The Platoon Split You May Have Never Heard Of
Writing about baseball isn’t the most predictable task. I often don’t know what my topic will be until the dust settles, hours after rummaging through a pile of numbers that, at first glance, makes little sense. For example, this article started off as an inquiry into Darin Ruf. Of all the journeymen to stop by the KBO, experience a resurgence, and return stateside, he’s by far enjoying the greatest success – who would have guessed?
As a right-handed hitter, Ruf’s primary asset is a knack for mashing lefty pitchers. He can hold his own against righty pitchers, too, posting a 126 wRC+ against them last season. But detractors might point to a .386 BABIP that buoyed much of that production. In other words, one could expect Ruf to become a bit more… rough in the future (sorry). A quick search reveals that he had a higher groundball rate against righties compared to lefties, which doesn’t bode well for future success, and not much else. The critics win this round.
Here’s the thing, though – he wasn’t alone. It turns out that in 2021, right-handed hitters had a higher groundball rate against right-handed pitchers; conversely, they had a lower groundball rate against left-handed pitchers. You can see for yourself:
| P Throws | vs. RHH | vs. LHH |
|---|---|---|
| Right | 43.4% | 41.4% |
| Left | 41.1% | 47.5% |
This is also true of left-handed hitters. Facing same-handed pitchers led to more groundballs, while opposite-handed pitchers led to, well, the opposite. The gap in groundball rate by pitcher handedness is greater for lefty hitters, though that may be influenced by the relatively few instances of lefty-versus-lefty matchups. Still, the difference, which appears on a league-wide scale, is significant enough to warrant an investigation.
So, what’s causing it? Just a fluke season, perhaps? Not really, because you could observe this trend in the pandemic-shortened 2020 season, too. Ditto the year before that. And if we keep going back, all the way to 2002, the first year batted ball data is available at FanGraphs, this is what we see. First, the righty hitters:
Then, the lefties:
Yep, it’s a similar story. This weird, obscure, yet entirely real divergence has been around for at least two decades, and there doesn’t seem to be prior research on the subject.
Maybe we can figure it out. Consider the different types of pitches hitters encounter, depending on the opposing pitcher’s handedness. Righty pitchers are more likely to offer righty batters breaking balls that swerve away from the barrel of the bat, in hopes of inducing weaker contact or whiffs. Against lefty batters, however, those same pitches swerve into the bat’s sweet spot – not good! It’s why righty pitchers will bring out their changeup against lefties, and why the development of one is still seen as a prerequisite for becoming a starter.
The idea: Pitchers tend to approach same-handed and opposite-handed batters differently, and this habit is driving the discrepancy in groundball rate. To test it, I scraped together every pitch put into play in 2021 – excluding bunts, because ew – and looked at the pitch type distributions (fastballs, breaking, offspeed) for each possible batter-pitcher matchup (e.g. RHH vs. RHP). Here’s the rough distribution of pitches that righty hitters put into play last season:
| P Throws | Total | Fastball% | Breaking% | Offspeed% |
|---|---|---|---|---|
| Right | 45,885 | 59.3% | 33.6% | 6.8% |
| Left | 26,102 | 58.1% | 19.3% | 22.4% |
Followed by the lefty hitters:
| P Throws | Total | Fastball% | Breaking% | Offspeed% |
|---|---|---|---|---|
| Right | 37,942 | 56.8% | 21.5% | 21.5% |
| Left | 9,824 | 62.3% | 32.6% | 4.7% |
These numbers aren’t jarring at first glance. More breaking balls versus same-handed batters? Sure. An offspeed bonanza instead against opposite-handed ones? Yeah! But keen readers might have spotted something peculiar. Offspeed pitches like changeups tend to generate low launch angles when put into play; it’s no surprise that they’re the preferred offering of soft-contact champions like Logan Webb and Framber Valdez. Yet confusingly, it’s the same-handed batter-pitcher matchup – which features an abundance of breaking pitches, not offspeed – that’s responsible for the higher groundball rate. That doesn’t add up. What gives?
There’s another pitch famous for creating grounders: sinkers. They’re here to provide us with a better answer. It’s obscured under the all-encompassing fastball label, but when sinkers are isolated, we can observe another handedness-based trend. Against same-handed hitters, pitchers have a slight tendency to throw more sinkers:
| P Throws | vs. RHH | vs. LHH |
|---|---|---|
| Right | 17.6% | 12.4% |
| Left | 15.0% | 18.6% |
And it seems like pitchers do so to gain an advantage – in 2021, for example, righty-on-righty sinkers returned a .341 wOBA on contact and a two-degree launch angle, whereas righty-on-lefty sinkers returned a .390 wOBA and an eight-degree launch angle. Digging through lefty pitcher splits and prior seasons returns similar numbers. Throwing sinkers to same-handed hitters is, without a doubt, the safer option. Tying this back to our investigation, that lower launch angle in tandem with an uptick in usage likely outweighs the effects of the offspeed pitches opposite-handed hitters receive. Moreover, unlike with sinkers, offspeed pitches don’t really have a launch angle split based on handedness. Righty-on-righty offspeed pitches averaged seven degrees; lefty-on-lefty ones averaged nine in the same year. Here’s a table that sums it all up:
| P Throws | vs. RHH | vs. LHH |
|---|---|---|
| Right | 2 | 8 |
| Left | 6 | -2 |
But that last detail makes this all the more confusing, and honestly, I’m stumped. Movement-wise, sinkers and changeups are extremely similar, with the latter dropping an extra few inches. Maybe minor differences in pitch shape are responsible, but that doesn’t seem like a satisfying conclusion. There are also count- and location-based factors to consider: Sinkers are used to nab strikes early in the count, while changeups are usually seen as an out pitch. And while sinkers can also thrive up in the zone, for the most part, changeups end up down and away – just where you would expect. Further research is required, no doubt.
