What the Dodgers Asked Yu to Do

Before his first start with the Dodgers, just after his trade from the Rangers, Yu Darvish sat down with general manager Farhan Zaidi for a conversation about that night’s game. Andy McCullough relayed some of the details:

At the team hotel in Manhattan, Darvish met with general manager Farhan Zaidi, who advised him on how to attack that night’s hitters. Zaidi opened a laptop and revealed how Darvish could optimize his arsenal, altering the locations and pitch sequences he utilized during five seasons with Texas.

What a fascinating moment! Where so many teams might have shied away from tinkering with a newly acquired, fully formed star, the Dodgers jumped right in with suggestions, and the player was all ears. The team came with so many adjustments in hand that, today, in a press conference, the starting pitcher joked that the only thing they didn’t ask him to change was his “beautiful face.”

So, what were those changes?

The obvious place to start is with pitch mix. No adjustment can be made as quickly as one that involves pitch selection. It’s certainly the least intrusive. That said, the differences between Darvish’s approach before and after the trade aren’t immediately obvious. He hasn’t thrown a slow curve with the Dodgers, supposedly, but he only threw 30 in his 22 starts with the Rangers, so that’s not a big deal.

He has nearly doubled his curveball usage. In that first game, for example, he threw more curveballs than he had all year. But he’s had higher curveball rates in previous seasons, so maybe that isn’t a big deal, either. More cutters, fewer sinkers… but all on the order of a couple different pitch types a game, and from a guy who has eight pitch types in some classification systems.

Things become a bit more stark if you look at what he picked in certain situations. As McCullough points out, sequencing and location were an important part of the adjustments they were asking him to make.

While Darvish has changed the location on his four-seamer to righties, it might just be a product of improved command. Early in the season, he was throwing too many four-seamers down the middle. With the Dodgers, he’s moved his heat map on the pitch more down and away. “Don’t throw it middle-middle,” isn’t what you’d call high-level analysis, though.

There’s something probably more significant to be found with Darvish’s cutters and curves, both of which he’s throwing more often to lefties.

Check out his cutter location to lefties early in the season:

And then with the Dodgers:

It appears as though he started throwing them higher in the zone to lefties.

As for the curve, he threw it low in the zone to lefties with Texas:

Then, with the Dodgers, he has really been burying the pitch against lefties.

These things go hand in hand. Throw the high cutter to set ’em up, bury the curve to finish them off.

This is what it looks like in action.

A high cutter or slider to Michael Conforto during Darvish’s first start with the Dodgers:

And then the buried curve:

There’s always going to be debate about the finer points of pitch classification — what one pitcher calls a slider another might characterize as a cutter — but thanks to Jeff Long and Rob McQuown of Baseball Prospectus, we can look at which pitches Darvish is throwing in certain sequences. And it seems, from the data, that two of the three pairs of pitches Darvish is throwing more often involve the cutter.

Yu Darvish Pitch Pairings by Team
Pairing Rangers Pre-% Dodgers Post-% Diff
SL|FA 127 7.8% 71 11.1% 3.3%
FC|FC 58 3.6% 40 6.2% 2.6%
FC|SL 67 4.1% 32 5.0% 0.9%
SOURCE: Baseball Prospectus
SL= Slider, FC = Cutter, FA = Fastball, SL|FA means slider, then fastball.

In addition to leaning on his cutter more often in certain sequences, Darvish has dropped his arm slot with the Dodgers. That new arm slot has added nearly an inch of horizontal movement to the pitcher’s slider, change, curve — while his cutter’s horizontal movement has stayed the same. He is getting an inch less drop on the cutter with the new slot, and that goes hand in glove with the new approach. Cutters and sliders thrown high but with less drop, followed by a more sweeping, back-foot curve, buried deep: this is Darvish’s new approach against lefties.

With the Rangers this year, Darvish was struggling against lefties, allowing a career-worst 4.41 FIP. That’s improved to a 3.93 number with the Dodgers, where he’s getting 1.8 ground balls for every fly ball instead of 1.3 like he did with the Rangers.

While we may not have access to the actual content of the conversation between Darvish and Zaidi, we can get a sense from the pitch-tracking data of what the latter’s suggestions might have been. First, drop the arm slot, thereby changing the movement on the cutter and curve. Next, throw that pair of pitches in different locations. If that was, indeed, part of the plan, it seems as though the Dodgers have helped the veteran become an even better pitcher. Maybe we’ll see it in action tonight as he takes on the Diamondbacks in Game Three of the National League Divisional Series.





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.

3 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Newcomer
6 years ago

I can’t help but see that the 4.41 FIP vs. lefties with the Rangers and the 3.93 FIP vs. lefties with the Dodgers almost perfectly matches the rough 0.50 ERA/FIP league adjustment I’m used to making to compare AL and NL pitchers, though. Which is to say, you’ve discovered some changes in approach, but I’m not sure there’s a real difference in performance outcome there. But I haven’t looked into if the leagues have grown closer together in a while.