R.A. Dickey’s Lost Velocity

Knuckleballers aren’t like other pitchers, or so the saying goes. Their pitches flutter like butterflies, they pitch at less than max effort, they don’t depend on velocity, and they can pitch into their fifties. All of these things seem true, and yet the more we know about knuckleballers the more they might actually be more like all the other pitchers out there. So when 38-year-old R.A. Dickey has lost some oomph on his seminal pitch, maybe it means something, just like it usually does for other pitchers.

First, the fluttering. That might be an optical illusion. R.A. Dickey doesn’t agree, but scientists and physicists that have studied the path of the knuckleball agree with Professor Alan Nathan that it behaves ‘generally’ like other pitches. The unpredictability might still be wonderful, but it’s physically impossible for the ball to break multiple inches in each direction. The ‘dancing’ might be our minds trying to figure it out.

Next the max effort thing. Dickey said he throws at about 75% of the effort he spent as a traditional pitcher. So that much may be true when it comes to rest and recovery.

But it takes effort to put velocity on the ball. And we’ve seen that Dickey’s ‘fast’ knuckleball is probably the best one in his arsenal. The pitcher himself told me in an interview last year that he’d been getting more swings and misses from the fast knuckler. Professor Nathan found that velocity and movement were negatively correlated for the knuckleball, and it seems that Dickey’s elite control for a knuckleballer might have been at least partially due to the velocity on his most important pitch.

Now Dickey’s three-year run of above-average control is in jeopardy. As you might have guessed, it’s not from a lack of movement. Professor Nathan plotted the movement this year and last year and found little change. But perhaps there is still something different about his knucklers this year. His velocity is down two miles per hour on average.

Dickey throws a few knuckleballs, and the error bars on his velocity readings are large. Let’s look at the distribution of his knuckleballs instead of a straight average. Using three-mile-per-hour buckets, I plotted all of his knuckleballs from four distinct periods below. The first is pre-breakout Dickey. The second is 2010-2011 Dickey, who was great but not elite. Then you have Cy Dickey and this year’s Dickey.

DickeyVelocity13c

Those purple bars might have a little too much in common with the blue bars of Dickey’s darker days in baseball. They certainly don’t fit with his Cy Young year, and in some ways, they don’t even look like the 2010-2011 version. In 2010-2011, 34% of the knuckleballs Dickey threw went faster than 78 mph. In 2012? 47%. And this year he’s been throwing those fast knucklers 16% of the time.

Dickey’s still getting whiffs this year. His 9.9% whiff rate puts him in the top 30 among qualified pitchers. But after three years spent walking less than six percent of the batters he saw, his walk rate is in the double digits this year. Maybe he has a little lost velocity to blame, since the pitch should have the least movement and was his go-to pitch in hitter’s counts last year. And, considering his back and neck problems this year, and his abdominal and hip and plantar fasciitis issues last year, it may be fair to ask if his body is robbing him of this velocity.

Is that too conventional a question to ask of a knuckleballer?

Here’s the distribution of Dickey’s knuckler velocity in half-mile buckets just for good measure:

DickeyVelo13last




Print This Post



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


53 Responses to “R.A. Dickey’s Lost Velocity”

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

    I hate seeing Dickey struggling like he has been but Im so glad Alderson sold high.

    Vote -1 Vote +1

    • dovif says:

      perhaps the Blue Jays can send him back to the Mets, they might need to include a prospect to offset the salary, but it might be better for them

      Vote -1 Vote +1

  2. suicide squeeze says:

    “the blue bars of Dickey’s darker days” is a great piece of alliteration.

    Vote -1 Vote +1

  3. FeslenR says:

    I wonder if the catcher makes any difference? Don’t forget Thole/Nickeas caught Dickey last season and the years before that. J.P. Arencibia isn’t known as a great defensive or game-calling catcher.

    Vote -1 Vote +1

    • nick says:

      But JP only caught one game. The rest of the time its been Blanco.

      Vote -1 Vote +1

    • tylersnotes says:

      whatever difference catcher would make, it wouldn’t be to velocity. It could make for a study to see if Dickey is getting more called balls than he did the last 2 years, which could be catcher defense but i’d bet the league change has as much to do with it.

      Vote -1 Vote +1

    • za says:

      Doesn’t hurt to give Thole a shot every few days. He’s more than holding his own at AAA with a 169 wRC+. Arencibia is a real butcher behind the plate. The only problem there is that Arencibia isn’t exactly a lefty masher and has extreme splits in 2013. Even taking that into account, it’s ridiculous that Thole is in AAA while Blanco is in the Majors. Adding Thole to the squad would make Toronto a better team; he’s a guy who could definitely put up a .350+ OBP from the catcher position and a .750 OPS, with appreciably better defense than JP.

      Vote -1 Vote +1

      • vivalajeter says:

        Sure, he *can* put up a .350 OBP with a .750 OPS. But it’s pretty unlikely, unless the ballpark really tickles his fancy. The only time he’s ever put up that OPS was in 59 plate appearances in 2009 (.752 OPS). The last two years, he was at .690 (2011) and .584 (2012). Unless he has a huge turnaround, I just don’t see any offensive potential out of him.

