Hitting the ball on the ground creates something of a dichotomy for hitters and coaches. Some demand it, trying to put pressure on the defense by forcing more fielders to handle the ball on each play. A ball in the air can be caught for an out, but a ball on the ground requires a catch, a throw, and another catch to retire the batter. Others would like to hit everything in the air because there is only so much damage that can be done by a ground ball. With these conflicting beliefs, it may be difficult to understand what approach is truly best for all hitters, or each individual hitter.
Confusing things further, both of these camps often will talk about using the same swing to get opposite results. You can swing down to hit the ball on the ground, or you can swing down to produce backspin and hit it in the air. It seems irrefutably silly that both of these philosophies are alive and well in the baseball world, at all levels (including the highest levels, by my sources). Not only is this a paradox, but it is also the wrong kind of swing altogether. Swinging down greatly reduces the chances of squaring a ball up when the timing of the swing is not perfect, which is something we’ll look at later. First, let’s focus on balls after they have left the bat.
At the amateur level, fielders and field conditions are not as excellent as in the major leagues. Fewer ground balls are converted into outs due to reduced range, more errors, and bad hops. Additionally, power is a smaller part of the game in high school and college since players are not as physically developed. There are fewer players able to drive the ball consistently to the outfield and over the fence. Given this, it seems more reasonable to try to hit the ball on the ground at these levels. However, I would not advocate it for professional hitters, or those aspiring to be. The data I am presenting tell a different story.
Let’s start with some numbers. If a major league hitter wanted to gear his swing toward one type of batted ball, which one should he pick?
|Batted Balls Outcomes for 2013 MLB regular season|
|Batted Ball Type||OBP||SLG||OPS||ISO||wOBA|
As most of us probably expected, line drives are the most desirable result. Fly balls have a large slugging percentage but the lowest likelihood of leading to a hitter reaching base. Ground balls own the lowest weighted on-base average of the three, and it’s not really close. Based on these simple stats, hitters should be trying to hit hard line drives to have the most success with a lean toward fly balls if they mis-hit a pitch. The 19 point advantage in on-base percentage does not make up for the 385 points in slugging percentage between ground balls and fly balls.
It should be noted that reaching on an error is much more likely on a ground ball than it is on a line drive or fly ball. Let’s assume that all reached on errors are from ground balls, and the full credit belongs to the hitter’s ability to get on base this way. The best reached-on-error rate last year was by Andrew McCutchen at 48 ROE/PA (reaches on error per plate appearance). If we crudely add that into the league average numbers, it equates to approximately 21 points in OBP, still nowhere near overcoming the enormous difference in production due to the extra-base potential of a fly ball. Again, this may play a bigger part in amateur baseball, but major league teams are pretty good at getting outs when you give them a ground ball.
This seems obvious from a majors-wide viewpoint, which is why so many pitchers try to force ground balls, and some get paid primarily for their ability to do just that. For another point in favor of hitting fly balls, I refer back to Tango and Lichtman’s analysis in The Book of groundball pitcher vs. flyball hitter matchups. Flyball hitters have a distinct advantage over groundball hitters against groundball pitchers. As more teams focus on stacking staffs with groundball pitchers, the importance of hitting the ball in the air becomes more important. To me, this has to do with swing path, which I will further explore later in this article.
Remember the numbers above are just for league average. What about for players who rely on their speed to produce offense? Perhaps they see a greater benefit to hitting the ball on the ground because they put more pressure on the infield defense to get the ball to first base. To explore this possibility, I looked at all of the qualifying hitters from the 2013 season and graded their speed using FanGraphs’ version of Bill James’ Speed Score. In the absence of reliable raw speed measurements, this gives a decent enough approximation of players’ absolute speed without too much added noise coming from baserunning intelligence.
Selection bias rears its head here, since players who are accruing enough plate appearances to qualify must be playing well enough to stay in the lineup. If the groundball approach is a viable strategy on any level, however, we should be able to find big-league regulars who succeed in this manner. We shouldn’t care as much about the approaches of players who can’t stick in a starting lineup. Here are the offensive numbers of these 140 hitters according to batted ball type:
|Batted ball outcomes for 2013 qualified hitters|
|Batted Ball Type||OBP||wOBA|
We still see similar relationships to the league-average numbers above, with a slight uptick in production across the board. Now, see the relationship in this graph between speed and wOBA on ground balls:
Though not a perfect correlation, the expected advantage of fast hitters versus slow hitters on ground balls is present. How about just the fastest 25 hitters? How do they do on all batted ball types?
|Batted ball outcomes for 25 qualified hitters with highest Speed Scores|
|Batted Ball Type||OBP||wOBA|
There is a much bigger difference in on-base percentage now between ground balls and fly balls, but the weighted on-base average rankings have not changed. Fast players still do better hitting the ball in the air than they do on the ground. Faster players also get a slight wOBA bump on line drives, likely by being able to take more doubles and triples than the average hitter.