It’s also interesting to consider how this ties into the notion of platoon splits. When watching a righty batter hit a moonshot off a feeble lefty, we might be inclined to think that contact quality is the most prominent aspect of a platoon split. But according to research from Baseball Prospectus, it’s the tendency for batters to strike out less and walk more against opposite-handed pitchers, rather than a metric like slugging percentage, that’s sticky year-to-year. Platoon splits, it seems, are a matter of plate discipline.
I’d argue, though, that the groundball rate split is nearly as important as the strikeout or walk rate split. It might not be consistent among individual players, but collectively, the league has long had an easier time elevating against opposite-handed pitchers. Righty hitters last season had a .184 ISO against lefty pitchers, a number that dwindled to .161 against righties. It’s roughly the distance between Mike Yastrzemski and Aaron Judge, all because of those pesky groundballs.
That in and of itself isn’t a seminal discovery, of course. But I’ve yet to come across an article that discusses platoon splits in terms of a discrepancy in groundball rate, which seems to be caused by… sinkers? Maybe this isn’t the most relevant subject, but the mystery of it all is fascinating enough for me. At this point, I pass the baton onto whoever stumbles upon this, granting them full permission to yell at me in the comments for omitting this or that. Same-handed batter-pitcher matchups result in a higher rate of groundballs. I have no further updates about major league baseball’s lockout.
Justin is an undergraduate student at Washington University in St. Louis studying statistics and writing.
i’d imagine a right on left changeup (for example) is relatively more likely to get a whiff whereas a right on right changeup is relatively more likely to induce a groundball and vice-versa. to what extent (if at all) could this explain the higher gb% that we see when the pitcher has the platoon advantage?
Very cool
Rarely do I read an article more than once. But I’ve noticed that I tend to reread an article from Choi than from most writers. This one is already saved for the first reread.
It makes sense. We’ve always known righties hit better against lefties and vice versa, and fly balls overall provide for a much higher OPS and wOBA than ground balls. Why would you expect anything different?
I’d offer that sinkers have a higher GB% than change-ups due to pitch velocity. As a hitter, to elevate pitches you want to get extensions. A change-up is essentially designed to cause too much extension and that’s why you swing and miss. So when a hitter just misses the timing on a change they’ll likely be extended and hit it weakly in the air. I think this would show up when you look at the GB% of power-changes usually being substantially higher than their slower brethren.
Exactly what I was going to say. It’s hard to pop something up when you’re late, and getting jammed often leads to grounders. Getting out front is more air balls (unless you roll over)
As a former high school sinker ball guy, this was my problem, nothing but grounders from righties. Sure changeups have the same pitch shape as sinkers, but against lefties they are supposed to run away from the bat head when a lefty is way in front for whiffs and nubbers off the end. A sinker is designed to look hittable and get a guy to drill it into the grass, the problem is with lefties they have it timed up and on the barrel unless you have Maddux or Hendricks like control and can put it on the black running away.
Matt made a good point, and from my experience as well.
If you are right-handed, and have a good sinker, it will tend to move down-and-in to a right-handed batter. That means you can jam him/her with it, it looks good initially, the batter takes a mighty cut with the result a grounder to third. If you make a mistake and get it a little bit over the plate, well then the most likely thing is still a grounder, but now to short, or perhaps a single through the hole.
Against a left-hander, you throw that pitch hard to the outside and they will tend to lay off it. You have to throw it slower, e.g. a change-up, to get them to bite. Trying to throw that pitch inside to a left-hander, they might bite, but the movement will bring it more to the barrel of the bat and they get more power on it.
the secret to Jamie Moyer, Tommy John, those kind of pitchers was about half their “sinkers” were actually slow cutters, so they were coming into the right-handed hitter aimed at their back foot. The hitter would probably foul that off to the third base side, then they come back with a sinker that looks the same, but moves away, and they roll it over. Easy to say, hard to do.
When hitters have the platoon advantage, movement from a sinker moves along their bat path, from the barrel toward the end of the bat, roughly along the axis of the bat.
When pitchers have the platoon advantage, the movement from a sinker causes it to dip almost perpendicular to the axis of the bat, making it much more effective, causing hitters to beat it into the ground.