Joba through the eyes of Pitch f/x


Before Joe Girardi managed to grab defeat from the jaws of victory with his bullpen moves, Joba Chamberlain pitched well in his first start of the year, allowing just one earned run in six innings. He threw only 88 pitches because he hadn’t gone beyond the 75 pitch plateau during Spring Training, and predictably worked off his fastball. Here’s the breakdown:

Pitch Selection

I’m guessing that the two seamers are just four seamers that Pitch f/x did a shotty shoddy job of identifying. The percentages (~69% fastballs, ~27% breaking balls, ~4% changeups) are right in line with what he’s done the past two seasons. Here’s the pitch trajectories, remember to click for a larger view:

Bird's Eye View

Fastballs and changeups go one way, breaking balls go the other. That’s typically how it works. Here’s the look in from first:

First Base View

It’s cool to see just how much Joba’s offspeed pitches drop compared to his heater. Seeing this, it doesn’t come as much of a surprise that he generates so many awkward looking swings and misses. Here’s what Jose Molina was seeing from behind the dish:

Catcher's View

I heart this view. It takes Joba’s fastball just under 0.39 seconds to go from his hand to the front of the plate, but it takes his slider about 0.45 seconds to make the same trip. Look at how similar the two pitches look until they’re about two-thirds of the way to the plate, when they split up and head in different directions. I’ll repeat what I said above: it doesn’t come as much of a surprise that he generates so many awkward looking swings and misses. Here’s a look at his release points:

Release Points

I’m going to start plotting these graphs on the same 2-foot by 2-foot grid that’s 18-inches from the center of home plate horizontally and 5-feet above the ground vertically. That way you can see where Joba’s pitches come from compared to say, Chin-Ming Wang’s or AJ Burnett’s. Joba was very consistent with his release point yesterday, fitting everything into a 9-inch tall by 10-inch wide box, except for that one extraneous fastball. He probably stood more towards the first base side of the rubber during that pitch, whether he realized it or not.

Much like Andy Pettitte on Friday, Joba released all his pitches from practically the same spot. Combine that with how similar his fastball and slider look alike until it’s too late, and … well … it doesn’t come as much of a surprise that he generates so many awkward looking swings and misses. Lastly, let’s take a quick look at his fastball velocity:

Fastball Velocity

The blue line is the velocity of the pitch out of his hand, the red line is the velocity of the same pitch when it crosses the plate. Joba sat 90-94 all day, touching 96, which isn’t the same 100 mph heat we’re used to seeing when he works out of the bullpen. He’s not going to throw that hard as a starter, he’s just not, but 90-94 touching a six is above average velocity and more than enough to be a dominant starter.

You can see that Joba’s fastball lost a little life as the game wore on, evidenced by the decrease in velocity at the plate. The ball was still coming out of his hand fine, but it just lost that little extra oomph as the game went on. No big deal, it’s actually pretty normal, but it’s just cool to see.

I’m not going to break down every start of the season like this, but I’ll check in every so often when someone puts together a great start or fires a stinker. Once we get a little more data for this season, I’ll take a look at some of the relievers, specifically Mariano Rivera and Edwar Ramirez because they’re “one trick ponies.” Oh, and if you’re every looking for one of these posts, search for the Pitch f/x tag.

Update (11:21am): A.D. asked to see the release point of the slider that Joba hung and John Buck appropriately deposited into the seats for a homer.It’s the big orange one:

Release Points with Buck Homer

Categories : Analysis


  1. Tripp says:

    His Change Up too comes in almost exactly the same as his slider and fastball two/thirds of the way in. Man, when he starts throwing a few more change ups consistently he’s really going to dominate.

  2. Somewhere, sometime, somehow, some day, I’ll actually be able to comprehend all of this =D

    (Well, I am better than I was, now I actually look at it…)

  3. Hobs says:

    How is it possible that the velocity of the ball leaving his hand increased as the game went on (trend line) but the velocity a the plate decreased?

    Change in release point? Funky wind/conditions? Shitty technology?

    I know you didn’t make the data, just plotted it…but that’s kind of strange.

  4. Adam says:

    I’m not sure I understand the release point data. Are the axes in feet?

  5. Slugger27 says:

    i think 90-94 is actually pretty encouraging, considering he was clearly not a fan of the rain. if its a sunny day with no precipitation, i bet its more like 93-96 for most of the game

    • Rich says:

      Especially since it was his first start of the season. Power pitches often don’t reach max velo for at least a month.

    • Joba sat 90-94 all day, touching 96, which isn’t the same 100 mph heat we’re used to seeing when he works out of the bullpen.