Let’s look at the batted-ball data from another perspective. How about power-handicapped players? If a hitter has a low likelihood of hitting the ball out of the ballpark, perhaps he would be better served trying to keep the ball on the ground. Here are the same numbers for the 25 worst home run hitters in terms of homers per fly ball this past season. The rates of HR/FB range from 1.50 percent to 6.80 percent.
|Batted ball outcomes for 25 qualified hitters with lowest HR/FB ratio|
|Batted Ball Type||OBP||wOBA|
Now we have some guys in the ballpark we were looking for. Looking at on-base percentage, there is a pretty wide gap, while in weighted on-base average it is virtually a coin-flip. Line-drive outcomes are not greatly affected by this group’s lack of power, still maintaining a .646 wOBA when the ball is hit square. Even more extreme, here are the same outcomes for the 10 worst power hitters:
|Batted ball outcomes for 10 hitters with lowest HR/FB ratio|
|Batted Ball Type||OBP||wOBA|
Awesome! The best HR/FB rate in this group belongs to Alexei Ramirez, at 3.60 percent. As a group, they have a 27-point advantage for ground balls over fly balls. So, if a hitter expects to hit fewer than about four percent of his fly balls out of the park, he should probably give up and hit the ball on the ground, right? Not exactly.
While grounders overtook fly balls in this table, line drives still have a huge production edge over the other two. If we added liners into the flyball numbers to show all balls hit in the air, there would be no comparison. These hitters still should be trying to hit the ball up, just not to hit it over the fence. Line drives are where it’s at, period. To show this relationship between power and wOBA difference, check out this graph of the full sample of hitters:
The better the home run-per-fly ball rate, the greater advantage in flyball wOBA over the groundball variety (obviously). You can see the small group above the x-axis where ground balls result in a higher wOBA that we already explored, all of whom had below-average HR/FB rates. The x-intercept of the linear regression line falls in at 4.15 percent, giving an estimate for the breakeven point in this sample.
You may have noticed the decreasing line drive wOBA in the last two tables as the power numbers have gone down. Most of this effect is likely because line drives hit by weak players are not as hard to defend as those hit by strong players. I also suspect that a hitter who sells out for ground balls will not hit line drives as often, or be able to achieve the same results from those line drives due to quality of contact. I attempted to find further statistical proof of this, but this is probably where the sample of hitters limits the analysis. Even the most groundball-heavy regulars must hit the ball well enough and often enough to warrant their consistent playing time.
Now for the swing itself. First, some logic, then we will look at a few examples of how hitters create these results.
To consistently hit the ball flush, hitters need to swing on a slight upward path to match the ball on the same plane. Even line drives must be hit at a higher angle than level to the ground. Nothing really earth-shattering, since other research points to this already happening naturally.
Baseball Prospectus recently spotlighted a 2011 article by Matt Lentzner, in which he used PITCHf/x data to explore different aspects of pitch movement. He runs more numbers than I care to repeat here, showing a spectrum of pitches crossing the plate on a 4° decline for a shoulder-height Justin Verlander fastball down to 12° for a knee-high Adam Wainwright curveball. In an interesting graph from the piece, he showed the lowest whiff rate was present on pitches that crossed the plate on a 7° decline, suggesting the average major league hitter swings on something like a 7° incline. Ted Williams advocated this approach in his book The Science of Hitting many years ago:
This is the same reasoning that likely explains why flyball hitters do better against groundball pitchers than groundball hitters do. A groundball pitcher tends to throw pitches with more negative vertical movement. The correlation between pitcher groundball percentage and average vertical movement on fastballs (PITCHf/x data) is slight but appreciable (R-squared = .14). Flyball hitters tend to swing on a sharper uphill plane, matching the steeper downhill plane of groundball pitchers. This results in more true contact and driving through the ball on the same level, producing more hard line drives and well-hit fly balls.
So what does this look like? For a demonstration, let’s take a look at the leader in line-drive percentage among the 140 qualifying hitters from 2013, James Loney. Here he is hitting a double on a belt-high pitch into right-center field.
