Breaking Down the Swing: Best Hitters of 2012 Part III

This is the final article detailing my mechanical study of the best 50 Major League hitters from last season.  After this I will begin to apply this information to the vast library of amateur hitter video that I have acquired in the last few years.  Feels good to close the book on this monster.  For this portion I will focus on the moves of the lower half, again analyzing these athletes to create a baseline for examining less experienced hitters.  To get a full explanation of my methods and previous analyses, feel free to check out Part I and Part II of this study.

Looking at the lower half is why I started this study in the first place.  I looked at hitters empirically, and it seemed as though the best hitters all created the same kinds of ground angles.  I wanted to see if that simple observation accounted for any of the variance in hitting ability in Major League Baseball, and by extension, amateur players entering the draft.  Evidently, this first idea has spiraled into a crazy world of numbers and spreadsheets that has expanded countless times into the work that I am now presenting.  I only include information that I have gleaned from these hitters’ swings, but will continue to explain how I understand these processes and teach them to players.

With the hitters we train, the two main things we talk about with the hips are sequence and direction.  In any rotational sport, you want the biggest muscles in the body driving the turn, which of course are the glutes (and the core in general).  When the glutes engage properly in the swing, the sequence and direction of the hip rotation are most efficient.  The hips fire first with the knee and foot turning as the hips bring them through, and the direction of the turn is more like a straight line directly toward the pitcher, relatively.  If the rotation is initiated by the back knee or by turning the foot (“squishing the bug”), the glutes disengage and the back leg gets disconnected from the ground.  This leads to a longer path for the back hip turn with more movement toward the plate before driving toward the pitcher, thus diminishing the strength and quickness of the lower half.  All the weight gets onto the ball of the foot instead of on the heel where the body is strongest.  The weight will transfer to the front of the foot or even completely off the backside as the swing approaches contact, but the idea is to initiate rotation while still firmly connected to the ground.  When looking at the swing mechanically, the measurements I made help to paint an objective picture of how this movement works.  To help illustrate this a bit, here is a demonstration, courtesy of Buster Posey:

Notice how the first move is the core turning slightly, then the knee making an inside move toward the pitcher, followed by the back heel coming over the shoelaces toward the middle of the field.  If he were to turn on the ball of his foot, you would see the heel spin around behind the toes before coming forward.  Also notice how he has both heels on the ground for a few frames, and it almost rolls sideways to come through rather than spinning.

The aspects of the swing that I chose to address here are hip slide, hip rotation, back leg angle and ground angle, front leg angle and ground angle, and posture angle.  For the easiest explanation of the angles measured, take a look at the picture below:

Picture 1

The orange arc shows how I measure posture, based on a line drawn from the middle of the head to the middle of the core at the belt.  The blue arcs are the front and back leg angles, and the yellow arcs show the ground angles of each leg.  Hip slide and hip rotation were measured by looking at the middle of the belt to show the core position, and the belt buckle to show how the hips rotated through each part of the swing.

For the positions of the swing, I used similar delineations as in Part II, just with a slanted eye toward the lower half, “Stance” being the first.  “Gather” is the point where the weight shifted back the most and the hands and elbows moved back, if either loaded at all.  “Stride” is the point halfway through the step or move forward, coinciding with the point right before the back elbow attacked.  “Toe Touch” is when the front foot first touches the ground, and “Heel Plant” describes when the front foot is planted firmly on the heel as rotation begins.  For some parts, I also use “Hands Commit” as the point when the hands begin to attack after the elbow has started.  “Heel Release” is another position unique to some parts of this analysis, where the back heel comes off the ground and the foot begins to turn over after the hips have begun to rotate.  Previously I called this “Foot Rotation”, but I like the new terminology better for multiple reasons.  “Contact” is self-explanatory, and “Extension” is the point when the hands are farthest out in front of the body after the swing.

One final note about the Toe Touch position for a few hitters:  because some hitters reach out their front foot as they gather, rather than as they move their hips forward, I had to fudge their position labels slightly.  Instead of the first instant their front foot touched the ground, I assigned Toe Touch to the point in the swing right before the front heel started to come down.  In essence, these hitters “stride” with their front foot already on the ground, which made the measurements of the swing much more similar to the other hitters.  This only applied to two players, Aaron Hill and David Wright, but I felt it a worthwhile distinction.

One last disclaimer:  there are many big league hitters who do not have efficient swings, especially regarding the lower half.  I truly believe the hands and upper body are what separate the best and worst hitters, and there is far more variance in the legs.  As a good rule of thumb, I like to think you hit .300 with great hands, and you hit 30 home runs with great legs.  Not every move has to be efficient for someone to be a great hitter, but the more efficient player has a better chance to be a productive hitter.  That said, some guys do just have bigger motors and do not need to be as efficient.  I will explore this more in the future amateur hitter study.


