Coming into 2018, Royals RHP Jakob Junis was one of my favorite sleepers in the AL. In August and September of last year, Junis had a 3.61 ERA and 50 strikeouts in 62.1 IP. By FIP, he was even better, with a 3.43 mark in that span, thanks to a measly 1.3 BB/9 and a HR/9 under 1. His slider was (and still is) nasty:
There were plenty of other positive indicators as well, and through the first 2 months or so, Junis was cruising. After 12 starts, he had a 3.61 ERA and 71 K’s in 74.2 IP. He had a bit of a HR problem, as his 12 allowed dingers was tied for 14th highest in baseball (out of 170 qualified SP), but he was limiting the damage everywhere else, and his .299 wOBA against was in the top half of pitchers in the league. He looked good, but that slight HR problem all of a sudden became a hell of a lot worse.
From June 8th to July 2nd (5 starts), Junis allowed a whopping 12 home runs in just 27 innings, ballooning his season ERA from 3.62 to 5.13. His HR total was now at 24, which not only was the highest in baseball, but was 20% higher than the second largest total in the league. Junis would hit the 10-day DL 5 days later with lower back inflammation.
If you know me or have read any pieces in the past, you know that, primarily, I choose to write about prospects within baseball. I dabble in the occasional MLB player analysis piece such as this every once in a while, because I love this stuff, but mostly, it’s prospect research/watching/writing. One of the first things you learn about a pitching prospect when attempting to project future command is the pitcher’s delivery and, more specifically, the repeatability of said delivery and the release point of the ball as it comes out of the pitcher’s hand. If a pitcher has an inconsistent delivery or the point in which the ball leaves the hand is inconsistent, chances are the pitch location will be equally as inconsistent.
Though the first 3/4 of the season, that is exactly what Jake Junis struggled with. His release point was not as tight as you would like it to be. As a result, the command came and went. Here’s his release point chart from his first 20 starts of 2018:
However, over his last 6 starts, Junis has found success, with a 1.89 ERA and 39 K in 38 IP, all while only allowing 1 HR. His release point chart over those 6 starts looks much better:
Comparing the two, the release point looks far more condensed. It’s a bit crowded, but by magnifying the area in question, it further supports the notion of more consistent release point (with the exception of a wonky trio of curveballs in chart #2):
Of course, as the sample size for the second data set grows bigger and bigger, it’s mathematically guaranteed that the chances a new extreme release point appears and furthers the outside limits of what we’re looking at, but it hasn’t happened quite yet. Take a look at the violin plots from the two stretches in question. A violin plot reads sort of like a box plot, so we can see the extremes on both sides of a distribution, but gives a broader view of the information at hand by showing the full distribution [instead of a box plot which just gives summary statistics such as mean/median and interquartile ranges].
The bodies of the plots and their medians don’t change all that much, but the difference of the range in which the plots extend is striking. In Junis’ first 20 starts (chart #1), the maximum and minimum release points were quite far away from the median. In his last 6 starts (chart #2) the density (consistency) around that median is much, much better and the max/min release points are closer. The same can be said for his charts for his horizontal release points, with exception, again, given to those damn few curves:
When a pitcher’s command comes and go, so do the quality of his starts, for one reason or another. In Junis’ case, his lack of command meant he was leaving way to many hittable pitches out over the plate. Here’s the zone breakout of his first 20 starts:
The pink zones, despite not being nearly as dark as the deep red, suggest very high consistency. Usually, unless it’s coming in at triple digits with movement or deception, it doesn’t matter what the pitch looks like; right down the middle is a hittable pitch. Junis doesn’t throw anywhere near triple digits.
In his last 6 starts with a more consistent release point, however, those pink zones (down the middle and the right outside edge) have cooled off
Some of the important middle zones, such as middle up, middle out, and middle in, have grown a deeper shade of blue as well. The decreased frequency of hittable pitches inside the zone have been huge for Junis.
His Zone% in his last 6 starts compared to the first 20 has dropped from 46.6% to 42.4%.
His Contact% has dropped from 80% to 76.2%.
His Z-Contact% has dropped from 90.9% to 88.1%.
His O-Contact% has dropped from 61% to 58.5%.
His SwStr% has risen from 9.2% to 11.1%.
All-in-all, Junis has just become… less hittable. A 4% drop in contact rate may not seem like a lot, but in your average 100-pitch start, that could be a difference of 1-3 different instances of contact. These could be 1-3 foul balls, sure, but they could also be 1-3 extra base hits.
Jakob Junis went on the DL on July 8th after a horrendous stretch of starts with a lower back issue. It is not my job to theorize whether or not Junis’ command problems were because he was, in fact, not fully healthy for that stretch, but it sure looks that way. In the same sense, I would be remiss if I didn’t acknowledge that the stretch of starts being analyzed here is an extremely small sample size of just 38 IP. However, Junis’ release point chart, at least over a stretch of half a dozen games, is much more consistent than any other 6-7 game stretch during his career, including the first 7 games of the season in which Junis pitched very well. That small sample size could make this all look very premature down the stretch should Junis lose his release point again, but for now, it looks like we could be looking at a brand new, completely healthy, very good Jakob Junis.