The swing starts with Loney’s back elbow dropping under his hands and toward the pitcher. The hands drop in on top of the elbow before leveling off through the contact zone, until after extension when the hands roll over. Because the shoulders tilt slightly, the barrel is working underneath the hands by the time it starts to pass his back hip. This brings the swing path on a slight uphill plane, generating natural loft without relying on backspin to make the ball carry.
For further illustration, I would also like to show the hitters on opposite ends of the graph of groundball and flyball wOBA differences. Chris Davis had a flyball wOBA 622 points higher than that on his ground balls this past season. Marco Scutaro was the most extreme of the few with a higher groundball wOBA, coming in at 100 points higher than his flyball result. First, let’s see Davis crushing a home run to center field in Fenway Park:
Here we see some moves similar to the swing above. The back elbow drops down below the hands, even more than Loney’s did. The hands work down on level with the elbow near the back hip before coming through the zone on an uphill plane. His shoulders are tilted toward the plate like Loney’s, perhaps more so, and his spine is much more angled back toward the catcher rather than relatively straight up. This positioning results in a very steep uppercut swing as he drives through the pitch. Even though some hitting coaches would refer to this as dropping the back shoulder, leading to too much length and uppercut, notice how tight the barrel stays to his body as his hands and elbow drop into plane. This keeps his swing short to the ball despite being in an uphill position with his upper body. Because he is so strong, he can gear virtually his entire body toward hitting the ball in the air and have tremendous success doing so.
For the other side of the spectrum, here is Marco Scutaro hitting a single into center field:
The first move is similar to the previous two swings, though the hands and elbow go together more. The hands push in front of the elbow earlier than the other two, taking some whip out but likely lending more control to his swing. The hands come through level or on a slight upward plane, dropping under his shoulders before driving forward. His shoulders also tilt toward the plate, bringing the barrel of the bat underneath the hands. On approach to contact, the bat is swinging on a slight upward plane that matches the line of the pitch.
Even for a guy who should not be trying to hit a fly ball, he still swings up through the pitch, just not to the same degree as a guy like Chris Davis.
A quick note about the “swing down to create backspin and lift” sentiment: stop that. It is okay for guys to think in terms like this to keep from getting long with the swing, but please realize that this does not optimize batted-ball distance in reality. Swinging down on the ball while the ball is also going down will more often result in hitting the ball down or missing it, unless one possesses unbelievable hand-eye coordination.
While backspin does enable a ball to fly further, taking a path down to the ball is nearly impossible to do with any kind of consistency. In a sport where hitting .300 is lauded, I do not think this person exists. Even if a hitter could make contact that same way every swing, it sacrifices too much energy from the bat-ball collision in favor of more backspin. Physicist Alan Nathan has a great research article posted on his website detailing the math behind hitting a home run based on experimental measured ball flight. In it, he summarizes the results pertinent to this common teaching axiom:
For a typical fastball, the batter should undercut the ball by 2.65 cm and swing upward at an angle 0.1594 rad.
That value in radians converts to a 9.13° uppercut swing, representing the maximized swing path for energy transfer and backspin using a typical major league hitter’s bat speed.
I do not believe this is common practice or knowledge in major league baseball, which is unfortunate. I have heard a lot of second-hand horror stories about the philosophies of many organizations in the game. Especially at the big league level, there is little evidence that a true groundball swing will lead to success. Line drives are the key to hitting, regardless of hitter attributes. Speed appears to have less of an impact on a hitter than what popular belief says. While speed may help boost a player’s batting average on balls in play, fast hitters do not have an automatic incentive to hit the ball on the ground, based on these results. They can turn doubles into triples rather than just outs into singles.
That said, there is some (but little) wiggle room for individualized approaches to swing path. To tell Chris Davis and Marco Scutaro to have the same swing would be ridiculous. There are data here that show hitters who are not strong enough to hit home runs consistently may benefit from hitting more ground balls than fly balls. Again though, line drives are still best for even the least physical hitters. A slight uphill path to the ball should be the goal of every batter in order to maximize efficient contact and batted-ball success, with even more uppercut swings reserved for those who can drive the ball out of the park. Every hitter is different and has different strengths, but the wiggle room lies more in how high in the air each hitter should aim.
The bottom line: leave swinging down to bad hitters at the amateur level, who have no chance at playing at the highest levels of the game. Otherwise, hit the damn ball in the air.