First up, let’s analyze how much hip slide each hitter has.  Hip slide simply refers to how much lateral movement the core of the body undergoes during the stages of the swing.  If a hitter has too much lateral movement between toe touch and contact, it can lead to sliding off his backside and not driving his hips efficiently into the swing.  Ideally, a hitter lands in a balanced 50-50 position and goes right into rotation.  Here are the numbers for Toe Touch to Heel Plant and Heel Plant to Contact, presented as a percentage of the batters’ heights at Heel Plant:

Toe Touch to Heel Plant
Average 4.78%
Std Dev 2.29%
High 14.71%
Low 1.49%
Heel Plant to Contact
Average 0.41%
Std Dev 2.01%
High 4.85%
Low -3.69%

Up to Heel Plant, there is a slight move forward for all 50 hitters, but relatively small and with a pretty narrow band of measurements.  The 14.71% belongs to Chase Headley, which is more than 5 percentage points more than the next highest number from Austin Jackson.  To get a visual for what this looks like, here is Headley’s move from Toe Touch to Heel Plant:

From Heel Plant to Contact there is virtually no slide from any of these hitters with an even more narrow spread in the numbers.  This certainly gives weight to the idea that the hips should be in place to deliver the swing once that front heel gets down, and perhaps even sooner.  I would suspect that we would see a higher level of consistency from MLB hitters who have the smallest lateral movements, but that is not the goal of this study.

For those who are interested in hip movement through the stride, here are the numbers from Gather to Toe Touch:

Gather to Toe Touch
Average 11.40%
Std Dev 4.12%
High 22.22%
Low 3.24%

From this data, as well as my work from Part I, it is apparent that stride length and movement forward during the stride are much too varied to say one way is better than another.  That is not to say a hitter cannot improve by striding differently, but there is certainly no one way every hitter needs to stride.


The next aspect of the swing we will discuss is the rotation of the hips.  I measured the movement of the belt buckle relative to the middle of the body to separate rotation from lateral movement of the core.  The main thing I wanted to quantify is how much rotation happens before the back heel comes off the ground and over the toes.  I also looked into how much these hitters loaded their hips during the gather, as well as how much rotation happened during the stride before the front foot touched down.  Below are the numbers from Stance to Gather:

Stance to Gather
Average -1.06%
Std Dev 0.97%
High 1.19%
Low -3.63%

From Stance to Gather there is a slight turn back of the hips, although nothing crazy at -1.06%. Remember these numbers are a percentage of batter height at Heel Plant.  Assuming a batter’s height to be around 6 feet, that puts this backward hip turn at less than an inch.  I point this out mainly because I have heard hitting coaches talk about how you have to rotate backward before you rotate forward.  The greatest rotation came in at 3.63%, which is only 2.6 inches for a hitter standing 6 feet tall during his swing.  To further de-emphasize how much load these hitters have, here is the pure lateral movement during the same period:

Stance to Gather
Average -1.89%
Std Dev 3.31%
High 3.44%
Low -9.65%

The average movement as these hitters gather is still only 1.89% toward the catcher.  More variability here for sure, but even the most dramatic move in the group is only 9.65% back.  You can do the math.  As with stride length, I am not convinced that the lateral movement backward matters much, but I do think it is important to note the relative lack of rotational movement back during the gather.  Now for the hip rotation during the stride and before the back heel comes over.

Gather to Toe Touch
Average -0.08%
Std Dev 1.23%
High 2.27%
Low -3.42%
Toe Touch to Heel Release
Average 3.70%
Std Dev 2.13%
High 10.32%
Low 0.00%

Next to no rotation happens between the Gather and Toe Touch positions of the swing, constituting the entirety of the stride.  This will likely be an important distinction in analyzing the amateur hitters’ swings, due to the tendency for less gifted hitters to begin turning before the foot is on the ground.  As for Toe Touch to Heel Release, every hitter was in the positive except for Shin-Soo Choo.  His 0.00% simply means his back heel began to release in the same frame that his front toe touched the ground.  For the other 49 hitters, they all had varying amounts of rotation between the front foot touching and the back heel coming over/around, but the magnitude is not important for the Major League portion of this study.


