The Secretly Good Curveball Of Ryan Yarbrough

Ryan Yarbrough has evoked hard questions and even stronger opinions among the Rays’ fan base. During the first half of this season, most of the debate was around whether or not he should be starting games or coming in behind an opener. His career splits show it actually doesn’t matter; he does tend to throw a few more strikes when he’s starting, but Yarbrough is essentially the same pitcher no matter what role he’s in.

But that’s not what I’m here to talk about. The current debate about Yarbrough stems from his overall effectiveness this season. We can see below that he hasn’t been great this season, but he really hasn’t been bad either by his standards.

The main things that really stick-out here are Yarbrough’s increase in HR/FB rate, dropping pitch velocity, and relatively high FIP and xFIP numbers. It would be easy to say, “well his pitch velocity is low so it makes sense that he gives up more home runs now, which is causing his FIP and xFIP to increase.” This would also help explain why his second half numbers have been worse than his first half; logically, his pitch velocity would drop as the year progresses if it’s dropping from season-to-season.

Not much has changed from Yarbrough’s first half to his second half other than a slight jump in BABIP which has resulted in a higher AVG and fewer men left on base. It’s also a smaller sample size, but there’s still a noticeable drop in his performance so far.

But like we said, Yarbrough’s average pitch velocity is at a career low this season. And while this is true, we actually can see that his pitch velocity is currently around the same speed it was back when the season started (excluding his start against Baltimore on 8/18; he was coming off the Covid list and didn’t have a rehab start to build his arm-strength back up).

Yarbrough was pretty effective early in the season with a lower pitch velocity, so why should we think a lower velocity is just now causing his performance to drop?

The reality is that it’s not the reason for his performance this year or the drop in performance in the second half. To further support this, we can see that Yarbrough is in the 99th percentile in average exit velocity and 97th percentile in hard hit rate, and he has been incredibly consistent with that the entire season. So he’s not giving up harder contact as a result of his velocity dropping.

A slower average pitch velocity does mean he has to be careful, but being careful and dancing around bats is something that Yarbrough has built his career on.

So if his velocity isn’t the cause for his recent decline, what is? What’s really interesting is that it’s most likely a result of him experimenting with his arsenal.

We can see Yarbrough has been using his curveball more than his sinker for the first time this season. And rightfully so; his sinker has been his worst pitch based on xwOBA for his entire career.

Historically, we’ll notice that his favorite way to attack LHH has been sinker-curve with an occasionally cutter. However, this season Yarbrough has been attacking lefties with more curveballs than sinkers (while still sprinkling in some cutters). We can see that his curveball has been very good his entire career, and Yarbrough is now starting to throw it more to both RHB and LHB.

For some context, here’s how good his curveball is compared to others in the league (minimum 50 plate appearances):

Something else of note is how Yarbrough is attacking righties. In the past, he has gone after them with cutter-change-sinker, but this season he has been more of a cutter-change-curve guy – with even more curveballs than sinkers as the season has progressed.

This is a great idea because, like we established earlier, the sinker has never been good and his curveball is surprisingly really good. So now you might be wondering, “why has he gotten worse this season if he’s throwing his curveball more and it’s supposed to be his best pitch?”

The answer to this is a little complicated, but that’s primarily because he’s trying to figure-out how to best incorporate his curveball to both RHH and LHH in terms of sequencing. First, we’ll look at how he has performed against LHH with his curve.

You’ll notice that he has really gotten a feel for that curve this season, and has continued to improve as the year has gone on. Now, lets look at how that curve has been against RHH.

Yarbrough only threw nine curveballs to RHH in 2020, so we can ignore that outlier. However, this season he has thrown 176 curveballs to RHH – this is over three times more than each of his totals in 2018 and 2019. He has gotten better results with his curve to RHH as the season has gone on which is great, but he’s still trying to figure out how to use it.

His curve has pretty average horizontal movement this season, but roughly 8% less drop than the average curveball this season. Additionally, Yarbrough has relatively low spin on it, ranking in just the 24th percentile. So if the pitch doesn’t have any special movement or spin, how is it so effective?

Because it looks almost exactly the same as his changeup out of the hand, but has inverse spin (and therefore inverse movement as well). Kind of like what we’ve seen with Rich Hill that made him so good in May – his fastball and curve look exactly the same out of the hand until it’s too late.

So we can see his curve and change both have similar amounts of spin, but in opposite directions. The curve spin pretty much goes from 3:30 to 9:30, while the change spin goes from 9:30 to 3:30; so while the spin goes in opposite directions, it’s still very hard for hitters to pick-up. All they see is the side-to-side spin, and it’s almost impossible to see which way it’s coming from. Additionally, the horizontal movement between the two pitches is less than two inches apart, so it looks almost the same on the way to the plate until the bottom falls out on the curve more than the change.

Yarbrough, Kyle Snyder, and both catchers are still trying to figure-out the sequencing aspect of incorporating the curve more which helps explain why his other pitches have been getting hit around a bit. He’s throwing his curve to LHB which is nothing new, but he’s using it a lot more to RHB. This season, the xBA from RHB on the curve is an incredible .187, but the SLG is somehow .486. It has xSLG of only .386 to RHH though, so we can expect regression to set in; it always does.

Yarbrough is currently moving away from his poor sinker in favor of the curve (some people might want him to move away from his cutter too because it’s just average, but he needs a pitch with that type of velocity and movement to keep hitters honest and not just sitting change or curve). It’ll take time before Yarbrough figures it out all the sequencing stuff, but his start against Baltimore on August 18th was a step in the right direction.

The offense being as good as it has been since Cruz has arrived (131 wRC+ from 7/23 to 8/19) gives Yarbrough more freedom to take chances. I’m excited to see him continue this experiment as we progress into October because it could very easily make him the number two starter by then.