As most everyone who likes baseball enough to read this blog knows, Jake Arrieta was a horse this year. Finishing the season with an eye-popping 22-6 record, 1.77 ERA, and an obscene 0.865 WHIP, Arrieta would be all but a lock for the NL Cy Young if it weren’t for a certain pair of Dodgers’ hurlers. In order to finish a season with numbers so incredible, one would assume that Arrieta steadily dominated hitters from his first start to his last with few exceptions. Fortunately for statistics nerd like me, that wasn’t the case for Arrieta this year. Why is that fortunate, you ask? Well, if Arrieta wasn’t dominant all season and still finished with such extraordinary numbers, that must mean there was at least a stretch that he was even BETTER than his final numbers showed. This stretch is what we as analysts are interested in, and finding significance in such bouts of success is easiest when we have something to compare it to. Fortunately (again) for us, Jake decided to go ahead and split his season into two parts for us, which from here on out will be referred to as Part 1 and Part 2. At the end of Part 1 (June 16) Arrieta’s ERA sat at a serviceable but unremarkable 3.40 (the highest it was all season, except for on May 7 when it sat at 3.41). Over his next 20 starts, Arrieta’s ERA began a descent of unrivaled proportions, beginning with a 4-hit complete-game shutout on June 21. Arrieta’s ERA over his last 20 starts was an outrageous 0.86, which brought his season total down to the aforementioned 1.77.
The interesting part about the change in Arrieta’s effectiveness is that he didn’t really strike out too many more batters – 26.89% in Part 1 against 27.27% in Part 2. Arrieta also didn’t really walk many more batters in Part 2 either – 6.34% vs. 5.01%, and the Part 1 percentage is inflated by a 6-walk performance in (coincidentally?) his June 16 start. So, those are just small improvements on already pretty good rates. This must mean that most of Arrieta’s changes came in the form of what happened on balls put in play. Let’s look at some charts and graphs, and from those we’ll examine just exactly what Arrieta did to be so much more dominant. First, we’ll take a look at the results-based portion of these charts, which would be the hit location heatmaps.
As you can see , Arrieta gave up FAR fewer deep balls to the outfield in the second part of his season, especially to right field. Arrieta’s ground balls are also farther away from the middle of the field and not quite as deep – which means fewer ground ball hits up the middle and easier plays from his middle infielders. Interestingly, when examining the direction of the grounders Arrieta induced in Part 2, they moved in the opposite direction as his outfield hits – Arrieta got significantly more grounders to the right half of the infield in the latter part of his season. All of this is nice to know, but ideally we’d like to know WHY Arrieta’s batted balls looked so differently in the second part of the season. This leads us to an examination of Arrieta’s pitch heatmaps and pitch breakdown charts.
Part 1: Pitch Breakdown (from Brooks Baseball)
Part 1: Pitch Heatmap
Part 2: Pitch Breakdown (from Brooks Baseball):
Part 2: Pitch Heatmap
In Part 1, Arrieta was far more erratic with his pitches, missing left-to-right more often than he probably liked. He also threw nearly twice as many four-seam fastballs as he did in Part 2, while throwing fewer sinkers (39% to 45%) and changeups (3.5% to 5%). His slider and curveball were used nearly the same. This (pretty drastic) change in pitch usage resulted in some good things for Arrieta – a lot more pitches inside the strike zone, and more pitches low and away (to a righty). Arrieta’s most frequent pitch location (yellower parts of heatmap) also shifted lower and away from a righty in Part 2. Interestingly enough, all of these changes point to exactly the differences we saw in looking at the hit heatmaps – fewer balls driven to the outfield and more balls hit on the ground weakly, especially to the right side. Additonally, Arrieta had far more success with his slider in the second part of the season – more whiffs, more balls put in play (seems contradictory, but this just means he threw it for more strikes and it was fouled off less often), FAR more grounders and fewer line drives. Almost all of Arrieta’s velocities say a slight uptick in Part 2 as well, which helps create weaker contact (usually), but that’s pretty typical for most young pitchers as their arm builds strength and endurance over the course of a 162-game (or, in this case, 33-start) season.
Arrieta’s huge increase in success in the second half was not an accident and it was not a fluke – it’s clear that he and the Chicago coaching staff made a conscious decision to try and pitch to contact more, using his best pitches more and his weakest pitch less. Arrieta clearly also focused on improving his slider command, which resulted in far more favorable results on that pitch than he had in the past. Arrieta’s ability to adjust so quickly and turn his season from good to amazing is something that speaks to his competitiveness and dedication to the art of pitching, which makes him so fun to watch on the mound. Thankfully, we won’t have to wait long to watch him again – he’s scheduled to take the mound tonight at Wrigley at 6PM ET. Judging by his stellar performance in the NL Wild Card game, Arrieta isn’t phased by the big stage, and I expect him to spin another gem tonight in front of what will surely be a raucous home crowd. Thanks for reading guys, enjoy watching Jake work tonight, and enjoy the games!