The Jump-Step and Other Unrepeatable Deliveries

Jordan Walden has a jump step in his delivery. Jordan Walden has control problems. Does one cause the other?

He’s not the only one who has this tendency. Ask around and you might hear about Javy Guerra and Trevor Cahill. With the sample so small, does it mean much? What about other unrepeatable deliveries, like the ones from Chicago relievers Rafael Dolis and Carlos Marmol? Is there something different about the jump-step that sets it apart from other difficult deliveries?

Walden’s jump-step is well-documented. Jeff Sullivan has some great pictures here, but the most compelling shows Walden with both feet in the air. That’s quite a feat:

That jump has to wreak some havoc on his front footfall. Which would, in turn, seem to change the direction he’s pointing as he releases the ball. That’s a theory, anyway, and there’s no way that jumping makes repeating the entire delivery any easier. Walden has consistently walked more than four batters per nine innings during his career, and there might be a cause we can find in his delivery. His current walk rate may have had something to do with the fact that he was moved out of the closer’s role with the Angels.

Another California closer has struggled with his control and was relieved of his duties. Guerra doesn’t seem to employ the hop-step, though. Check out this nice multiple-exposure that Kirby Lee (US Presswire) shot. The back leg stays rooted:

No hop-step there. But ask Brandon Lennox, of TrueBlueLA, and he’ll tell you that the crow-hop does have something to do with Guerra’s control problems: “Control issues have been Guerra’s calling card since early in his career when the Dodgers removed a ‘crop-hop’ from his delivery. Ever since then, he has struggled to repeat his delivery, which causes him to be wild.” Removing the jump was probably a good idea — Guerra has had a 3.51 BB/9 in the major leagues after regularly walking more than five-per-nine in the minors; but that wasn’t the whole idea, since the change came well before the improvement.

So now we have two former/future closers — one who is struggling with his control with a jump-step, and the other who might be struggling with his control despite removing the jump-step from his delivery. Let’s wade into less black-and-white territory. Does Trevor Cahill have a jump-step in his delivery? Check out the video around 1:45. Does that back toe leave the ground? It’s very close.

Fittingly, Cahill’s walk rate (3.36/9) is only a tick above the league rate, which has been hovering around 3.2/9 recently. And if you think he’s got a jump-step, then you’d probably accuse Tim Lincecum — and perhaps Trevor Bauer — of the same. But if we did that, then we’d be opening up a whole new bag of worms — that is, considering the strong proponents of their deliveries. Then again, none of these guys has better-than-average control, and you can see that there’s a little jump in their deliveries (here’s a good one of Bauer), as they push out to get the largest stride they can manage.

We started with two closers in California, though, so let’s move to two closers in Chicago before we’re done. Carlos Marmol has terrible control, but he usually gets the strikeouts to make things work out for him. He has been failing at getting swinging strikes recently, though, and he was demoted from his role. Now Rafael Dolis is giving is giving it a shot. Dolis has terrible control, but he usually gets ground balls to make things work out for him. Can we see anything in their deliveries on which we can blame their control issues?

It’s a little crude, but here’s a random sampling of pitches from Dolis in his last start. Look how his left-to-right lean is different in some of the pictures. (The easiest way is to compare how much of his jersey you can see on the left side of the white line, or how far down on the left his head is.)

It’s not about about a jumpstep here, but it looks like he’s having difficulty repeating his delivery — at least to this lesser-trained eye. And since we can’t see the foot fall, it’s unclear if his feet are the problem. We do, though, track release points over larger samples. From Brooks Baseball, we find that Dolis’ horizontal release points have ranged a full foot, while his vertical ones have been as much as six inches different. Look at Marmol’s release points, and they’re over a foot diverse vertically and horizontally. That seems bad. Then again, Roy Halladay‘s release points seem to fall right in between these two control-challenged pitchers.

It’s clear that there’s no direct, linear relationship between a delivery and the control it produces. But it also seems clear that the footfall is an important moment in the process, a key to that relationship. After Kyle Drabek began his career with years of control problems, the Blue Jays spent this past offseason focusing on his footfall. The team drew a yellow box on his practice mound and told him to land in it every time. Sportsnet columnist Shi Davidi had the column and the picture that said it all:

After working ‘in the box,’ Drabek was upbeat:

“Really, everything has transformed. It also has allowed me to repeat my delivery on each pitch 80% or more, that’s what I’m looking for, to be able to repeat my delivery, that I’m not showing something different on certain pitches.”

And yet Drabek — with a walk rate more than five-per-nine — still is struggling with his control. Even though it’s a small sample, it goes to show that this is a tricky science and that the type of large-sample, statistically robust research that we’re accustomed to is lacking here. Do diverse release points correlate negatively with walk rate? Does precision in the footfall influence control? Those are just two questions, and yet they represent the difficulty in meshing bio-mechanical scouting knowledge with a statistical view of the game.

As we gain more publicly available data, though, we can begin to answer these questions. It does seem, at least anecdotally, that the jump-step is a bad idea — not so much because the pitcher is flying, but more because it leads to an uncertain footfall that can influence the ball’s direction. As much fun as it is to watch Walden leap at a batter, it might be detrimental to his game.




Print This Post



Graphs: Baseball, Roto, Beer, brats (OK, no graphs for that...yet), repeat. Follow him on Twitter @enosarris.


8 Responses to “The Jump-Step and Other Unrepeatable Deliveries”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. thatwasweird says:

    just for better review of walden’s entire motion – http://cdn1.sbnation.com/imported_assets/794721/WaldenMotion.gif.opt.gif

    Vote -1 Vote +1

  2. What is the rule on this? I can’t see anything in the MLB rules so I must be reading the wrong things.

    I assume a pitcher can’t take a couple of steps towards the batter before throwing, or run all the way up, but I don’t see anything in the rules that describes how it has to be.

    Vote -1 Vote +1

    • Eno Sarris says:

      From the comments on Sullivan’s piece:

      8.01a

      He shall not raise either foot from the ground, except that in his actual delivery of the ball to the batter, he may take one step backward, and one step forward with his free foot.

      – There is no indication how long the pitcher must keep his pivot foot on the rubber though nor in which ways he’s able to leave the rubber.

      Vote -1 Vote +1

  3. Terry says:

    I saw Guerra pitch in high school. They called him hopping Javy. His motion was ten times more extreme than Walden’s current motion. The other teams would yell “balk” on every pitch.

    Vote -1 Vote +1

  4. jw says:

    Nice read. Thanks.

    Vote -1 Vote +1

    • Eno Sarris says:

      You know, I write mostly because I like to write, but it is really nice to get positive feedback to balance out the negative, so I thought I would thank you for this simple comment. So, thanks!

      Vote -1 Vote +1

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>