothing short of incredible this season, putting up one of the best relief seasons in recent memories and re-establishing himself as one of the truly elite relievers in the game. He’s done it all for the team this year, and when there were some questions in the eighth inning early in the season Kimbrel even came in and locked down that frame. He hasn’t been needed in that role as much after the first month or two of the year, but for the most part he’s been the same old Kimbrel. That is, until recently.

Photo by Rich Gagnon/Getty Images

Over the last six weeks or so, Kimbrel hasn’t quite been the same dominant force. Don’t get me wrong: He’s still been great. He’s still one of the best relievers in the game and I don’t feel any less confident when he’s in. That being said, things haven’t been quite as smooth lately. Since the last week of June, he’s pitched to a 2.81 ERA (which, obviously, is still pretty damn good) while allowing a .267 batting average and a .756 OPS. None of that is bad at all, but it looks worse when you compare it to his 0.85 ERA, .086 AVG and .278 OPS from before that point.

Surely, some — probably even most — of the shakiness we feel around the closer of late is because he’s set himself up with impossibly high expectations with his early-season performance. Relievers can sustain such absurd performances over a whole season, but it’s always unlikely. Putting the impossible expectations aside, there are a couple of changes that stand out over this recent stretch of baseball for Kimbrel.

The first thing you notice is that, by peripheral numbers, he really hasn’t been all that different. He’s still striking out an absurd number of batters (he struck out half the batters he faced in July) while keeping his walk rate below seven percent. The issue has been on batted balls, as the once unhittable Kimbrel is now mildly hittable. Looking at his batted ball splits, the first thing that jumped out on me is how much more often batters are pulling the ball off him since the start of July. Over the first three months of the year, opponents were hitting the ball to their pull side under a quarter of the time they put the ball in play, but since the start of July they’ve been doing so about half the time. This suggests they are a little more comfortable at the plate and are figuring out how to get out in front of the heat and stay back on the curveball. Along with pulling the ball, opponents are also hitting more line drives and generating more hard contact, per Fangraphs.

There are almost certainly a lot of factors that play into this adjustment made by the league against Kimbrel, but one thing in particular stood out to me. Starting, again, in the last week of June, we’ve seen the closer start at bats off with ball one more often than he was earlier in the year. This hasn’t resulted in the walk issues we saw in 2016, but it has created hitter’s counts much more often. Obviously, this is a good recipe for hard contact. Below you’ll see a graph that looks at each five-game grouping of his from the season — so, games 1-5, then games 2-6, and so on — and it will compare his first-strike percentage with his opponents’ batting average on balls in play and opponents’ line drive rate. The correlation is pretty clear.

Hopping over to Brooks Baseball, only one of Kimbrel’s two pitches has been hit harder in this stretch, and that’s the fastball. This is a result that makes sense when paired with throwing fewer first-pitch strikes. The closer has been pitching from behind more often, and pitchers throw more fastballs in that situation. Hitters also know what’s coming, which helps explain why they’re suddenly able to pull the ball more often. Opponents are hitting way more line drives off his fastball of late and have as many home runs as ground balls against the pitch.

There’s no reason to be worried about Kimbrel long-term. He’s an incredible pitcher and has been all year long. The issue is that he’s no longer pitching like the best reliever in baseball as he was for much of the season. That’s okay. There are some things to work on — most important is simply getting ahead in the count. When Kimbrel is ahead of you, things likely aren’t going to go well. That will come, and there’s no reason to believe he won’t be a huge part of this team heading into the postseason.