Mechanical Analysis: Matt Garza

[Note: I talked to a few people after this article was posted and stewed on a some things I mentioned and decided to change a few parts from the original post, regarding the benefits of a faster tempo and frontside mechanics. The ultimate conclusion and overall message are still the same, but I wanted to alert readers who read the original version that this version has some slight edits.]

One of the more intriguing trades this off-season centered around the Rays shipping Delmon Young (and others) to Minnesota for Matt Garza (and others). The fortunes of these two emerging talents will decide which club ultimately will win the trade. I’ll take a crack at Matt Garza today, and will tackle Delmon Young sometime in the future (sorry Delmon, but throw a bat at an umpire and lose your spot in the front of the line). I haven’t had the chance to scout Garza in person, so I can’t give a comprehensive scouting report based on just his video. I can, however, breakdown his mechanics and get a good impression of what kind of pitcher he is and cast some guesses as to what he will become. And away we go after the jump…
Here’s the quick look at Garza’s motion. I made it a thumbnail that you have to click to see in it’s own window.

Now I’ll go break it down piece by piece.


We’ll start with tempo, because that’s what I was taught to start with, and because it just seems logical.

Carlos Gomez (the former internet video scouting guru, now a scout–not the Twins outfielder) likes the aggressive arm stroke Matt Cain takes, and it’s a great example to illustrate my point, so I’ll put Cain’s arm stroke and Garza’s side-by-side. One standard way to measure tempo is frames of video from the top of the leg kick to the release of the ball. Cain is among the fastest, a model for power pitchers like Garza (you’ll see how much power later in the article), and Cain’s speed is 21 frames while Garza is at 28 frames. While tempo is by no means a be-all-end-all measure of effectiveness or mechanical efficiency, it certainly does help create arm speed, velocity, and snap on a breaking ball.

[Quick tech note: depending on your internet connection speed, theses two GIFs may not be synched (the yellow numbers should pop up at the same time). Just trust me, they are going the same speed and the the commentary below is still right. It should still make sense, you’ll just have to concentrate a little harder.]

If you watch these two again, you’ll notice I gave Garza a head start and synched (or tried) to release. When they both reach a common area (ball straight behind them without the shoulder stretching back yet), I put the frame it was for each to get there. Garza took 19 frames to get there and Cain took 11 frames. That’s Garza being 8 frames slower to get to that point, and he’s only 7 frames slower over the whole motion. So, the whole disparity in tempo occurs from the time the ball goes from out of the glove to behind him. Now, from the ball being behind him to release, Garza is actually faster than Cain, who is one of the fastest in the league. That explains how Garza creates his velocity (a quick finish), but he could create more by tweaking the beginning of his arm stroke.

Garza’s arm almost comes to a complete stop in the back, which is generally a red flag of possible future injury. This is because that near stop lends itself to the arm lagging to speed back up to the body, dragging the arm behind the body, and the body forcing the arm through release point, thus stressing the shoulder and losing velocity and command. Garza could combat this by either leaving the ball in the glove longer and going from glove to release point in a shorter time (sounds good) or slowing his body down to the tempo his arm takes (two wrongs don’t make a right).

It appears in frames 0-19 for Garza that he was taught to “tall-and-fall” at a young age and became comfortable with that, then at a later age, someone else taught him to explode to the plate and that’s when his velocity showed up. I’ll bet when his velocity jumped from high 80’s to mid 90’s, it was when his finish got more aggressive. The problem there (not really problem, more inefficiency) is that Garza didn’t make the beginning of his motion as quick as the finish and he looks to be leaving more velocity on the table.

To clarify, the speed of the arm in the takeaway from the glove is somewhat irrelevant–it is a negative action in relation to the throwing of the ball. Saying a faster takeaway automatically equals a faster arm (and more velocity) is like saying going backwards before a 100 meter sprint would net a better time. But in Garza’s case, he slows down enough in the back that it’s a concern to me and thus getting him to a consistent speed like that of the rest of his arm stroke, could really help.

