FanGraphs Baseball


RSS feed for comments on this post.

  1. “There have been many knuckleball pitchers before, but none has ever finished higher than third in the Cy Young voting before. Dickey looks like he will break that glass ceiling this year ”

    Really, you are already giving Dickey a top 2 finish in the Cy Young this year. Dude it’s June.

    Comment by Mark — June 21, 2012 @ 10:37 am

  2. I wonder if even though there is less movement the higher the velocity of the pitch, iif it doesn’t present some sort of visual difficulty for the batter to gauge the velocity since there isn’t any rotation.

    Comment by Nathan — June 21, 2012 @ 10:38 am

  3. His 13K performance featured a strike 3 on a knuckler that didn’t even top 60. It’s the complete unpredictability of location, movement, AND velocity which makes Dickey truly impressive this year. He’s playing confuse-a-batter out there every time using every available trick, and he’s winning big.

    Comment by Well-Beered Englishman — June 21, 2012 @ 10:42 am

  4. Agreed that it’s way too early to talk about Cy Young. But he’s got a good shot of being the starter of the all-star game, which is pretty cool.

    Comment by Brian — June 21, 2012 @ 10:54 am

  5. Along the same lines, I wonder if the difficulty in hitting a knuckleball comes less from the magnitude of movement than the unpredictability of the direction of movement. If the latter, higher velocity could possibly make it harder to hit even with reduced movement, simply because it gives the batter less time to recognize how the ball is moving as it nears the plate.

    I guess the truth is that despite the explosion in pitch-level data we still just don’t know enough about the knuckleball to draw any meaningful conclusions.

    Comment by mcbrown — June 21, 2012 @ 10:58 am

  6. This article would have been a lot easier to read if the three graphs all had the same X-axis scale. Otherwise I can’t really compare the difference without remapping them in my head. Otherwise, very very interesting. Thanks.

    Comment by Adam — June 21, 2012 @ 11:01 am

  7. Thank you sharing this. But, the effects would be more clear if the dimension of the X and Y axis on the “Knuckler Velocity” graphs were even….and if you overlaid the 2011 and 2012 distributions!

    Comment by Matt — June 21, 2012 @ 11:13 am

  8. Concur! I’ll add that a consistent Y-axis would help as well.

    Still, solid read. The next time I see Dickey pitch will be the next time he pitches. His last three starts have been very fun to see. I’m thinking about getting mlb tv just to follow his season (my team’s in market, so I haven’t been able to justify it yet).

    Comment by fresheee — June 21, 2012 @ 11:18 am

  9. If the axis were the same on all three graphs, the graphs would actually look pretty similar. The author’s distortion of the images supports his argument.

    Comment by rz — June 21, 2012 @ 11:20 am

  10. Eno, nice work, I was at the Rays/Mets game last week and Dickey’s knuckler was completely mystifying the Rays, who I remember had a good track record against Wakefield in his last few years.

    You make a great case for Dickey’s increased Velocity over the past three years, but you don’t present evidence for “less Movement” over the past three year, in Dickey’s case at least.

    With the information presented above, I could just as easily argue that one of the reasons that Dickey is doing so well this year is that he has been able to increase his velocity without significantly impacting the movement on his knuckleball.

    Is there any way to test that hypothesis? Or is the data Dickey’s knuckleball movement not available from prior years?

    Comment by JKB — June 21, 2012 @ 11:21 am

  11. I’ll rein in my excitement/homerism haha.

    Comment by Eno Sarris — June 21, 2012 @ 11:57 am

  12. rz I obviously disagree. Look at 74 MPH knuckleballs. I probably should have overlaid, but I’m not the whizziest whiz at excel. But the count determined the axis, and you can see that not only are there no super-slow-mo knucklers so far this year, but he’s used the 74 MPH knuckleballs less.

    Comment by Eno Sarris — June 21, 2012 @ 11:59 am

  13. Follow the links. Those guys did that work, so I tried to focus on the velocity.

    Comment by Eno Sarris — June 21, 2012 @ 12:00 pm

  14. I have been spending lots of time recently looking at RA’s pitches from the 2010-2012 seasons, courtesy of Harry Pavlidis who supplied the PITCHf/x data for me. I have tried to quantify, among other things, the dependence of movement on speed. I am finding that the mean movement is pretty flat at around 10″ in the 70-75 mph range, then gradually decreases to about 8.5″ over the range 75-83 mph. For the PITCHf/x experts, the total movement is sqrt(pfx_x^2+pfx_z^2).

    My own view is that Dickey effectiveness is due not only to the random direction of the movement but to his ability to mix in different speeds. So, even with less movement at the higher speed, the fact that the speed is different makes the batter’s life difficult in and of itself.

    Comment by Alan Nathan — June 21, 2012 @ 12:01 pm

  15. I think Dickey’s season this year is a perfect illustrative example of overkill with the numbers (credit to Keith Hernandez who makes this point all the time) – anyone who has watched Dickey the last three years can notice four major changes beginning after his start in Atlanta earlier this year:
    1) He is getting WAY more borderline called strikes – that’s the biggest difference by far.
    2) He is throwing his fastball less and his “hard” knuckler more often, and has become a master at changing speeds on the pitch – he is also “jamming” LH batters with the knuckler, especially on two-strike counts. But throwing the fastball less is more a product of consistently getting ahead and staying ahead in the count.
    3) Batters are having a much more difficult time recognizing which knucklers are in or out of the strike zone (see # 1) – kinda reminds me of Glavine in his prime when RH batters would consistently swing at pitches 9 inches off the outside corner ’cause they knew it would be a called strike.
    4) He is throwing a different “rising” type of knuckler that appears to have upward movement that batters are really struggling with – Dickey has acknowledged this new pitch in interviews.

