It’s become a cliche, but the sentiment still holds true: the best pitch in baseball is a well-located fastball.

Overwhelmingly, the most commonly thrown pitch in the game is the fastball — whether it’s a four-seamer, cutter, or sinker — and it has been since the sport was invented. But as more data becomes available, there’s an interesting trend developing among starters. Fastball usage is on the decline. Since 2002, the first year PITCHf/x data is available, starters have gradually thrown the heater less.

There are likely many reasons for this. One cause is breaking balls are simply not hit as hard as fastballs, and the growing trend of using them in traditional hitter’s counts (2-0, 3-0, etc.) rather than a fastball right down Broadway keeps hitters off balance. Against fastballs, batters are managing an average exit velocity of 89.2 mph and a .453 slugging percentage in 2018. Versus breaking balls, those numbers drop to 87.0 and .353, respectively.

But there’s also been an emphasis on pitchers throwing their best offering more frequently, and it has occurred both with players who do and do not have a quality fastball.

Clayton Kershaw has done it. Masahiro Tanaka, along with basically every other Yankees pitcher, has done it. Patrick Corbin’s breakout season is in part attributed to it. But the most notable example in recent memory is Lance McCullers Jr. Yes, the same guy who threw 24 straight curveballs in Game 7 of the ALCS to help send the Astros to the World Series last year.

McCullers is one of the most extreme examples of “pitching backwards,” as his curveball is his primary pitch. Starting in 2016, he stopped trying to throw a fairly ineffective fastball the majority of the time and turned to the biggest weapon in his arsenal — a lot. Over the past three seasons, McCullers has thrown his curveball 46.9% of the time, by far the most in baseball among starters. He still uses his fastball, now preferring a sinker to a four-seam, but he’s cut its usage down to 40.9%. This season, he’s throwing it less than ever (39.7%) as he has mixed in a changeup more often.

For McCullers, the primary problem with his four-seam fastball had nothing to do with velocity — he could fire it at 99 mph, after all. Rather, command was the issue. Despite his electric stuff, he struggled to get ahead in the count with his fastball and the pitch got crushed when he came in the zone with it.

Sound like anyone Reds fans know?

While not an identical case, Robert Stephenson fits a similar mold. The former first-round pick can dial his fastball into the upper-90s and average around 93-94, but it has been an ineffective offering because he can’t control it consistently. Traditionally, the fastball is meant to be a get-ahead pitch. Although Stephenson has used it on 57.4% of his big-league pitches, it has found the strike zone only 48.6% of the time. Among all pitchers who threw the pitch 500 or more times in 2017, only eight had fewer called strikes for every ball thrown than Stephenson (0.40).

But Stephenson continues to rely on it when he falls behind hitters anyway. In hitter’s counts, he’s thrown the fastball 58.2% of the time. With three balls, 63.9% of his pitchers are heaters. It’s not particularly surprising, then, that on 59 of his 72 career walks, ball four came on the four-seamer.

When he does throw it in the strike zone, he doesn’t excel at painting the corners or keeping the ball high or low enough that hitters swing through it. In his 121 2/3 major-league innings, the four-seamer has been crushed for a .324/.451/.606 slash line, and 16 of his 21 home runs have come on the pitch. Here’s a look at the career heat map:

What Stephenson is missing in fastball effectiveness, however, he makes up for with his secondary offerings. His curveball held batters to a .148/.172/.259 slash line last year and had a respectable 13.0% whiff rate. His changeup registered a 21.2% whiff rate. Now we’re talking. Only 16 other pitchers in baseball who threw the pitch 200+ times had a higher percentage of swings and misses. Despite the strong whiff numbers, Stephenson doesn’t have the same dominant results (.263/.290/.404), but the pitch is a solid one when he keeps it down in the zone.