      Since 93 with movement is >>>>>>>>>>>>>>>>>>>> 99 with no movement, there is absolutely nothing in the world wrong with Joba eschewing going max effort to try to reach triple digits. I though Kyle Farnsworth and Armando Benitez taught us that lesson.

      We have fallen way too in love with the 99mph heat. It’s overrated.

  6. rafael says:

    Okay, a little off-topic, but…
    Perhaps I’m missing a joke or something, but I always thought the word was spelled “shoddy,” not with Ts…am I wrong? The curiosity has been building as I’ve seen it repeatedly over the last couple days (I know, I wonder about weird things).

  7. A.D. says:

    Do you guys know which of the slider release points was the hanger that Buck hit out?

  8. Hobs says:

    Woah, did you guys see Santana’s comments basically throwing Murphy under the bus for blowing yesterday’s game?

  9. FL Yank says:

    You forgot to mention how many awkward swings and misses Joba generates.

  10. Tom says:

    Hi, are there any physicists or people who have read that book on the physics of baseball that could explain better how a ball could lose speed at the plate compared to out of the hand over the course of the game? Could it be less spin on the ball? The wet conditions? Errors in data collection? Its interesting.

    • Hobs says:

      Yeah that’s what I was wondering too.

      If the data is accurate, it’s got to be that he as he was tiring the spin on his 4-seamer wasn’t as ‘tight’. This probably prevented the ball from ‘cutting’ through the air as smoothly, increasing it’s drag.

    • Well, I’m not a physicist nor have I read the book, but it seems fairly straightforward that every pitch naturally declines in velocity as it travels through the air and wind drag slows it down.

      Fastballs decline in speed less, since they have a tighter spin (and thus, generate less drag?), but every pitch will arrive in the catchers mitt at a slower speed than it left the pitcher’s hand.

      • Whozat says:

        Yes, but the magnitude of the drop on speed increased over the course of the game

        • A.D. says:

          My guess in this game, the weather definitely had something to do with that.

          • kunaldo says:

            yeah cutting through rain is sure to create some additional resistance…so that’s probably it

          • whozat says:

            Sure; but how? What was the mechanism? If is was harder to grip the ball, did that mean that Joba couldn’t generate as much backspin? Maybe his forearm tired, and that led to less backsping? Perhaps his whole body tired, and that meant he wasn’t driving as hard with his lower body, and it has nothing to do with backspin? Maybe he was being slightly more careful about his footing?

            Just to be clear, I am NOT concerned about Joba here…I’m just curious about what leads to this phenomenon in general, because I enjoy the understanding of things.

            • “I’m just curious about what leads to this phenomenon in general, because I enjoy the understanding of things.”

              And that’s why we like you.

            • A.D. says:

              Well in terms of the mechanism for the weather as it began to rain harder the ball is going to have more resistance for it to reach home plate. Give that we’re saying the velocity out of the hang is staying the same, that means the ball is facing more resistance, thus slowing it down, before reaching home plate. It yesterdays game the rain seems to be the very obvious culprit, especially as it was raining harder as Joba’s start went on, and then eventually letting up.

              If we see something similar in a clear day, in which there doesn’t appear to be outside factors then perhaps its something with follow through, or some other pitching mechanics/fatigue phenomenon.

              • Rain + humidity. Don’t forget that.

                More humid air causes balls to pick up more drag and thus, lose more speed inflight.

                This is why the thinner, less dense, less humid air at Coors allows balls to travel faster, and why power pitchers fare better there than breaking ball pitchers do. Fastballs are faster, breaking balls break less.

                (and mistake pitches are crushed harder, of course…)

        • Meh, like Mike said, it’s common. And again, consider the scale: that’s a 1mph drop in end velo over the course of those 88 pitches. No big deal.

          Virtually every power pitcher, even the best, will have a redline with a more pronounced drop than his blueline. Halladay, Johan, CC, everybody, their lines either stay the same or drop an MPH or two over the course of their start.

          • kunaldo says:

            tsjc, i dont think he’s a) questioning why a starter’s velocity would drop towards the end of the start or b) that the release velocity is different than the plate velocity…

            he’s simply wondering why a lower release velocity earlier in the game led to a higher plate velocity than later in the game(ex. 93 release -> 88 plate, then later in the game, 96 release -> 86 plate)…like is it less spin on the ball, the rain…what would make it disproportionately change like that

            • Ah, okay.

              Perhaps, as whozat suggested, as a pitcher tires he loses the ability to generate enough arm action to get backspin as tight as his earlier fastballs?

              I quibble with your example, though, because it’s a bit of a red herring that can lead us down the wrong path. “93 release -> 88 plate, then later in the game, 96 release -> 86 plate” isn’t really what’s happening here, IMO.

              More like: 94 release-88 plate (loss of 6mph), then later in the game 93 release-86 plate (loss of 7mph).

