The Cards have lost 4 in a row and are losing right now. Rather than confronting this reality and my subsequent anxiety about making the playoffs, today I am going to write about a general baseball topic that was raised this week. On Thursday on Fangraphs Bradley Woodrum wrote an article analyzing hitter swing planes and how they might determine a hitter’s success against ground ball pitchers vs. fly ball pitchers. This is a fascinating topic that elicited a strong response in the comments section, including two aggressively long posts from yours truly and responses from hitting analyst Chris O’Leary.
When it comes to predicting hitter-pitcher matchup outcomes, broadcasters tend to revert to numbers on platoon splits (how the batter performs against all lefties or righties) and the historical matchup between the specific pitcher and hitter. While platoon splits are generally helpful, one would like to think we would have more predictive power if we used more specific information about the matchup. Using the historical outcomes between the players solves this problem, but almost always implies such a small sample size that we cannot draw any meaningful conclusions. The best approach is to find an adequate set of comparable pitchers that the hitter has faced, which increases the sample size and uses more information than just the handedness of the pitcher. I have heard of ongoing work on this topic, but Bradley’s article is the first I have seen looking at how common swing mechanics and swing planes may determine performance against pitchers with a common trait–high GB rates.
I almost always believe numbers should be combined with scouting when formulating an opinion on player evaluation and I think this is a topic with real implications for how teams manage rosters and lineups. That being said, I don’t have any definitive answers as to which kind of swings will perform particularly well against which kinds of pitchers. I would, however, like to offer some thoughts on what factors should be considered if we want to make meaningful predictive claims about matchup performance.
Analyzing swing planes can be equally complicated and controversial. In some cases the controversy is just a function of semantics–i.e. what do we mean when we say uppercut? The analysis is complicated because every hitter is unique and, more importantly, every pitch is unique.
Let’s consider the same pitch, a low fastball down the middle, to three different hitters: Dustin Pedroia, Jason Heyward, and Carlos Pena, who range in height from roughly 5’9″ to 6’5″. So what are we likely to see on this pitch from these three hitters? First, let’s talk about what we will see from all hitters.
All hitters make contact with the barrel below their hands. It’s the physics of gravity. The degree to which this is the case will vary greatly and it was this angle between a hitter’s hands and barrel that Bradley used to define swing plane in his article. All hitters have a swing path that initially takes the barrel down into the hitting zone and eventually back up to their finish. This is partly a function of the first point. No hitter begins his swing with the barrel below his hands. Therefore, the barrel has to come down to make contact and at some point transitions from moving down to moving horizontally and up. Now, in my opinion, differences between hitters in swing plane and who “uppercuts” vs. who has a “level swing” are predicated on how quickly the barrel goes from moving down to moving horizontal and up AND the relative degree of the up angle the barrel moves at through the hitting zone.
Now, back to our three hitters. As Chris O’Leary points out on his site, Dustin Pedroia really hinges his back knee, dropping him lower to the ground than he already is and effectively turning every pitch into a high pitch. Pedroia has a relatively “level” swing in that his barrel quickly flattens out to get into the hitting zone, then stays in the hitting zone for a very long time moving at only a slight up angle. This leads to his relatively low finish and provides a greater margin for error when his timing is off. Finally, Pedroia maintains his upper body posture, meaning his head and torso usually don’t lean very far out over the plate at contact. The effect of this swing path, posture, and the hinging of his back knee is that at contact Pedroia’s barrel will only be slightly below his hands, even on the low pitch we are considering.
Bradley and myself tend to respectfully disagree on how to define Jason Heyward’s swing plane. Heyward starts his hands relatively high and certainly doesn’t have a large hitch. He also finishes relatively low, which suggests a relatively “level” swing. Indeed, this slow motion video promo seems to indicate that Heyward flattens out the barrel pretty quickly and stays in the hitting zone a very long time with only a slight upward angle. What does this mean for Heyward on the low fastball? Normally it is difficult for guys with flatter swings to get to low pitches. A loopier arc either through an uppercut or a steep swing path–meaning a guy who starts high and finishes high (picture a swing path that looks more like a V than a flat U)–tends to handle low pitches more effectively than a flat swing. In the case of Pedroia this is overcome by hinging the back knee. In the case of Heyward this is overcome a couple of ways.
The first is that despite being 6’5″, Heyward sinks into a wide base, meaning at contact he is effectively shorter. Second, because he is 6’5″ Heyward has very long arms, which also helps. Finally, unlike Pedroia, Heyward gets a large degree of tilt in his torso toward the plate, as this picture somewhat demonstrates (it was the best I could find, but for a better look check out the pictures on Bradley’s Fangraphs article). The effect is to both bring Heyward’s body and hands slightly closer to a low pitch and to essentially rotate the angle created between the barrel and his hands. To demonstrate this point, stand with your arms straight in front of you. Now bend at the waist toward the ground. The angle of your arms relative to a vertical line changed, but all you did was lean over. To an extent, this is what happens because of Heyward’s posture at contact. His swing plane is still relatively flat, but by tilting toward the plate the low pitch ends up almost underneath him and the angle between his hands and barrel is artificially increased.
Ok, so we’ve seen two guys without “uppercuts” who have differing methods for getting to low pitches, how about an uppercut guy? When I think uppercut, I think Carlos Pena. Pena starts his hands very low, quickly flattens out and gets his barrel going uphill. Posture is again an important factor in generating this swing path, but for Pena it is tilt backwards rather than towards the plate that matters. All good hitters drive into a stiff front leg and have some degree of tilt backwards, but as the picture below demonstrates Pena’s extreme tilt generates an uppercut that allows him to catch low pitches on the way up in his swing path and generate loft and power.
Hand position, posture, height, and hinging the back knee are all factors that contribute to the way a hitter attacks a low pitch, irrespective of swing plane. For this reason, I believe that analyzing which types of hitters will handle low pitches better than others is more complicated than simply looking at swing plane. It is certainly more complicated than looking at the barrel-hand angle at contact, which is mostly driven by pitch location anyways. This is only the tip of the iceberg, though, because figuring out how hitters will perform against a group of pitchers depends on a lot more than analyzing pitch location. Velocity, movement, and pitch mix may all contribute to making a pitcher a “GB pitcher” as much as their ability to keep the ball down. It may be these factors as much as pitch location that align with hitters’ swing mechanics or approaches to determine their splits against GB vs. FB pitchers.