Neither of those would be the pitch to take the McCullers approach with, however. His best offering is, funnily enough, the newest addition to his arsenal: the slider. He just picked it up last year, and it’s already an elite-level weapon. The opposition managed an 83.8 mph average exit velocity (league average for SPs: 85.2 mph) and a pitiful .136/.177/.203 line against it when they put it into play, which wasn’t very often. Per Statcast, Stephenson got a whiff on the pitch 25.0% of the time he threw it. Of the 221 pitchers who used the slider 200 or more times last year, only 15 had a higher whiff rate than that. Among the names just ahead of Stephenson: Carlos Carrasco, Max Scherzer, Clayton Kershaw, and Zack Greinke.

Would increasing his slider usage and decreasing his fastball percentage unlock Stephenson’s potential? It seems plausible based on last year’s results. As the chart below shows, he gradually became more comfortable with the new breaking ball as the year moved on, using it around 30% of the time in his August and September starts. And the results were promising. In those two months, he posted a 2.50 ERA and 24.3 K%. But that came at the expense of his curveball rather than the fastball, and his walk rate remained high at 13.1%.

Would decreasing his fastball usage help him sort out the command problems? There’s no way of knowing unless he does it. I’m not a pitching coach, and I don’t pretend to be. The Reds have significantly more data than I do, so they may have concluded that this approach would be a complete disaster for Stephenson. Maybe he’s also unwilling to give it a shot. But just looking at these numbers, it’d be an interesting experiment to try.

Stephenson is again in the midst of an inconsistent season in Triple-A Louisville (4.11 ERA, 26.6 K%, 13.9 BB%) as he tries to claw his way back to the major leagues. His strikeout rate indicates his raw stuff is still strong, though the walk rate would tell you that fastball command remains an issue. Perhaps unfurling his best pitch with more regularity could be beneficial for the 25-year-old and help him become a mid-rotation-level starter or better. For a rebuilding team whose ace is arguably a pitcher with two starts under his belt in the last year-and-a-half, that would be a significant development.

22 Responses

  1. seanuc

    Awesome post. Pulling for Stephenson because we need him; guys with his stuff and easy delivery are so hard to find.

  2. Tom Mills

    Well researched idea. I hope someone in the Reds system sees it.

  3. James H.

    I think you’ve struck gold, TBH. Maybe send this in an email, and not just a link to your article? What could it hurt?

  4. Scott Carter

    Good stuff. I hope somebody in the Reds organization sees this and convinces Bob Steve to try it. Man we need him at the MLB level. Still the best pure stuff I believe in the organization, with perhaps the exception of Iglesias.

  5. SultanofSwaff

    Thank you!!! I’ve been advocating for BobSteve to ‘pitch backwards’ for him for a long time.

    On a more macro level, I think you could argue this approach should be employed across the board in the organization. When I look around the league I’m always amazed at how average the ‘stuff’ of other pitchers is compared to our pitchers and yet our guys get knocked around like pinatas. There has to be more to it than ‘our guys suck’. We’ve eliminated one variable (Price/Jenkins), yet the pitch sequencing is still dogmatic in hitters counts. Maybe it’s time to reassign the advance scouts in favor of some new blood. Finally, you can’t absolve Tucker Barnhart either. I’d be curious to know how he processes and implements his information. Of course, this would require reporters who ask real questions.

  6. Bill

    Am I wrong, but it seems like most left handed batters like the low and most right handed batters like the ball up. I know not all do but most, so why not try to do the opposite of what they like.

  7. big5ed

    Stephenson has some interesting splits. He has 46 Ks & 14 BBs in 129 ABs against RH hitters, but 21 Ks and 21 BBs in 81 ABs against LH hitters. LHs have hit only .148 off him, but have 5 HRs. RHs hit .233 against him, and have but 4 HRs. His BABIP against LHs is abnormally low, at .127, but is .329 against RHs, as I calculate it.

    His problem, then, is an abominable 20.6% walk rate against LHs, and a 4.9% HR rate. (The almost 10% walk rate against RHs isn’t good, but it is manageable.) The good news is that this would seem to be a mental issue, rather than anything wrong with his arm or his stuff.

    Maybe he just needs to work on his pitch selection against lefties.

  8. Tom Mitsoff

    How is his control of his curveball?