              Joba’s mean end velo dropped a single MPH, his mean start velo remained level. It’s not like his fastball was losing only 5mph during flight at the start of the game and then started losing 10mph at the end of the game (like you claimed in your proposed example). That’s a mischaracterization of the data.

              I’m also interested in learning more about the physics of it, but let’s not start creating bridges out of thin air for us to leap off of.

  11. You added a pie chart.

    Somewhere, DBHOF is rolling over in his grave.

  12. Andy In Sunny Daytona says:

    No Gazoo….:(

  13. Steve B-BALL says:

    Blah, Blah Blah
    I enjoy reading all the “stuff” on River Ave Blues, but give me a break , what’s all the B.S. about release points etc.
    If you pitch a good game you Win if you pitch a shitty game you lose!
    Joba was fine…. It was Girardi and his awful handling of his Bullpen….. COke o my gawd worthless, not to mention Cody ” what am I doing in the Majors” Ransome
    It’s no wonder he never made the “Show”… he has no idea how to hit a Curveball or anything other than a dead red fastball.
    And Centerfield…… oh no What a choice Gardner or Melki
    Gardner and his power stroke… can he hit it out of the infgield on a fly?? Not to mention his fielding has (to be kind) sucked, especially his throw to home!!! or should we call it his roll to home!
    Let’s hope that Tex is back at first tonite and Wang can give us 7 good innings!

  14. Adam says:

    What exactly defines a consistent release point? How many inches deviation is considered good versus bad?

    My b if you already explained this in another post.

    • whozat says:

      I don’t know that you can define it like that, necessarily. In an ideal case, all pitches are released from the same point so that they all look the same to the batter for as long as possible. And, also, consistent release point is one component of consistent mechanics, which leads to better command and control.

      But then you get a guy like David Cone, who threw all kinds of pitches from all kinds of different angles later in his career. But, basically, the tighter clustered the release points, the better.

      I’d love to see some analysis of Mo. I betcha his release points are all within a 2×2 in box, and we eventually realize that he is a pitching robot who was manufactured in a secret Panamanian lab in the early 70′s.

      • But then you get a guy like David Cone, who threw all kinds of pitches from all kinds of different angles later in his career. But, basically, the tighter clustered the release points, the better.

        This is a great point. A guy like Cone had so many release points it made it harder for a hitter to pick up the ball initially, changing the plane of his vision. That can be a great weapon.

        I’d guess though (without actually looking at Cone’s release point plots) that he probably still had between two to four clusters of release points, generally speaking. His “over the top” release points, his three-quarters release points, his sidearm release points, etc. It probably resembles three groupings of plots rather than a full, unbroken arc.

        Making him consistently inconsistent. Just my guess.

        • andrew says:

          Agreed. The idea behind that is if you can control where your arm is going, you can do a better job of controlling where the ball is going. So, it wasn’t as if Cone was flailing his arms about randomly. I’m guessing, as you said, that he chose a few different slots to throw from, and was able to hit those slots pretty consistently. If he didn’t hit those spots consistently, he probably wouldn’t have been an effective pitcher. Repeatability in a delivery is very important to being able to control your pitches.

  15. Pel says:

    Nice work, Mike.

    Also, the stray release point furthest to the right is the John Buck error-single-error.

  16. UWS says:

    Mike, when you have a chance, do you think you could do a comparison between CC’s f/x in his first and second start? I think it might be interesting to see how he adjusted.

  17. CB says:


    As you point out those two pitches identified as two seam fast ball could have been four seamers but I’m not sure that’s the most probable explanation.

    If you look at the pitch spin data those two pitches had a spin which would have placed them at the very extreme of the range at which Joba’s four seamer works at.

    This is particularly clear if you look at the release point corrected spin direction angle data. If they were truly four seamers they would have been at the extreme range of Joba’s typical four seam range and they would have also been comparatively low velocity for his four seamer(both pitches were around 89).

    I think it’s possible that two other pitches pitch f/x classified as 4 seamers were also two seamers.

    I bring this up because Joba had quite a bit of success last year when he used his two seamer when he was starting. I’d like to see him use it more often.

    BTW this pitch f/x section you guys have started is great. Congrats.

  18. Elle says:

    Just got into a fight with my boss about the immortal Joba-should-be-in-the-bullpen issue. Thanks for all your great posts about how stupid that is, so I could tell my boss he was being stupid. He still disagrees with me, but he has nothing to back it up.

  19. Just curious where do you guys find all this info? I assume you put the graphs together yourself, but where do you get the info?

Leave a Reply

You may use <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong> in your comment.

If this is your first time commenting on River Ave. Blues, please review the RAB Commenter Guidelines. Login for commenting features. Register for RAB.