Craig Kimbrel’s four-seam fastball has not been nearly as good in 2018
Featured image courtesy of Zimbio.com: (June 13, 2018 – Source: Lindsey Wasson/Getty Images North America)
The Boston Red Sox pulled off a victory in a close 2-1 game against the Seattle Mariners Thursday night. It was a statement win for Boston, highlighted by an excellent start from David Price and a solo-shot courtesy of Xander Bogaerts which was the go-ahead homer in the 6th. The game was not without drama, however. In the bottom 9th, Boston summoned closer Craig Kimbrel to finish Seattle off and it was not smooth.
He started off by walking the first hitter Mitch Haniger, missing with his fastballs by a considerable margin. He followed that at-bat up with another free pass to Nelson Cruz, despite the slugger’s attempt to help him by offering at curveballs in the dirt. Seattle had first and second with no outs in the inning for a team that has 21 one-run victories and a penchant for clutch hitting. Kimbrel was able to get out of the jam unscathed, striking out Seager on fastballs up and in and then eliciting a groundball double play off the bat of Ryon Healy.
Even though he was able to notch the save, it did not come easy for the impending free agent and that has more or less been a theme for him throughout 2018. In fact, looking at Kimbrel in every important statistical category shows negative regression from his previous season. Check out this table comparing his 2017 and 2018 season.
His strikeouts are noticeably down while his walks and homers are trending in the other direction. The increase in BB/9 is not too surprising, considering it falls closer to Kimbrel’s career norm (3.33 career BB/9) but the other two breed concern. Since debuting in 2010, he has never featured a strikeout rate this low (which is still well above-average for a reliever, by the way) and his homerun rate is astronomical by his standards and demonstrably higher than the average for an MLB reliever.
The most long balls the closer has every allowed in a season is six, coming in 2017 and 2015. Not even halfway through this season, he has already given up five round trippers. Among 174 qualified big-league relievers, he possesses the 30th-highest HR/9, which, while not being unbelievably bad, is still well in the upper (or lower?) percentile. The highest HR/9 he has ever had in a season was 0.91, so it has never been above 1 in any year before 2018. Granted, homeruns tend to be a volatile statistic, with not much predictive value. It appears this enormous spike in HR/9 could be the result of some bad luck/random variance and perhaps we will see some regression to the mean to his career norm as the rest of the season unfolds.
According to Brooks Baseball, though, all five homeruns this season have come off his four-seam fastball, not his knuckle-curve. This is concerning because his 4-seamer has lost a bit of its bite and thereby effectiveness this year. Due to the fact Kimbrel throws as hard as he does, people may not have noticed that the average velocity on his fastball is actually down a bit. In 2017, he averaged 98.7 mph on the heater and, in 2018, it has fallen to 97.4 mph. Yes, he still throws the baseball incredibly hard, but every mph of velocity matters in terms of messing with the reaction time of hitters. The faster the baseball is thrown, the better chance the pitcher has of recording outs and strikeouts with it. Losing 1.3 mph on the pitch is not nothing and has probably made it easier for hitters to barrel the baseball against pitch, ergo the five dingers off it this season.
Last year, hitters only managed to muster a .134 average and .246 slugging percentage off the 844 four-seam fastballs he threw. Meanwhile, this year, hitters have compiled a .185 average and .457 slugging percentage off the 324 fastballs he has offered. Putting it another way, hitters have produced 11 extra-base hits off the four-seamer in less than half of a season in ’18, while they accumulated 11 extra-base hits in all of ’17, a much bigger sample. Further, the Whiff% on the offering is down from 21.21% last season to 17.90% this season, illuminating the fact the loss in velocity has made it easier to make contact with. It also helps partially explain why his strikeout rate is down.
Anyway, over at Fangraphs, they have a linear weight statistic that attempts to measure how many runs a pitch is worth based on its effectiveness. It is an imperfect stat because pitch effectiveness is often influenced by the other pitches, but when there is a monumental change year-to-year, meaning can be extracted. This is the case with Craig Kimbrel, whose 4-seam usage is only down slightly this year. In 2017, his fastball was worth a sterling 19.0 score, which was the fourth best for a qualified reliever. This year, however, it has been worth just 0.9 of a run. Yeah, that is a major difference.
Outside of the velocity, though, what has been the culprit of the diminishing value of his fastball? Is solely attributed to a one and a half mph of velocity loss? Well, no, probably not, although it explains a good deal what is going on. He is really not locating it differently to opposing batters from either side. But he is using the four-seamer to right-handed batters very differently than he did in 2017.
In 2017, right-handed hitters saw a fastball 74% of the time, while, in 2018, they have seen it only 66% of the time. On the first pitch, righties could expect the fastball a majority of the time (64% in 2017) but this season he has gone to the curve more than the fastball to begin a plate appearance, throwing a fastball on the first pitch just 48% of the time. In other words, against right-handed hitters he is throwing a first-pitch fastball 16% less this season opposed to last season.
That is a demonstrable difference, which probably is the reason his overall F-Strike% (first-pitch strike percentage) has decreased from 63.8% in ’17 to 55.7% in ’18. Kimbrel is getting behind in more counts because he is relying on the curveball to evoke a swing-and-miss to begin a plate appearance and shying away from the fastball, which obviously ends up in the strike zone more frequently than a breaking ball. When he gets behind in the count to begin an at-bat, he is evidently starting with a disadvantage.
Even with two strikes, Kimbrel has been much less likely to go with the four-seamer versus same-sided hitters. The right-handed pitcher uses it 76% of the time with two strikes against them this year, a discerning difference from the 85% he used it in the same situation last year. It should be clear that the 30-year-old has changed his approach against right-handed hitter, for better of for worse. His numbers are down against both sides of the plate, even though he still has dominated same-sided competition.
He seems to have a tad less horizontal movement on the heater this year, so it could appear less deceptive to batters. Regardless, Kimbrel has loss a significant beat with his four-seam fastball and, for a two-pitch pitcher, that is a little concerning. Relative to other pitchers, though, he still owns one of the best heaters in the game, even with less horizontal movement and velocity. Additionally, his curveball has been just as good, if not better, this season and he goes to that quite frequently.
Even though he has not been as effective for Boston, he still has been one of the game’s most incredible relievers. It will certainly be interesting to see how his fastball usage against right-handed hitters progresses over the remainder of the season and if he can regain some of the velocity he has lost on the pitch. In a contract year, it would be ideal for Craig Kimbrel to feature the dominant fastball we have seen over his Red Sox career. Also, he has an unsustainable .196 BABIP (batting average on balls in play) this year, a far-cry from his career .262 BABIP. Just throwing that in there.