Scherzer’s Justified Confidence

At the end of May, when his seasonal numbers were still pretty ugly, Max Scherzer was interviewed by our own David Laurila. At the time, people were puzzled by Scherzer’s high strikeout rate (12 per nine innings) accompanied by a very high ERA (5.67). In the interview, Scherzer acknowledges the problems he has had, and does not use BABIP “luck” as an excuse. But he also says that “My stuff, right now, is where I want it to be.”

That may have seemed a bit cocky coming from a pitcher with a 5.67 ERA. But after April’s 7.77 ERA mess, Scherzer’s performance as measured by ERA, at least, has improved each month: 4.04 in May, 3.86 in June, 3.62 in July, and 2.25 so far in August. All the problems of ERA are exacerbated by dividing an already spartial-season sample into months, but it does catch the eye. Did Scherzer change something that spurred the improvement, or is this just the vicissitudes of in-season variation striking again?

Scherzer’s stuff has always been impressive, as have his strikeout rates. But when he has struggled to have an ERA that matches up with his peripherals, he has been compared to, say, Brandon Morrow. Morrow, curiously enough, managed to bring his ERA down to his peripherals so far this year. For the first time since 2008, his ERA (3.06) is lower than his FIP (3.39). Morrow seems to have done so while lowering his strikeout rate rather dramatically in 2012 to about 22 percent per nine innings after being over 25 percent in the previous two seasons. Morrow’s walk rate has also improved significantly this season.

Whether this is sustainable for Morrow is something for another time, but it does raise an interesting contrast to Scherzer. While Morrow has improved his runs allowed from the previous two seasons by lowering his strikeout rate, Scherzer does not seem to have “given in.” In fact, while Scherzer’s April strikeout rate was already excellent (22 percent of plate appearances), every month since then has been higher, if not following the linear path of improvement of his ERA.

This is not to dismiss the importance of his improved walk rates since April — that is at least equally significant. The point is simply the contrast with Morrow going away from strikeouts as a path to improvement. Scherzer was confident in his stuff at the beginning of the year despite the struggles with walks and, of course, a really high BABIP. Has he changed anything?

Without doing an in-depth analysis of pitch trajectory, nothing stands out. Compared to previous seasons, Scherzer is throwing basically the same mix of pitches. (Scherzer may or may not have experimented with a sinker this year, depending on how one interprets the pitch classifications, but it was just a few pitches, and he seems to have abandoned that experiment). One difference from previous seasons is that he is throwing his four-seam fastball about 94 rather than 93, which is, of course, usually a good thing.

However, the contrast we are looking for is not so much between 2012 and 2011, but between April 2012 and the rest of the season. Again, if one looks simply at pitch usage from month-to-month in 2012, one basically sees the same mix of pitches: each month, Scherzer has thrown about 60 percent fastballs (including the occasional sinker earlier this year), and about 15 to 20 percent change ups and sliders.

One should be cautious before reading too much into any month-to-month variation of anything in baseball. It is also worth remembering that the variation in the number of change ups and sliders from month-to-month might at least in part be explained by Scherzer facing different numbers of left-handed hitters in some months than others, and the change up comes out more often again them. On change within the season perhaps worth noting is that Scherzer has thrown far fewer changes to right-handed hitters in July and August. However, but the improvement in his ERA and FIP came before that.

Overall, there seems to be little reason from the evidence examined to assume that Scherzer made some big change in his approach after his terrible April. The linear decrease of his ERA by month catches the eye, but things like that can lead us into formulating misleading narratives. Scherzer’s FIP does not quite follow the same path, although it is close: 4.45, 3.83, 3.20, 4.03, and 2.50 by month. His xFIP has not been improving in a linear fashion, either: 4.72, 2.35, 3.12, 3.88, 2.53. This is not to say that his ERA was completely misleading, but that we should not place to much emphasis on a “story” simply for the order of performance, especially when that “order” depends on setting the data into arbitrary monthly sets that are more skewed by random variation than a full season.

Part of the reason Schezer’s problem attracted so much attention was likely that it occurred in the first month of the season. If Scherzer’s April and July performances this year were “flipped,” the bad month would not have received the same sort of attention, as his full-year stat line would still have looked acceptable at the end of July with an ERA in the mid-fours rather than the 5.67 from the end of May. We might have seen stories about “Scherzer has had a rough July after a good start to the year,” but they would not have been as prominent or as panic-striken as the concern about Scherzer’s numbers in April or May.

Scherzer did have a rough start, and he may have made some minor adjustments that have helped him throughout the season. Simply pointing to a pitcher’s FIP as opposed to his ERA is not a solution to every question. In this case, though, it does appear that Scherzer simply had the misfortune of running into the bad end of random variation right at the beginning of the year rather than in the middle. He was not as bad as he looked in April, and he is not as good as he has looked so far in August. At the end of May Scherzer expressed confidence in this stuff, and since then he has shown that that confidence was well-placed. If the Tigers make the playoffs this year, he is going to be a key part of a formidable rotation.