Let’s move further down the body and focus specifically on the angles created by the back leg.  I used to think that hitters really had to drive the knee forward to create lateral force and propel the rest of the swing toward the pitcher.  The earlier they did that, the more momentum they got from their legs and the harder they could drive the ball.  Referring back to the diagram above, this would mean that we would see the angle of the back leg start or very quickly decrease to below 180°.  Here is what the data for back leg angle at each stage of the swing shows:

Average 184.8
Std Dev 15.4
High 207
Low 151
Average 190.7
Std Dev 14.1
High 210
Low 164
Average 192.5
Std Dev 10.5
High 215
Low 172
Toe Touch
Average 186.8
Std Dev 10.5
High 207
Low 164
Heel Plant
Average 166.8
Std Dev 14.3
High 198
Low 134
Average 114.7
Std Dev 7.8
High 132
Low 101

The greatest angle on average is actually during the Stride phase of the swing.  All the way up through Toe Touch, the angle is actually greater than 180°.  However, there is a pretty wide range of values for all 6 of these positions.  A closer look at how the angle changes is warranted, so we can figure out a more easily understood pattern.  Let’s first look at the change from Stance to Gather, and Gather to Stride:

Stance to Gather
Average 6.9
Std Dev 7.4
High 37
Low -3
Gather to Stride
Average 2.2
Std Dev 8.2
High 17
Low -17

From the Stance to the Gather, only 5 hitters’ back leg angles decreased.  Of those 5, 4 of their angles increased in the Stride phase.  So, most of these hitters’ angles increased during the Gather, but a few increased a bit later.  Ben Zobrist had the only swing in the sample whose back leg angle did not increase after Stance, though the angle he started at was rather large at 203°.  It happens at slightly different points, but this is a pretty overwhelming case for hitters not trying to pinch the knee in as they stride.  Let’s look at a hitter whose back leg works close to the average from Stance to Gather, Josh Willingham:

And the hitter whose back leg angle increases the most during the same stretch, Torii Hunter:

Both were taken through the Gather stage and almost to Toe Touch.  Pay close attention to how the back knee almost looks like it turns back just as they start to move forward into the rest of the swing.  Because each hitter gets into this move at slightly different times, I will simply refer to the point where the angle is largest as “Max”.  For all but Zobrist, that ends up being the larger of the two positions, Gather and Stride.

If you think about how the legs work best in any other discipline, this makes a lot of sense.  When doing squats in a gym, your knees stay over your toes and sometimes slightly turned out, depending on your bone structure.  This is the same sort of leg base taught in karate—often called the “horse stance”—useful for allowing short, quick firing of the hips to begin a strike.  Might be useful in hitting to follow similar instruction as a field of study that has been around for thousands of years…

It must be right; look at the bodies behind him!  This position allows the hitter, or Bruce Lee, to initiate a strong move with the lower half by firing the glutes, as mentioned above.  The heels are connected to the ground, and the knee only comes through when the hip turns far enough to bring it through.  Having the back knee start more over the back foot puts the hitter pretty close to the same position.

Going back to the numbers, let’s now look at the change in the back leg angle from Max to Contact:

Max to Contact
Average -80.2
Std Dev 11.7
High -50
Low -101

A fairly large range shows up here.  Perhaps the best way of measuring how deep into the legs each batter works is simply to look at the position at contact from above.  Too deep may hinder hip rotation, and not deep enough could mean not using the ground to drive the swing.  Again, let’s save the details for the amateur study.


Next up are the ground angles created by the back leg.  In light of what I now know about lower half mechanics, I do not think that ground angles provide more information than what we already have here.  However, I feel obligated to post them since it motivated me to start this project in the first place.  I include these for future reference or if you would like to draw your own conclusions from them.  Here are the simple positioning numbers:

Average 77.5
Std Dev 9.9
High 94
Low 50
Average -30.9
Std Dev 8.3
High -15
Low -49
Toe Touch
Average 67.9
Std Dev 8.1
High 84
Low 48
Heel Plant
Average 54.5
Std Dev 9.6
High 75
Low 33
Heel Release
Average 45.2
Std Dev 9.2
High 65
Low 23
Average 23.6
Std Dev 6.2
High 37
Low 12

Each position carries a wide range of numbers.  Perhaps something can be gleaned from the more narrow range at Contact, but I am not too optimistic, since the range is also the closest to 0°.  One especially interesting piece of data shows up when looking at the ground angles at Hands Commit and Heel Release to see the timing of the two moves:

Hands Commit to Heel Release
Average -9.5
Std Dev 7.7
High 17
Low -23

Only 2 batters in the entire sample committed their hands after Heel Release.  This means that the hands started to swing while the back foot had not yet turned over for nearly every hitter in this study.  Something to explore further for sure, especially since it helps describe the link between the upper and lower halves.

The last thing I would like to present regarding ground angles is the relationship between the back leg and front leg at Toe Touch and Heel Plant, calculated as front leg angle subtracted from the back.

Toe Touch
Average 11.16
Std Dev 10.49
High 42
Low -11
Heel Plant
Average -13.28
Std Dev 12.30
High 15
Low -35

At Toe Touch we see the back leg angle is greater than the front for the majority of the hitters.  This touches on the previous theme of the knee staying over the back foot so the hip can drive the swing.  At Heel Plant the opposite is true, though with a decent number of exceptions in both examples.  Also, with the relatively large standard deviations present, this is more so just an interesting observation rather than anything really conclusive.