As for Cain, while he does a great job being aggressive here and I wouldn’t change it (don’t like fiddling unless I must, and “slow down” isn’t a lesson I like teaching), it’s possible that he’s so fast in his takeaway (remember, that’s the glove straight back to the farthest point behind you, before the arm goes forward) that the speed he’s creating in a “negative” move (in relation to the “positive” move of throwing at the plate) could do more harm than good given that the arm has to harness that negative energy away from the plate and turn it around. It’s a better problem to have than Garza’s, but I bet if you slow down Cain’s arm a tick but keep the smooth motion, he wouldn’t lost velocity and might have a slightly lower risk of injury. But, again, I wouldn’t recommend for Cain to do that.

So, to tie this section together, Garza can improve his tempo by fixing the hesitation in the back of his arm action in one of two ways: 1) leave the ball in his glove longer, or 2) speed his body up so his arm can’t afford to linger in the back. I would generally lean towards #2 because it’s easier to speed the whole motion up a tick and take out the pause than to try to “slow down” how quick the ball comes out of the glove and risk an over-correction where he whips his arm around faster than Cain does. This, of course, is all assuming the Rays’ staff feels Garza is open to a change and can feel he has the mentality and athleticism to handle such a tweak without altering what he does right. If they don’t feel he can succeed with what I’m suggesting, there are smaller things a pitching coach can tweak that to fix part of the problem.


That last section got pretty long, I’ll try to get to the point here. This should be easy though, the weight transfer is where Garza makes his living, as stated above.

We’ll start with the positive; Garza really gets after it with his lower half late in his delivery and creates a lot of torque. Look at frames 13 thru 15 when the numbers pop up in the lower left hand corner. In frame 13, his front hip is pointing at the plate and his arm cocked back (known as loaded scapula) using the elastic qualities of the shoulder as a slingshot–this is where you create velocity. Two frames later, less than 1/10th of a second, the arm is taking off towards the plate and his hips are turned almost 90 degrees. You might think it’s a red flag that his body is already turning home while his arm looks to be behind. This is good, the arm isn’t lagging, it’s loading up–it comes to release faster than his body does from that point forward. This is a really aggressive move and gives Garza the velocity to be what he is–a potential frontline starter.

One of the things to look for in the weight transfer is the front foot, once extended toward the ground makes a quick “step-over” move into the stride toward homeplate. Imagine a can of beans sitting about 5 feet in front of the rubber and when the pitcher strides that far, he has to pick his foot up a little at the end and stride just a little farther, to step over it–that’s an outward sign of a lot of power about to be unleashed. If that makes any sense, this is what you’re looking for, in conjunction with an aggressive hip turn. All that to say, Garza’s step-over isn’t much because it happens in frames 10-12 (of this clip) and he starts exploding in frames 13-15. There’s a little velocity he’s leaving on the table.

There’s one more thing in the weight transfer that raises some red flags for me. See how in the first few frames his leg is reaching out straight to the plate, without any bend? It’s good to want to get a longer stride and make that hip move more aggressive, but when the leg loses its bend, the knee gets locked, the leg becomes stiff and it affects your command. In this particular clip, Garza is delivering a letter-high fastball. The catcher is set-up high, so he hit his spot, but if this happens with a low spot and just misses by a bit, it’s going 95 mph the other way. That’s just a glimpse of the sort of thing that can cause command problems, usually up in the zone. In other pitches from other games I’ve looked at, Garza does this when appearing to aim low. Having superior velocity to get away with this mistake at times almost encourages young fireballers to keep doing it, mainly because it looks cool and will get you on Sportscenter. It will also get you off the 40-man and onto a non-roster invitee list, just ask Ramon Ortiz.


As a general rule-of-thumb, the more “elbowy” the arm action is the better; Garza is pretty “elbowy” in the back, an indicator of torque, which causes velocity. Garza does a good enough job out of the glove keeping his elbow flexed (fully-extended=bad, due to flexing muscles being weaker than relaxed ones). He also does a good job of letting his elbow “pick-up” the ball, which is another good thing, rather than “flipping” like some top pitchers do. The flip is essentially the arm pointing down and to transfer it to pointing up, there’s an abrupt flipping motion to get it there, rather than a full-range-of-motion, smooth sweep back. The unneeded torque and stress the flip creates it the equivalent of a shotput ball in the same position with the speed the arm performs the flip. Just ugly all around.

