Marcus Stroman Discovered Roy Halladay’s Sinker

You’ve heard it all before. Regress to the mean. Don’t make too much of a small sample. Don’t believe in the predictive power of second-half statistics, if they look particularly different from the first-half statistics. You know all the ways you are and aren’t supposed to interpret a player’s numbers. But you also know the key to exceptions, which many try to exploit: when a player makes a legitimate change, his prior numbers become less useful. A change, I mean, to his approach, or his mechanics. The White Sox don’t care too much about Zach Duke‘s history, because he recently changed his delivery. The Tigers don’t care too much about J.D. Martinez’s history, because he recently changed his swing. Marcus Stroman was never bad, but he, too, made a change. It’s real easy to spot on the following image, from Brooks Baseball:

That’s Stroman’s big-league 2014, broken down by month. There’s no arguing the major trend: over time, Stroman threw more two-seamers, or sinkers, and far fewer four-seamers. It’s a dramatic shift, and it’s a dramatic shift in the middle of a year. Stroman became something he hadn’t been before.

This, naturally, has been written about. Virtually everything’s been written about. When Stroman talked with David Laurila in 2013, he noted that he pitched off his four-seamer, with almost no two-seamers to speak of. When the two spoke in 2014, Stroman discussed his increased two-seamer usage. He’s always been a tinkerer, motivated by his own curiosity, and it turns out Stroman just found a sinker grip he liked in the middle of this past July. From Eric Koreen:

Stroman started to use the sinker more often in his July 19 start against Texas, after finding a grip he was comfortable while fiddling around with a baseball at home.

And from Jamie Ross:

“I have a sinker now, and it’s good,” Stroman said, proudly. “It’s a good pitch, and I get excited when I throw it. I get excited on the mound. It’s a pitch I started throwing like a month ago, and I’m already using it in games and having success with it.”

In the game against the Rangers, where Stroman debuted his new two-seamer, he threw it six times. In his next outing, he threw it 18 times. The outing after that, 31 times. He topped out at 46. It’s evident that Stroman fell in love with his sinker almost immediately, and this visual might indicate why. I don’t know the source, but, via Reddit:

I’ve been making .gifs of pitches for a long, long time. Back when I was first getting started, and more comfortable with the process, I started to pay close attention to the catchers. Something I noticed quickly is that catchers almost always close their eyes as the baseball is arriving. Of course, it’s not conscious — it’s just the automatic, instinctive response to deliberately getting in the way of a projectile flying at 90 miles per hour. Now look at Dioner Navarro. Look at his eyes. He never blinks. Even he can’t take his eyes off of Marcus Stroman’s two-seamer.

It’s one thing for a pitcher to introduce a new pitch. It’s another thing entirely for a new pitch to be introduced and then used so heavily, so rapidly. Stroman’s sinker, in this regard, is exceptional, and I wanted to try to find some comparisons. When I’ve played with this kind of methodology before, I once found that Stroman throws a slider that’s a lot like Corey Kluber’s slider. This time, I was just focusing on sinkers, and I pulled up the Baseball Prospectus PITCHf/x leaderboards.

I decided to look at starting pitchers from 2008 – 2014. I narrowed it down to righties, and then I looked for sinkers within half a mile per hour of Stroman’s, on average. From there, it was just a matter of selection and elimination by movement. I looked for sinkers within half an inch of Stroman’s in terms of horizontal movement, on average. Then I looked for sinkers within half an inch of Stroman’s in terms of vertical movement, on average. This was to be my pool of comparable sinkers. The resulting pool of comparable sinkers:

• Roy Halladay’s sinker

And that’s it. That’s the whole group. Granted, in 2013, Halladay was bad. In 2012, he was close to average. But from 2008 – 2011, Halladay was perhaps the greatest starting pitcher in baseball, and his sinker was a huge reason why. Over those four years, his sinker was an even better comp for Stroman’s. He used it to generate 62% grounders. Two-thirds of those sinkers went for strikes. Stroman threw 70% of his sinkers for strikes. It just generated 66% grounders.

Marcus Stroman didn’t simply add a new pitch: he added, to his already broad repertoire, a pitch that compares well to a primary pitch thrown by a possible or probable Hall-of-Famer. And Stroman must’ve known immediately that he’d added something good. From Brooks, look at his first- and second-half pitch usages, broken down by handedness and circumstance:

• against lefties

First Half	LHB, 4-Seam%	LHB, 2-Seam%	Second Half	LHB, 4-Seam%	LHB, 2-Seam%
All Counts	54%	1%	All Counts	20%	35%
First Pitch	65%	1%	First Pitch	23%	39%
Batter Ahead	59%	1%	Batter Ahead	22%	39%
Even	52%	1%	Even	19%	36%
Pitcher Ahead	53%	1%	Pitcher Ahead	20%	27%
Two Strikes	51%	1%	Two Strikes	22%	34%

• against righties

First Half	RHB, 4-Seam%	RHB, 2-Seam%	Second Half	RHB, 4-Seam%	RHB, 2-Seam%
All Counts	51%	0%	All Counts	25%	30%
First Pitch	54%	0%	First Pitch	33%	20%
Batter Ahead	58%	0%	Batter Ahead	29%	36%
Even	46%	0%	Even	26%	26%
Pitcher Ahead	52%	0%	Pitcher Ahead	19%	31%
Two Strikes	47%	0%	Two Strikes	19%	27%

Stroman was comfortable using the sinker when behind in the count against all hitters. Same when he was ahead in the count. Same when he was even. Same on the first pitch. Same with two strikes. He’s got other pitches, a lot of other pitches, and he still has his four-seamer, too, when he wants to elevate some heat, but the sinker has given Stroman something he didn’t have: a groundball pitch he can locate and use to get quick outs.

Usually, you don’t see a big change in a pitcher’s groundball rate unless he’s made a big change to his arsenal or approach. In 2013, in Double-A, Stroman posted a league-average groundball rate. In 2014, in the majors, he was just short of two standard deviations above the average, and that includes his first half in which he didn’t throw sinkers. What Stroman will sacrifice are a few strikeouts. He’ll also probably sacrifice a few pop-ups. But now Stroman will be better able to keep the ball in the yard, and he’ll be better able to work deep.

In 2013, in Double-A, Stroman averaged 4.1 pitches per plate appearance. In last year’s first half, that was also 4.1. In last year’s second half, he trimmed that rate to 3.8. Previously, Stroman would’ve averaged about 24 batters per 100 pitches. In the second half, that improved to 26, demonstrating his better efficiency.

So now look what Stroman did down the stretch, after introducing the two-seamer against the Rangers:

• FIP-: 7th out of 134 starting pitchers
• xFIP-: 18th
• GB%: 5th
• Zone%: 5th

Stroman posted a better second-half adjusted FIP and xFIP than Max Scherzer and Jordan Zimmermann. And Stroman wasn’t even pitching to a particularly good receiver, in Dioner Navarro. For the season ahead, Stroman will get to work with Russell Martin. That would make almost anyone better. As much as people have talked about the Blue Jays maybe needing to add an ace, it would appear they might already have one. Stroman hasn’t yet proven himself over a full year, much less a series of them, but the talent is obvious and the results are encouraging. Stroman was a quality young pitcher before adding Roy Halladay’s sinker. Then, he found it, almost by chance.

About a year ago, the questions concerned whether Stroman would be able to hold up as a starter in the major leagues, given his size. To that question, we don’t yet know the answer, but it’s at least clear the size won’t keep Stroman from pitching like a No. 1. He’s already done that. And it looks like he ought to do it again.