The last piece of the lower half I analyzed is how the front leg firms up during the swing.  In my work with hitters, occasionally we focus on making sure the front leg does not stiffen too early in the swing.  This can cause the hips to get too uphill and inefficient, and sometimes make the front shoulder pull off early.  If it never firms up, the body may continue drifting forward causing a hip slide and a power leak in the lower half.  Let’s see what the numbers bear out:

Toe Touch
Average 173.8
Std Dev 11.7
High 197
Low 140
Heal Plant
Average 189.5
Std Dev 12.6
High 216
Low 164
Average 182.1
Std Dev 8.8
High 207
Low 165
Average 178.1
Std Dev 8.7
High 206
Low 160

This is a bit tricky since there are two ways that hitters can register an angle under 180°, depending on which way the knee is facing.  For example, here are two hitters whose front leg measures out to 166°, Justin Ruggiano at Toe Touch and Andre Ethier at Extension.

Picture 1 Picture 2

Since Ruggiano’s knee is slightly turned back, his measurement comes out to be the same as Ethier’s slightly hyper-extended front leg.  Going back to the numbers in the table, it does appear as though the averages do correspond to firming up just after contact.  To get a clearer picture, here are the changes in front leg angle between the same 4 positions:

Toe Touch to Heel Plant
Average 15.7
Std Dev 8.0
High 34
Low 3
Heel Plant to Contact
Average -8.4
Std Dev 16.7
High 23
Low -46
Contact to Extension
Average -3.0
Std Dev 9.7
High 35
Low -22

Nobody lands firm, as everyone in the sample has an increase in front leg angle going into Heel Plant.  Heel Plant to Contact has a huge spread in values.  This is due to having some guys that firm up with the knee still turned slightly back (positive change), while others have the knee facing forward at Heel Plant (negative change).  Still others rotate so that it faces back at Toe Touch and forward at Extension, throwing the calculations off even if the bend in the knee never changed.  This is a little ambiguous, and the same issue pops up in Contact to Extension, though in a lesser manner.  This may not be something that can be numerically analyzed from only a side view, but I will still look into it with the amateur hitters.  Perhaps I will have to use discretion in measuring the angle based on which way the knee is facing.


The last piece of the swing I measured was the posture at each of the phases we have discussed.  However, in looking at multiple swings from the same hitters, I am convinced that this shows the greatest variability of all the things I would like to record.  To properly analyze posture changes, it would only make sense if the pitch offerings were standardized.  If a hitter is fooled on an off-speed pitch, he will tend to lean forward more, even if he still maintains a balanced position.  If a pitch is up, it may cause him to lean back more get on plane with the ball.  There are too many outside variables that change this part of the swing for me to feel comfortable attaching a number to it for players based on one swing each.  That said, it can be a valuable teaching point for coaches.

On that note, I will share a couple ideas that we use to help influence hip action and hand path by focusing on posture.  Often hitters will slide their hips because their head stays too far behind the body, or their hands will create too much of a loop because they are so uphill.  We talk about trying to keep the head out in front of the body’s center of mass (belly button for simplicity) and feel the hips loaded up behind the head.  Many times a player will hear “stay back” and try to keep their head in one place.  Then they end up lunging for the ball anyway because the hips are not stable underneath the upper body, or they end up slicing balls to the opposite field because their bat path is too much of an uppercut.  We have guys focus on staying back by keeping the hips loaded up behind them, creating a “K” shape with the body, like the front side is crunched down slightly while the back side is more straight:

Trout K

If you read Part I of my study, you will remember how everyone’s head moved forward in some capacity, and some moved down as well.  Every swing works a little differently, so there are varying levels of this feeling for different hitters, but it is an important point to address.

So in this final installment for the Major League portion of my study, we addressed many aspects of the lower half that play a part in the efficiency of the swing.  Each of these topics will be covered again once I get around to analyzing my amateur swing videos, with the goal of determining what use objective mechanical analysis can have in projecting unproven talent.  I sense that the results will be very enlightening regardless of the outcome.  I hope you have enjoyed this series; it has helped me organize my plans for my next big project while also giving me a forum to throw out some key hitting ideas that are misunderstood or not usually discussed.  Please ask questions about the data and feel free to challenge any of my thoughts you disagree with.  I always have an open mind about baseball.  It is far too interesting of a sport to disregard new ways of understanding it.


*Edited by my lovely wife, Nicole, who is willing to proofread a 4000-word article she has little interest in reading.

Print This Post

Dan is Fangraphs Lead Prospect Analyst, living in New York City. He played baseball for four years at Franklin & Marshall College before attending medical school. Follow him on Facebook and Twitter @DWFarnsworth.

Sort by:   newest | oldest | most voted

You owe the lady big time ‘cake’. Great work.