To finish, the most important part of the arm action for creating velocity, as described above, is “loading the scapula,” which basically means when your arm is fully extended behind your head and about to start coming over the top to deliver the ball to the plate, the shoulder joint allows the arm to stretch farther back behind you than you could move it on your own–like someone is putting all their weight toward pushing your shoulder joint from the front. Imagine the force it would take to move the joint back a noticeable amount, and now consider that his arm is moving so fast that Garza is making his own arm do that, and every millimeter farther back he goes, creates a bigger “slingshot” reaction forward and more velocity. It also creates more of an injury risk but it generally considered a good gamble to take your chances in that part of the motion.


It was tough to find footage from the homeplate view of his frontside that is viewable in this article. You can refer to some of the above video taken from behind him to get an idea of what the glove and arm are doing at and after release.

One issue is that Garza’s propensity to have a stiff front leg makes his body want to lean back after he’s released the ball. This makes sense, if the front leg is locked and you can’t follow-through over it, you’ll fall back. And a bad side effect of this is that after following-through, Garza frequently recoils his arm back where it came from like a whip. This is a big no-no and can create arm problems for the simple reason that you aren’t completely decelerating your arm, and on top of that are yanking back in another direction. The tiny parts in your elbow and shoulder could get stressed much more quickly and nobody wants that.

One more thing late in his motion that Garza does well is clearing out with his front side. For those of you who read Carlos Gomez’s stuff in the past, here’s one spot where we disagree philosophically. I like Roger Clemens as an example of my preference, and PED allegations aside, he stayed healthy doing that so I’ll stick with what he does. Basically, the ideal motion (in my opinion) is to extend your glove arm (just the elbow is fine, you can also stick out the lower arm for a “swim move”) and help the throwing arm accelerate by pulling the lead elbow laterally (not down, everyone agrees down is bad). And the key here is to not overdo it, because that causes you to fly open, and Garza has a tendency to do that at times and looks to over-rotate in some clips I’ve seen. Toning down that clear-out move should improve Garza’s command and lower his injury risk, but it creates some of that elite velocity he has, so I would not change too much.


Like I said at the top, I can’t really give you a traditional 20-80 look at Garza’s stuff on the scouting scale, but I think these two videos may assuage any fears about him coming short in those departments (click to watch in a new window).

Wowza. Might as well call him Captain Hook. The funny thing is that given the action on that pitch, I think that was his two-plane, low-80’s slider. He also throws a mid 70’s curve that has deeper but softer break. Those two pitches are similar to Roy Oswalt’s hard and soft variations of a curveball. On this last video (click to watch in a new window), check the mph in the lower right for both pitches, then on the 2nd pitch notice the location, that Garza is in the stretch, and the look on Travis Hafner’s face. Poor Pronk.

I haven’t figured out a snappy way to show the stats on this page, so here are two links to all the (non-subscriber) Matt Garza stats you’ll want: The Baseball Cube and FanGraphs

The things that jump out at me are his walk rates throughout the minors were very good, below 2.5 BB/9 IP at almost every stop, and his K rate was great, above 10 K/9 IP at every stop until AAA. These show me that these command issues were there in the minors and the margin for error at each level is almost perfectly shown in the progression of his stats. If the above mechanical adjustments are made and the performance improves like I think it will, I don’t see why his big league ratios can’t sit right between his career minor and major league ratios (roughly 8.5 K/9 and 3.0 BB/9). He has flyball tendencies but comes in close to league-average in this regard when he’s pitching well and keeping the ball down. His profile is very similar to a right-handed version of Erik Bedard.

Baseball Prospectus’ PECOTA projection system is the best in the business and seems to think (subscription required), reasonably, given that the program only has stats to go on, that Garza has some upside (comparables include John Smoltz, Don Larsen, and Chris Carpenter), but ultimately won’t completely “figure it out,” (most of the 17 other top 20 comparables inspire a gag reflex) saying he’ll be more of a league-average #3 starter; which isn’t bad at all.


Matt Garza is an impressive young pitcher with a realistic ceiling of, with a few small tweaks, a #2 starter, supported by flashes of some of the best stuff in baseball when he’s on. There have been some whispers about his makeup and coachability, but thus far in Tampa none of these have appeared to be a problem. If those things continue to not be a problem and he makes the slight tweaks mentioned above to shore up the command and consistency, I don’t see why he can’t be a legitimate #2 starter in the big leagues as soon as 2009. Do I think the Rays got the better end of the deal? Check back here later for the Delmon Young breakdown.

You may also like