Advance Scout: The Difficulty of Scouting Yu Darvish

Today brings baseball’s first wild-card play-in games. It also brings another baseball first: Yu Darvish’s first start against the Baltimore Orioles, scheduled to get underway at 8:37 PM ET.

You can bet that the prospect of facing Darvish for the first time in a high-stakes game has the Orioles worked into an advanced scouting frenzy. Their season—a magical one, at that—hinges on their ability to analyze (and effectively attack) a pitcher whom their hitters have never seen.

There is some good news for the Orioles: when our own Russell Carleton took a look at historic batter-pitcher matchups, he found that while a pitcher has an advantage over a hitter in their first few confrontations, it’s not a particularly large one, and it dissipates quickly. It should be noted that the historic matchups Russell was researching preceded modern advances in video scouting and PITCHf/x, so modern hitters may have even less catching up to do. Adam Jones believes that reviewing video of Darvish is “about one percent” as helpful as hitting against him, but Baltimore batters’ mileage may vary.

As the Orioles’ analysts, coaches, and hitters study the data on Darvish, what useful information might they turn up? Before we begin, please don’t let me fool you into thinking I have any real experience at advance scouting, which involves very systematic analysis of a pitcher’s approach, including his mechanics, skill at restricting the running game, etc. This is merely a look at what a season’s worth of data has taught us about Darvish’s repertoire.

Let’s start with what makes this assignment particularly challenging: the sample size is small, since Darvish has pitched in the majors for only one season. He throws a mix of seven (!) pitches—and that doesn’t even account for the fact that he delivers the curveball in two different ways. He throws 95. He throws 65. He can be wild, but he can also be sharp. All of this uncertainty, combined with lethal stuff, has conspired to give him the highest whiff rate on his fastball among major-league pitchers (minimum 500 pitches).

Oh, and, if you’re looking for tendencies, avenues of attack, or rules to hammer into your hitters, keep in mind: he’s dramatically changed his approach over the course of the season.

If you’re hitting against Darvish, you have to be aware of all his offerings. This is no simple task. 

True, he throws the changeup and split relatively infrequently. But everything else is always on the table. How can a hitter narrow down the range of possible pitches Darvish is going to throw?

Succeeding against Darvish depends on being able to anticipate which pitches he’ll lean on in particular counts. The problem is that his tendencies have changed over time.

Let’s take a look at some early-season (vintage?) Darvish, from May:

Look at the approach against left-handed hitters: 41 percent four-seam fastballs to start at-bats, and nine percent cutters. If he was ahead of you, he went to the curveball—an odd selection for a right-handed pitcher against a left-handed hitter. He also started righties off with the fastball nearly 60 percent of the time, but when he was ahead, he expanded his repertoire to include five pitches over 10 percent frequency.

Think advance scouting is easy? Let’s fast-forward to September.

Suddenly, against lefties, he’s almost always starting out with the cutter—83 percent of the time! And when the batter is ahead, he goes to the cutter 81 percent of the time! Suddenly, he’s hardly throwing the curveball to lefties at all, and he goes to the split with two strikes 30 percent of the time. It doesn’t get any easier to diagnose what he’s going to do against righties: he’s now using the cutter to get back into at-bats and the slider to put away hitters, and he’s hardly ever throwing his sinker.

With such variable pitch-type choices, it hardly makes sense to break down Darvish by location. But just for the sake of completeness, let’s take a look at where Darvish has thrown the ball and how he’s fared in those locations.

One tendency that Darvish has is to locate the ball low over the middle of the plate. This tendency is apparent in his approach to left-handed hitters:

…as well as in his approach to righties:

Over the plate and low tends to be a good area for hitting, and Darvish has paid the price for his willingness to throw there, surrendering a .369 batting average and a .607 slugging percentage in that location, while getting only 16.5 percent whiffs. So, one useful piece of advice the Orioles might give their hitters is that he’ll frequently attack that low part of the strike zone—and this trend, unlike some of the others that we saw, has held up through September.