Randy Dobnak’s adjustments have backfired

Randy Dobnak’s Twins career has been filled with unexpected twists and turns. You’re familiar with his story and his results so far. After an amazing (and improbable) beginning in 2019 that carried over to the first half of 2020, Dobnak’s trajectory has been pointed down. The second half of 2020 saw him struggle for the first time at the highest level and resulted in him losing his rotation spot and being optioned to the minors.

Following a surprise, team friendly early career contract extension before this season, Dobnak made the 2021 Opening Day roster in the bullpen. Disaster ensued and he was optioned again to build up for a return to the starting rotation. Upon returning, Dobnak has continued to struggle while trying to work through a couple of finger injuries that now have him on the injured list. Since the middle of last season Dobnak’s results have generated far more questions than answers and the depth of his most recent struggles make the slope of his career outlook point more sharply downward than it has at any point thus far.

Data sourced from baseballsavant.mlb.com

I’ve written in depth about Dobnak twice. The first time was in the middle of his great start to 2020 and I tried to determine if he had staying power. The cliff notes version of that piece was that he had some skills that could enable him to be successful long term — namely, the ability to generate lots of ground balls and strong control — but his upside would be limited by his lack of bat missing stuff and dependent on the quality of the Twins infield defense.

The second piece was during this past offseason and I tried to make sense of what happened with him in the second half of 2020. The takeaways from that were that Dobnak’s control wavered as last season progressed. His approach and stuff generally stayed the same, but inconsistent control led to fewer pitches in the strike zone, falling behind in the count more frequently, and higher rates of free bases allowed via walks and hit by pitches. Collectively, those factors helped to further reduce his already very low strikeout rate and may also have contributed to increased difficulty handling left handed batters.

Dobnak’s worm-burning, contact heavy profile is contrary to the way many pitchers are finding success in the strikeout oriented game today. His recipe for success is a throwback built around working ahead in the count, getting ground balls, and avoiding home runs and free bases. With two of the three core tenets of fielding independent pitching well covered, it’s a profile that can work, but it is also one without much margin for error.

Given that, it is understandable that the Twins pitching department and Dobnak have been seeking adjustments that could unlock some more missed bats. Last season, the main change toward that end was dropping Dobnak’s release point about 3 inches from 2019:

Instead of creating more swings and misses, though, that change actually decreased Dobnak’s whiff rate on each of his four pitch offerings. This was especially true on his slider, which went from a 46.3% whiff rate in 2019 to 33.8% last season. Those lost swings and misses were offset a bit by an increased ground ball rate (62.1%, highest among all pitchers to throw at least 40 innings). Again, this was especially true for his slider which had a 57.8% ground ball rate last season, up from 30.8% in 2019.

In spring training, it was not surprising when much was made about the Twins working with Dobnak to adjust his slider to be more of a bat missing weapon. Facetiously or not, Dobnak was quoted after a good spring training outing against Tampa Bay (6K’s in 3IP, 5 with sliders) saying, “I’m actually a strikeout guy. New Year. New me.”

Well, the results are in. And… no he’s (still) not.

Still operating from the lower arm slot, Dobnak has again generated a high ground ball rate (55.3%, 39th of 350 pitchers with at least 100 batters faced) and his walks (6.0%) and hit by pitches (just 1) seem back in line after last season’s second half troubles.

But his 12.5% strikeout percentage is the lowest mark of his career and is in the 1st percentile (i.e., worst) of all pitchers. Not only has the hyped up slider not resulted in more strikeouts as intended — it has been far and away his worst performing pitch.

By FanGraphs’ pitch type linear weights, which use the changes in run expectancy from one count to the next to measure how well a player has performed with a certain pitch, Dobnak’s slider has been more than 11 runs below average. By Statcast’s similar run value method, it has been 13 runs below average. It has been the worst slider in baseball per run values at both FanGraphs and Statcast.

Translating those metrics into more traditionally familiar terms, opposing batters have hit .333, slugged .804, and launched 7 home runs against his slider. Those results are starkly different from last season when his slider was his best pitch by the run value measures (5 runs above average by FanGraphs, 6 runs above average by Statcast). Then, batters hit .219 and slugged .234 with no homers and just a lone double against it.

Instead of being a new swing and miss weapon, the new slider has mostly been prone to extra base damage. While its whiff rate is up just a bit to 37.3%, it’s also become much easier for batters to elevate. The average launch angle for batted balls off Dobnak’s slider has increased to 16 degrees (from 7 degrees last season) and the slider’s ground ball rate has dropped to 34%.

