Tyler Duffey, the Shockingly Interesting Twin

I decided not to write a post about it, but earlier I went into the spreadsheets to compare some of the fan projections for pitchers to some of the Steamer and ZiPS projections. The idea was basically to see if there are guys the fans are particularly high on or particularly low on, and as the former is concerned, the fans are higher on a bunch of relievers. Let me tell you, there are some real believers here in Mychal Givens. And that’s great! I love Givens, too. He’s really interesting, but he’s also a reliever, and I found myself scanning for starters. A name that quickly turned up is Tyler Duffey. Steamer and ZiPS see him good for a 4.37 ERA. The few fans who participated see him good for a 3.58 ERA. That was enough to grab my attention, and now we have an article.

Duffey, despite a strong 2015 debut, remains pretty anonymous. If it weren’t for the headline, I wonder how many of you would’ve known he pitches for the Twins. We’ve been conditioned to mostly ignore the various Twins starters, and for the most part that’s been a pretty sound policy, but Duffey has some unusual things about him. He was also all but guaranteed a rotation slot the other day by Paul Molitor. Consider this, then, a Tyler Duffey introduction, in case you’ve been in need of one. Twins fans know what’s up, but I’m guessing the others are almost all in the dark.

Duffey just made 10 starts in the majors, down the stretch. He averaged a little shy of a strikeout an inning, and he wound up with the same ERA- as Marco Estrada and the same FIP- as Gio Gonzalez. As I write that I realize the names might not have the strongest impacts, so let’s try that again: Duffey wound up with a better ERA- than Madison Bumgarner, and he wound up with a better FIP- than Cole Hamels. He was a 24-year-old righty, and over 32 starts in the majors and the high minors, he allowed a total of just five home runs. It was a big, big year for Tyler Duffey.

When Duffey was drafted in 2012, he was drafted after having been a reliever. The Twins made him a starter in 2013, and his minor-league ERA is in the upper 3s. He never ranked as a Baseball America organizational top-10 prospect. Kiley didn’t even mention Duffey in his Twins system write-up after 2014. Duffey didn’t just have success last season at the very highest levels — he had success after having come almost out of nowhere. He emerged, like pitchers do sometimes, and now we have no choice but to take him seriously.

Now for the matter of the Duffey repertoire. The repertoire might help explain why Duffey used to be almost invisible. Yet the same repertoire also makes Duffey’s success impressive. Pulling from our leaderboards, last year, Duffey threw almost 60% fastballs, and he threw almost 40% curveballs. The fastball averaged about 90 miles per hour. A little before Duffey’s promotion, he talked about the improvement of his changeup in the minors, but then he stayed away from the changeup with Minnesota, not throwing a single one after the start of September. Duffy leaned heavily on his curveball — he had a higher curveball rate than any other starter, while as breaking balls go, he was ever so slightly eclipsed by Tyson Ross and Chris Young, who threw about 41% sliders.

It’s tempting to call Duffey a two-pitch pitcher. We’re always supposed to be skeptical of two-pitch pitchers. And Duffey, for sure, would benefit from having a reliable changeup, but he actually has three regular weapons: the curveball, a four-seamer, and a two-seamer. Chris Young has talked with Eno in the past about how the fastball can be any number of different weapons, depending on how you use it. Duffey occupied three areas with his three pitches. Pulling from Baseball Savant, here’s where Duffey put his curves:

Here’s where Duffey put his straighter fastballs:

Here’s where Duffey put his tailing fastballs:

There is some real diversity in there. Duffey threw his four-seamer to the glove side, and he threw his two-seamer to the arm side. Throw in the curve and you have three pitches Duffey came with on a regular basis. Yeah, two of them were fastballs, but the important thing is they were different fastballs. So while another pitch would be nice, I’m not actually convinced that it’s necessary. Especially given how Duffey can change the speed or look of his curve. For a young pitcher, he has an excellent feel for his breaking ball.

The breaking ball is what stands out. Even if Duffey does develop a useful change, the curve is his real bread and butter. It takes a high degree of confidence to use a secondary pitch two-fifths of the time, but Duffey used the curve to get 85% of his strikeouts. Here are two strikeouts, showing different ways Duffey can use the pitch. He can bury it, or he can use it for freezing purposes.

There’s a really neat thing about the curveball. How many times have I talked about my pitch-comp formula? Using PITCHf/x information, I can calculate similar individual pitches by looking at the average velocities, horizontal movements, and vertical movements. I decided to run pitch comps for curveballs thrown by starting pitchers in 2015. The curveball most similar to Duffey’s, on average: none other than Felix Hernandez’s curveball. And that particular curve was arguably the best in the game. Felix has about one extra mile on Duffey, but the movements are almost identical. Now, of course, Felix also has his changeup, and he shows a slider, so he has more going on overall than Duffey does. But if Duffey is known for anything, he’s known for his hook, and it turns out it compares well to the hook of one of the elites.

Another side effect of Duffey’s curve being so good — he picked up a lot of called strikes in the zone, and he also generated plenty of swings out of it. As a simple little measure of deception, you can subtract a pitcher’s out-of-zone swing rate from his in-zone swing rate. Last year, 186 starters threw at least 50 innings, and they had an average difference of 33%. Duffey had a difference of 23%, which was second only to Hisashi Iwakuma’s 22%. Duffey was about tied with Masahiro Tanaka, and he was a little in front of Felix. This is a stat that holds up to some degree year-to-year, and for Duffey it’s a positive indicator. He wasn’t easy to pick up, so batters got themselves in trouble.

That’s Tyler Duffey. He’s an unfinished product. Like everyone, he could use better command, and he’ll probably need to show his curveball earlier in counts to keep hitters honest after scouting reports go out. Maybe Duffey really will need that changeup after all, if hitters start picking up the curve or just eliminating it. Duffey isn’t on the verge of becoming an ace, and perhaps he’ll end up being the one to get the boot for Jose Berrios, but for right now, Tyler Duffey is a pretty interesting looking Twins starting pitcher. A year ago he was on nobody’s radar. These are the breaks that can help an organization get itself back to contention. The Twins, for their part, might argue it was no break at all.