        Vote -1 Vote +1

      • Dave says:

        Even if Thole didn’t hit at all, you’d think Toronto would have the catcher most familiar with Dickey up with the team and playing, especially given how heavily they’re invested in RA. Thole would be replacing 41 year old Henry Blanco it seems like a pretty easy move to make.

        Vote -1 Vote +1

        • Canadian_Eh says:

          The Blue Jays asked R.A. which catcher he felt most comfortable with behind the plate and he requested Henry Blanco. Can’t argue with that logic

          Vote -1 Vote +1

        • Dave says:

          Didn’t know that makes more sense now obviously.

          Vote -1 Vote +1

  4. attgig says:

    would really have to think that the back/neck problems are causing him to not throw the fast knuckler as often as he did last year. He needs to take 15 days, and get his body healed, so that he can pitch like he did last year.

    +8 Vote -1 Vote +1

    • Jason B says:

      Agreed – take 15 days to try and get healthy (healthier?), it’s a small price to pay to have a quality pitcher (hopefully!) for three years.

      Vote -1 Vote +1

    • vivalajeter says:

      This is the type of issue that perplexes me. Why would take the mound every 5 days and suck, rather than miss 2-3 starts and come back stronger? It doesn’t make sense why pitchers try to tough it out like that. Same thing with Niese. He’s been getting crushed, and it turns out he’s not fully healthy. He’s not helping his team by pitching like crap – so just get healthy, and come back pitching up to your abilities.

      Vote -1 Vote +1

      • Wally Pipp says:

        That doesn’t always work out either, man.

        Vote -1 Vote +1

        • BronxBomber says:

          As WP says; Especially when considering that for a 38-year old it may take 2-3 months instead of 2-3 starts to come back to 80 percent instead of 100 percent.

          Vote -1 Vote +1

        • Jason B says:

          Either way, I would still rather wait two months and have 80% of excellence than slog through at something like 50% and hope for the best.

          Vote -1 Vote +1

    • Jaker says:

      This. I’ve been following Dickey’s velocity issues since the start of the season. First off, he generally pitches slower in April. But he’s been way down this year and up until his last start was trending downwards with each start. This seemed to coincide with his back/neck issues.

      I can’t imagine why they don’t rest him for 2 starts! I know the Jays having nothing else but they’re basically out of contention at this point. They should be doing whatever they can to preserve their ‘ace’, even if he himself says he’s fine to pitch. You don’t just lose 5 MPH on a pitch at random.

      Vote -1 Vote +1

      • Bob says:

        Oh you know maybe because Johnson, Happ, and Morrow have been hurt this year and Romero is learning to pitch again in the minors.

        Vote -1 Vote +1

        • Jaker says:

          Yeah, I understand that’s why. But I question whether it’s worth it. Why risk long-term injury? He seems to be pitching well today and perhaps it wasn’t serious enough to warrant a DL stint. Will be interesting to see what his velocity was today.

          Vote -1 Vote +1

        • Eno Sarris says:

          four knucklers over 79, most in second inning. better but not quite Cy Dickey

          Vote -1 Vote +1

  5. Myles Handley says:

    Why would you change the color of the bars from one graph to the next? That’s such a bad idea…

    Vote -1 Vote +1

  6. AC_Butcha_AC says:

    All this optical illusion stuff is killing me! It is like 1910 again, when they thought curveballs were an optical illusion.

    Every researcher out there… if a dancing knuckleball is an optical illusion try to explain THIS:

    http://cdn.fangraphs.com/not/wp-content/uploads/2012/12/knuckletastic.gif

    have a nice day

    Vote -1 Vote +1

    • Eno Sarris says:

      I wouldn’t say the movement is an optical illusion. But even that one has to obey the law of physics and go from his hand down. That one breaks a bit in each direction, but it’s not a slider that turns into two-seamer or anything.

      Vote -1 Vote +1

      • Dustin says:

        I read a book about the physics of the knuckleball. The movement comes from the rate of air flow over the seams. Air flows faster over the seams which creates a different in pressure, which creates movement. A 0 spin knuckler should move consistently in the same direction. The book figured the “perfect” pitch has exactly half a turn. That should result in an S curve.

        That’s all theoretical of course. The orientation of the ball, plus the variable amount of spin, uneven air distribution and wind all play a role. But it is physically possible for the ball to move in two distinct directions.

        Vote -1 Vote +1

        • Alan Nathan says:

          “But it is physically possible for the ball to move in two distinct directions.” Indeed that is correct, as discussed in my analysis of that one pitch. The question I have been working on more recently is whether that particular pitch is truly exceptional (i.e., occurring only rarely) or is much more common. I don’t have an answer yet.

          Vote -1 Vote +1

      • AC_Butcha_AC says:

        Well this isn’t the only sport in which there can be crazy knuckleballs. I am German and living in Germany and I watch tons of football (soccer). There are absolutely CRAZY knuckleballs changing direction in mid-flight. It is called “Flatterball” in German. I Absolutely think these things can happen with baseballs, too.