You might be wondering how much of a factor luck is playing in Dobnak suddenly becoming homer prone. After all, he’s given up 11 total homers already this season, after only allowing 4 the prior two seasons combined. The data suggests that it’s a bit of a mixed bag in this area.

On one hand, a flukishly high 29.7% of all the fly balls Dobnak has allowed this year have gone over the fence. That mark is easily the worst in the majors among the 154 pitchers who have thrown 40 innings. It is nearly four times worse than his 7.8% combined mark in 2019 and 2020 and more than double the 2021 league average of 13.6%. So there might be some bad luck at play. On the other hand, Statcast’s estimates of Dobnak’s expected home runs based on the batted ball distances and ballpark dimensions is 11.3, suggesting that he’s earned the homers he’s allowed.

So, what’s not working?

By Stacast’s measurements, his pitches are behaving mostly the same as they always have. The velocity readings on all his offerings are in line with the past. He’s getting similar movement on his slider. If anything, he’s getting more movement on his sinker. His changeup looks the same. His four-seamer has a lot more movement than the past, but he doesn’t throw it that much.

What’s mostly changed is the consistency of his command. Yes, his walk rate and hit by pitches are consistent with his past, but those stats are rough measures of control. A generally accepted distinction between control and command is that being able to throw pitches over the plate is about control. Being able to throw pitches where you want to throw them in and around the strike zone equates to command.

Dobnak has had good control. He has not had good command.

Command is a difficult thing to measure because it requires us to interpret intent. We don’t always know what a pitcher was trying to do with a certain pitch, so it’s not a place where we have detailed measurements to analyze.

But, we do have lots of pitch location data parsed out by attack zones. If we generally assume that pitchers are trying to avoid throwing pitches over the heart of the plate and non-competitive waste pitches far away from the strike zone, it gives us a crude starting place for assessing command.

In Dobnak’s case, both of these have trended in the wrong direction in 2021. This season, 26% of his offerings have been located in the heart, up significantly from last season’s 21%. At the same time, 12% of his pitches have been classified as waste pitches, up from last season’s 8%.

Together, those factors serve to make it easier for the hitters and are indications that Dobnak is really struggling to command his pitches around the edges of the strike zone where it is much harder for batters to discern balls from strikes. Naturally, if heart and waste pitches are up, then the shadow zone and chase pitches that are toughest on hitters are down.

This is clearly demonstrated with Dobnak’s slider. Here is the location heat map of all 264 of Dobnak’s 2020 sliders:

And here are all 231 sliders so far in 2021:

Last season, his sliders were tightly clustered in the shadow zone low and away on or just off the edge of the strike zone. Those are very difficult locations for opposing hitters.

This year, his sliders have been sprayed all over that side of the plate with some of the most prominent red areas in the heat map being in the middle of the zone or all the way out in the left handed batter’s box. Far too many of these offerings have been easy takes or opportunities for hitters to feast.

This is especially problematic because Dobnak most frequently relies on his slider as his out pitch when he’s ahead in the count. This season, about 55% of his offerings when he has had the count advantage (and more than 65% in 0-2 and 1-2 counts) have been sliders.

Working ahead is usually an advantage for the pitcher. Collectively, the league has batted .190 / .200 / .300 with a .215 wOBA when the pitcher is ahead in the count this season. Contrary to that pitcher dominance, Dobnak has allowed .345 / .356 / .828 / .494 and 8 of his 11 home runs with the count advantage. The shaky slider command is no doubt a big reason why. The fact that 6 of the 7 homers he’s allowed on sliders have come when he was ahead in the count supports that notion.

It’s fair to wonder how much the release point adjustment last season have contributed to the bouts of inconsistent command he’s experienced over the past two seasons. It’s also very probable that the series of finger maladies he’s suffered this season factor in significantly to his current struggles. The lack of a clear role, place on the 26-man roster, and regular workload could also be a factor.

In any event, one thing is clear. The adjustments Dobnak and the Twins have tried in their attempt to get him to the next level have backfired and led to regression. Now they must hope a return to health and more time and repetitions to get comfortable with the changes will let Dobnak get back to executing his pitches as intended. Otherwise, it might be back to the drawing board to find more tweaks once again.

---

John is a contributor to Twinkie Town with an emphasis on analytics. He is a lifelong Twins fan and former college pitcher. You can follow him on Twitter @JohnFoley_21