Who Needs a New Pitch the Most?

I love it when research underlines conventional wisdom. Like when Mitchel Lichtman found that, the more pitches a pitcher had in his arsenal, the better his chances the third time through the order. Even if it was only on the order of a few points of weighted on base average, it was a real finding that functions as a virtual nod towards all those scouts and pitching coaches who’ve wondered about a pitchers’ third and fourth options. You might not need a changeup specifically, but you need other pitches if just to put more doubt in the hitter’s mind.

Given that finding, I thought it might be fun to try and use it in reverse. Who were the worst pitchers in baseball last season when it came to the third time through the order? Who saw their talent drop off the most upon seeing a batter the third time?

Thanks to Jeff Zimmerman and our splits database, we can produce the following table. It features those pitchers who (a) recorded at least 10 starts last year and (b) experienced the biggest decline in strikeout rate the third time they saw a lineup. I’ve left in the FIP split, too, so we can see their overall results.

Worst Third-Time-Through-the-Order Pitchers, 2016
Name Split K% FIP Split K% FIP FIP Diff K Diff
Juan Nicasio 1st Time 36% 3.19 3rd Time 16% 2.71 0.48 19.8%
Joe Ross 1st Time 28% 1.66 3rd Time 12% 4.72 -3.06 16.3%
Luis Perdomo 1st Time 23% 3.26 3rd Time 8% 6.50 -3.24 15.5%
Dylan Bundy 1st Time 29% 3.61 3rd Time 15% 9.30 -5.69 14.8%
John Lamb 1st Time 24% 5.19 3rd Time 10% 5.65 -0.45 13.7%
Matt Harvey 1st Time 27% 2.47 3rd Time 13% 4.73 -2.26 13.5%
Alex Wood 1st Time 31% 2.46 3rd Time 18% 4.33 -1.87 13.4%
Tim Adleman 1st Time 24% 3.18 3rd Time 11% 6.51 -3.33 13.0%
Adam Conley 1st Time 26% 3.17 3rd Time 13% 6.31 -3.15 12.4%
Jered Weaver 1st Time 20% 4.22 3rd Time 7% 8.45 -4.23 12.3%
Phil Hughes 1st Time 19% 5.08 3rd Time 7% 6.45 -1.37 12.2%
Jacob deGrom 1st Time 31% 2.48 3rd Time 19% 4.36 -1.88 12.0%
Mike Clevinger 1st Time 20% 5.67 3rd Time 9% 10.90 -5.22 11.5%
Ryan Vogelsong 1st Time 22% 5.11 3rd Time 10% 4.84 0.27 11.3%
Kenta Maeda 1st Time 28% 3.08 3rd Time 17% 4.34 -1.26 11.0%
Jorge de la Rosa 1st Time 20% 4.99 3rd Time 9% 6.60 -1.61 10.9%
Aaron Sanchez 1st Time 27% 2.26 3rd Time 16% 4.85 -2.59 10.7%
Zack Greinke 1st Time 24% 3.15 3rd Time 14% 5.20 -2.05 10.0%
J.A. Happ 1st Time 24% 3.53 3rd Time 14% 5.40 -1.87 9.9%
Zach Eflin 1st Time 14% 5.58 3rd Time 4% 6.67 -1.09 9.9%

Juan Nicasio aside, you’ll immediately notice that a lack of pitches isn’t the key problem for all of these players. Matt Harvey and John Lamb have plenty of pitches — so many, it’d be hard for them to add another without doubling up on one they already throw. What they did have last year was trouble with stamina. Between the first and sixth innings, Harvey and Lamb (and Dylan Bundy) lost over a tick on their fastballs. (Mike Clevinger lost almost two ticks.) That’s much more than the average, which is only around one-third of a mile per hour.

But that doesn’t mean there aren’t some obvious short-arsenal pitchers here. If you add up fastball usage with the first primary out pitch, Nicasio, Joe Ross, Aaron Sanchez, Luis Perdomo, and Zach Eflin were all over 88%, with Nicasio topping out at 97%. If you count cutters as fastballs, you can add Phil Hughes (at 95%!) and Ryan Vogelsong. All have been relievers recently or will be relievers this next year, so it’s not too surprising to see them grouped here.

At the nexus of both of these effects is Nationals righty Joe Ross. He lost about a full mile per hour between his first and sixth innings last season, and he also throws one of two pitches nine out of ten times. He went from a 28% strikeout rate and lights-out results the first time through the order last year to a 16% strikeout rate and pedestrian results the third time through. His difference in results was 22nd worst, behind some past or future relievers like Aaron Blair, Luis Severino, Alfredo Simon, and Jake Thompson.

We know very little about pitch types and injury, even if there’s a whiff of injury around throwing a slider as often as Ross does (38% career). Or as often as Tyson Ross does (36%). You’d think that throwing something with slightly different mechanics would help activate different muscles and ligaments and not stress the same ones over and over again, at least. And, considering their similarities and histories, injury is a key word for the Ross brothers.

But just as important is turning that lineup over that third time. To his credit, Joe has been throwing his change more than his brother ever did, and by results (whiffs at least), it’s been almost average. Brooks Baseball says that Joe only threw 11 curves last year, and maybe those were just slow sliders, but they did drop three inches more than his sliders, and do offer Ross some upside.

More changes or more curves, though, and Ross could even out those splits and become the pitcher his slider suggests he can be. I’m sure he’s never heard that piece of advice before.

With a phone full of pictures of pitchers' fingers, strange beers, and his two toddler sons, Eno Sarris can be found at the ballpark or a brewery most days. Read him here, writing about the A's or Giants at The Athletic, or about beer at October. Follow him on Twitter @enosarris if you can handle the sandwiches and inanity.

Newest Most Voted
Inline Feedbacks
View all comments
7 years ago

Does anyone know if, beyond the base effect of having multiple pitches, there are pitch usage patterns that correlate with relative success 3+ times through? I often see pitchers and announcers cite that as a strategy; for example, a pitcher may hold one of his pitches back first time through and only through it later in the game, or he may dramatically reduce his fastball percentage as he goes. Both seem like they would work for the same logical reason that having more pitches would work.

7 years ago
Reply to  Bip

Only “through” it, sure, that’s real special.