  9. Nick Carrington

    Nice post, Matt. Stephenson’s “success” at the end of 2017 was unsustainable because of his 14.1 BB%. No one succeeds long term walking that many guys. If this strategy could reduce some of those walks, then it’s worth a shot.

    It’s hard to give up on Stephenson because of the stuff he has. His slider is one of the nastiest pitches I’ve seen. But, I’m confused as to why the Reds continue to give him opportunities over other young pitchers who have seemingly improved on their deficiencies. Stephenson’s control remains his biggest problem, as it’s been since 2014.

    • Nick Carrington

      I’m suggesting that results in a short sample aren’t all equal. The way that Castillo performed last year was sustainable based on the peripherals. Stephenson got results by doing it in a way that no has never led to success long term.

      So his ERA in those 11 starts means nothing to me. If he’s going to succeed, he’s going to have to do it in a very different way than those 11 starts. He might do that. But equal results in a short sample don’t mean equal performance. .

  10. big5ed

    We don’t know specifically what DeShields was referring to, and I assume it was more than “Just throw strikes, son.” Maybe he won’t slow his mechanics down. Maybe he overthrows. Maybe he won’t collect and load himself better over his right leg. Maybe he won’t throw inside to left-handers. And maybe they’ve made a bunch of suggestions and given him drills to do, that he just won’t follow through on. Or maybe he’s just raw and wild, still.

    So, who knows, really, why he walks so many guys, particularly left-handers?

    But I am generally with the group here. Bring him up. He is better than Romano.

  11. Bill

    Maybe bring him and let him be wild at 98 mph it may have the same affect it had on John Kruk when Randy Johnson pitcher to him, he just swung at 3 pitches so he could get back in the dugout after Johnson threw a heater over his head.

  12. Chris

    Great read. I pulled up the tubular table and played with it, and unless i did something wrong, here is what i found. 549 0-0 counts. 67.4% of his first pitches are fastballs. If he gets ahead, the hitter has a 34.9% chance of seeing another fastball. If he falls behind, 1-0, the hitter has a 69.6% chance of seeing another fastball. If the count goes to 2-0, the chance of a fastball is 86%. 3-0 is 100% chance of a fastball coming. 3-1 is a 96.3% chance of a fastball. 3-2 is a 76.7% chance of a fastball. Forget tipping pitches. Maybe this is the reason he is in AAA.

    • Matt Wilkes

      Great data, Chris. You highlighted exactly the reason I think pitching backwards would help Stephenson. He’s way too reliant on his fastball, and its not a particularly good pitch.

  13. Tom Mitsoff

    If Finnegan can pitch effectively as a starter, I certainly don’t want him traded. But he hasn’t yet shown any signs of being able to consistently dominate Class AAA hitters, which to me is a good indication that someone is ready to take the step to the next level and be effective. Someone who is average in the minors does not inspire me at all as an answer for the big-league rotation.

  14. Tom Mitsoff

    To that point, Stephenson has a 4.11 ERA in Class AAA. What do you think that will equate to at the major league level? At this point, Stephenson, Finnegan and Romano are all interchangeable, in my opinion. You hold on to them for as long as you can in case one of them figures it out. There is absolutely no reason to trade them, unless another team asks for them in a deal you’re trying to finalize.

    If you get to the point where they’re out of options and you have to say bye-bye, you’re saying good-bye to a guy who couldn’t get it together for years in the minors, not a top prospect.

    I hope they all figure it out, but the best-case scenario is probably that one of them MAY figure it out.

  15. Jeffery Stroupe

    looks like 6 against 9 to me

  16. Streamer88

    I have hope still for Bob Steve. This article’s outline of strategy should serve as his raison d’etre from here on out.

    As a pitcher you can deceive with pitch type, pitch speed, and location. Bob Steve should abandon perfecting the latter most, and instead focus on the former two variables, as you suggest.

    One extreme is Greg Maddux who came up with erratic, electric stuff, opted to tinker by developing about 12 iterations of his pitches and finally became the greatest control pitcher of all time.

    He went stuff, pitch diversity, control, in that order, for his development. Why not Bob Steve too (in a non HOF fashion)?