        And you are right, that balls obey the laws of physics. I even study physics. I am just saying, that these calculations are nice and stuff… but from guys who never really faced a knuckleball to make that claim is a bit weak.
        And even at 80 MPH, these subtle changes in direction are huge and exactly what Prof. Nathan thinks is impossible.

        Vote -1 Vote +1

        • Eno Sarris says:

          Nobody thinks it’s impossible for it to move in two directions. I put a qualifier in about the severity of the movement in those two directions, as you can see. Nathan himself dissected the knuckleball in that GIF and came to many of the same conclusions. But it cannot rise, and it cannot show the same severity of overall movement of a two-seamer in one direction. But it can go an inch or so in different directions, and I’m sure it’s possible that our minds filling in the blanks make it rise and fall like a butterfly.

          http://baseball.physics.illinois.edu/DickeyPitch103a.html#update

          Vote -1 Vote +1

        • Alan Nathan says:

          I don’t think Prof. Nathan ever said subtle changes in direction are not possible. In my earlier analysis of PITCHf/x trajectory data, I concluded that there was no evidence within the overall precision of the tracking data (~0.3″ in one case) that the trajectory of knuckleballs are less smooth than those of ordinary pitches. In that entire set of data, I found no convincing evidence for multiple breaks. That is not the same as claiming that such effects are not possible. Simulations of pitch trajectories using realistic forces that have been measured in wind tunnel experiments can predict multiple breaks (certainly up to 2), as seen in the particular pitch from TBA last year. So there is no doubt that such things are physically possible. The question is, how often are such effect realized in practice.

          Vote -1 Vote +1

    • Alan Nathan says:

      I think I already have explained that pitch:
      http://baseball.physics.illinois.edu/DickeyPitch103a.html.

      The pitch clearly breaks two different times. However, before you conclude that there is some sudden break, you should realize that the video clip that I and others (including 60 Minutes) have been staring at for a few months has a missing frame at just around the time of the first break. When I went over the video with the 60 Minutes producer, he caught that immediately. So, one must be very very careful looking at video and drawing conclusions about whether the trajectory of the pitch is smooth or not.

      Vote -1 Vote +1

  7. Ender says:

    Injured player has control problems, seems completely normal to me. If he gets fully healthy I expect him to be good, if he has nagging injuries I don’t.

    Vote -1 Vote +1

  8. ryan emperor says:

    I was just offered AGon/Dickey –for– Panda/Lohse…

    Was thinking about it…

    Then I read this…

    Thoughts?

    Vote -1 Vote +1

  9. Fitz says:

    Dickey is only getting strike calls on 71.5% of his pitches in-zone. Not sure what he was at last year, but the fact that nearly 30% of would be strikes are being called balls could be an explanation. Of course, that depends on what he has had in the past.

    Vote -1 Vote +1

    • za says:

      I’ve watched a few innings and he absolutely isn’t getting calls he should have been getting and did get as a Met. Inexperienced umpires maybe? The Jays should put Dickey on the DL and let him heal while concurrently calling up Thole to platoon with Arencibia. Neither is the perfect catcher, but Thole’s .270/.350/.380 with improved defense could help the team over Blanco.

      Vote -1 Vote +1

    • Qd6 says:

      Where did you get the source for your information? I’ve been seeing a lot of his pitches on gameday that look to be in the zone go for balls but I don’t know if that’s gameday’s fault or the catcher’s fault of not framing it properly.

      Vote -1 Vote +1

    • Jaker says:

      Source? Very interested to read this.

      Vote -1 Vote +1

    • Kevin says:

      anecdotally this is what i’ve been seeing too. most of the misses appear to be at the bottom of the zone, perhaps because umpires are calling the pitch where the catcher catches it (which can be ankle height on a ~75mph knuckler) rather than where it crosses the plate. now the pitch tracker they show on jays broadcasts isnt perfect, but other pitchers do not seem to lose as many low strikes as dickey does. overall i think the injury and his reduced velocity are bigger issues than the strikezone though.

      Vote -1 Vote +1

  10. Impossibles says:

    RA’s march/april 2012 is nearly identical to his march/april this year. Way too early to come to any conclusions.

    Vote -1 Vote +1

  11. SFDave says:

    Has anyone taken a look at the Rev per 60′ 6″? It’s possible he’s just not throwing as good of a knuckler. More spin on a knuckler is just plain BAD. That would be an exhausting study however, looking at all those pitches in slow motion and counting rotations. Sounds like FUN! Where can I get access to that kind of video?

    Vote -1 Vote +1

    • Alan Nathan says:

      I can vouch for the fact that the suggestion is an exhausting study. I am doing it and it is exhausting. Specifically, my collaborator Rod Cross and I have gone through many of Dickey’s games from 2012 and identified specific pitches for which there is good enough slo-mo video to see clearly the seam pattern and rotation rate. We then obtained raw tracking PITCHf/x data for those pitches. We are currently analyzing.

      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>