Greg Holland’s fastball won’t cut it anymore

I’m someone who loves to break down players when they are in the news, and Greg Holland is no exception to this rule. On Wednesday, the Diamondbacks agreed to a one-year, $3.25 million deal with Holland, if only to recapture enough magic to flip him at the 2019 trade deadline.

It’s not a bad idea. I’ll remind you of this: for a short time, Holland was one of the best relief pitchers in baseball. During his three-year peak from 2012 to 2014, Holland was worth a total of 7.3 fWAR. That ranked third among all relief pitchers, behind just Craig Kimbrel (7.9 fWAR) and Aroldis Chapman (7.9). His 1.88 ERA ranked fourth-lowest.

Holland’s had a hard fall in the years since, marked most notably by Tommy John surgery which forced him to miss all of 2016. Even when healthy, Holland has not been the same. Holland has posted a 13.0 percent walk rate since leaving the Royals, compared to just a 9.6 percent rate while in Kansas City.

He did lead the National League in saves with the Rockies in 2017, and signed a one-year, $14 million deal with the Cardinals last offseason. He was released in early-August after walking as many batters (22) as he struck out in 25 innings. The Nationals picked him up shortly thereafter, where he posted solid numbers — a 0.84 ERA alongside a 25 to 10 strikeout-to-walk ratio in 21 ¹⁄₃ innings pitched.

So, Holland has had both ups and downs since leaving Kansas City. Needless to say, he’s been significantly worse overall, especially in the control department.

I came here today to try to pinpoint the real issues behind Holland’s recent struggles, in hopes that I can suggest a solution. I’m no pitching expert by any means, but data can tell us a lot. Perhaps a deeper dive in the data is all we need.

With all of the injury issues that Holland has had, his fastball velocity has taken a tumble. In 2013, his fastball was averaging 96.8 mph. By 2018, it fell to 92.9 mph. As a result, Holland couldn’t blow his fastball by any hitters. For an easy comparison, in 2013, batters hit just .260 with a .321 wOBA against Holland’s fastball. In 2018, those numbers jumped to a .297 batting average and a .411 wOBA. For reference, batters were hitting better than Alex Bregman (.286 BA, .396 wOBA) when Holland threw the fastball.

This velocity loss hasn’t only resulted in more batters being able to barrel up Holland’s fastball; it has also caused him to generate fewer strikeouts. Batters swung and missed at 27.3 percent of Holland’s fastballs in 2013, resulting in a 23.8 strikeout percentage on the pitch. With the lower velocity this past season, batters swung and missed just 16.8 percent of the time and struck out just 8.0 percent of the time against the fastball.

Because Holland has seen the effectiveness of his fastball significantly decrease over time, he’s throwing fewer fastballs for strikes. This has ultimately resulted in more walked batters on the pitch. Here’s a comparison of Holland’s 2013 fastball...

...to his 2018 fastball:

As you can see in the images above, Holland threw a significantly fewer percentage of his fastballs in the strike zone in 2018. Quantifying that drop, Holland threw 47.6 percent of his fastballs in the strike zone in 2013 compared to just 40.2 percent in 2018. Throwing fewer pitches in the zone isn’t a problem when you’re able to blow the pitch by a hitter. But with his decreased velocity, Holland hasn’t been able to do that. Thus, his walk rate on the fastball has spiked from 13.1 percent to 24.1 percent. That’s not good, to say the least.

Here’s a chart that compares Holland’s average fastball velocity to his fastball strike zone percentage over time:

Holland didn’t pitch in 2016, and he may have had struggles controlling his pitches in 2015 due to his torn UCL. So, even if you ignore 2015, there’s a clear difference in the number of fastballs he is throwing for strikes as his velocity has dropped over time. This would make it easier for hitters to detect and lay off the pitch.

Because of these poor results, Holland’s fastball was worth -2.0 weighted runs above-average in 2018. This ranked 144th out of 191 relief pitchers with more than 40 innings pitched, putting him squarely in the 25th percentile in this metric. At 33, it’s unlikely that Holland will be able to recover much of the velocity that he has lost on his fastball. It seems that it is time for him to find a new weapon.

As you can see above, Holland’s slider has consistently been his best offering. Since returning from injury, he’s thrown it at a much higher rate than he did in the past, likely to compensate for an ineffective four-seamer:

What’s most interesting to me, however, is that Holland’s slider percentage saw a steep drop after his move from the Cardinals to the Nationals last season:

Holland’s April, May, June and July were spent solely with St. Louis, while his August and September were spent solely with Washington. With the Cardinals, 48.2 percent of his pitches were sliders. With the Nationals, this number dropped to 41.3 percent.

Perhaps due to this drop (perhaps not), Holland’s results on the pitch became significantly better. Just consider his wOBA by month on the pitch:

• April 2018 (Cardinals): .353 wOBA against in 16 plate appearances
• May 2018 (Cardinals): .414 wOBA against in 19 plate appearances
• June 2018 (Cardinals): .000 wOBA against in 9 plate appearances
• July 2018 (Cardinals): .283 wOBA against in 18 plate appearances
• August 2018 (Nationals): .079 wOBA against in 20 plate appearances
• September 2018 (Nationals): .157 wOBA against in 20 plate appearances

With the exception of Holland’s June — which is an abnormally small sample — his two best months of slider performance came with the Nationals. His April, May and July were three of the eight worst months of Holland’s career in slider results. Batters this past May posted the highest wOBA against his slider of any month during his career.

Thus, throwing the slider more might not be the answer to Holland’s struggles, though even that can be disputed. In 2017, when Holland was better than he was last year, he threw his slider 49.4 percent of the time, a career-high. The results on his pitch that year were phenomenal — a .137 batting average against with a .174 wOBA against. It was dominant.

If the results about the his slider percentage are inconclusive, what about Holland’s curveball? There isn’t a whole lot of data on the pitch, considering he has thrown it just 254 times over his eight year career. In 2018, he threw it 10.8 percent of the time, a career-high. Yet, it posted mixed results. Batters hit .263 against it (5-for-19), which isn’t great. They swung and missed 29.0 percent of the time, which is about three percentage points lower than the league-average curveball whiff rate (31.9 percent). But, his xwOBA against the pitch, which takes into account launch angle and exit velocity, was just .217. That was significantly below the league-average .254 mark against the curve.

What does all of that mean? Holland’s curve had a mixed bag of success last year, but it might be worth his time to tinker with it. After all, with the Diamondbacks, he is mainly looking to rebuild his value. With a larger focus on his curveball — it doesn’t have to be his No. 1 offering by any means — he might be able to do just that.

Regardless of whether Holland wants to focus on the slider, the curve or a combination of both, one thing is for certain: Holland needs to throw his fastball less. It has lost its effectiveness. As a result, Holland has not been able to control it like he did during his peak.

---

Devan Fink is a Featured Writer for Beyond The Box Score. You can follow him on Twitter @DevanFink.