    He is no doubt benefitting from his evolving reputation as a pitcher who throws strikes, and just like Maddux and Glavine is taking full advantage of it. Congrats to R.A. on a dominating season to this point!

    Comment by Chris — June 21, 2012 @ 12:13 pm

  16. Joe Niekro threw a hard knuckle. Reportedly 77-83 mph, and sometimes mid-80’s. While looking for the exact velo, I came across:

    Joe Niekro’s Top 10 Excuses – August 6, 1987 – David Letterman Show

    10. The emory board is a new super-grip popsicle stick.
    9. I only used it to apply Vaseline to the ball.
    8. I needed it to scrape dried wads of chewing tobacco off the bullpen telephone.
    7. Delicate double-knit uniforms easily snagged on rough nails.
    6. I was using it to make a statue of commissioner Ueberroth.
    5. I used it as a bookmark for my dugout copy of Shirley MacLaine’s autobiography.
    4. Rules of fair play are for saps and squares.
    3. I’ve been hypnotized by evil dogs.
    2. It was all William Casey’s idea.
    1. I like to give pedicures to ballboys.

    Comment by Choo — June 21, 2012 @ 12:32 pm

  17. You can manually select the range of the axis in Excel, though, so the count didn’t have to determine the axis. It would be helpful to have the same scale on all three graphs even if that meant chopping off some outlying data that is insignificant to the point under discussion.

    Comment by steex — June 21, 2012 @ 12:47 pm

  18. I like making wild claims too!

    Comment by Nik — June 21, 2012 @ 12:56 pm

  19. you guys only have one dickey article today? not getting it done

    Comment by wily mo — June 21, 2012 @ 1:12 pm

  20. Well, Vegas currently has Dickey as the favorite to be the NL Cy Young winner. I don’t agree (he’s far too difficult to project to name him the favorite) but I don’t think it’s a huge crime to say that he could finish in the top 2.

    Comment by Craig Glaser — June 21, 2012 @ 2:30 pm

  21. Really, are you ignoring facts. Dude he leads the league in wins, ERA, 2nd in strikeouts and innings pitched, 3rd in BAA, 5th in K/BB.

    Comment by BlackOps — June 21, 2012 @ 8:04 pm

  22. ask and you receive. I think that last one, with the blue line lowest in a percentage look, shows this effect the best. Thanks for pointing this out, still trying to get better at Excel and visualizations.

    Comment by Eno Sarris — June 21, 2012 @ 8:38 pm

  23. Although I am not a Mets fan, I love to watch knucklers pitch, so I am a Dickey fan. I feel, it is far and away the coolest pitch in baseball!!! 100 mph fastballs are fun to watch but a guy swinging way early or nowhere near close on a knuckler is even more fun to see for me!

    Comment by kick me in the GO NATS — June 22, 2012 @ 2:04 am

  24. Wow you even strike out on your posts!

    Comment by kick me in the GO NATS — June 22, 2012 @ 2:05 am

  25. There’s all this talk in the article of three clusters, but the graphs I see above suggest that Dickey’s throwing lots of pitches at all the varying speeds between those “clusters” as well. If you just claimed that Dickey generally varies speeds between 73 and 80 mph – sometimes throwing a bit harder, sometimes a lot softer – it would match my interpretation of the data better than the claim than there are three clusters.

    Comment by Jon L. — June 22, 2012 @ 4:50 am

  26. My PhD was all about fitting gaussians to noisy graphs, if I tried to claim the above could clearly be fit to three Gaussians my supervisor would have asked me to kindly return to my analysis package (or words to that effect).

    Comment by Simon Orr — June 22, 2012 @ 11:32 am

  27. Could you make the graph embiggen-able?

    Comment by John Peterson — June 22, 2012 @ 11:47 am

  28. Perhaps I saw what I wanted to see. Past research has suggested (and I linked to it) that once you add in movement to the velocity, he has two knucklers and a superslowmo, which he used rarely in 2010 and 2011 and hasn’t used much this year.

    Comment by Eno Sarris — June 22, 2012 @ 11:49 am

  29. I have no issues with saying he “could” finish in the top 2 but that’s a big difference from saying he will. I would think if they voted now he would win it.

    Comment by Mark — June 22, 2012 @ 12:03 pm

  30. “Dickey could break that glass ceiling this year.”

    ‘Could,’ not ‘will.’ Very big difference.

    Comment by Jeremy — June 22, 2012 @ 12:34 pm

  31. Yes, it’s a big difference. And the original article said ‘will’.

    Comment by vivalajeter — June 22, 2012 @ 12:57 pm

  32. The reaching to give any legitimacy to Dickey’s nice little run is hilarious. The knuckle is a gimmick with properties and statistical outcome completely known by now. There are three reasons why Dickey is making it seem like he’s different:

    1. As pointed out above, some generous umpire love;
    2. As pointed out by me elsewhere, 2012 the highest strikeout rate in MLB history — batters are taking awful approaches to hitting him; and
    3. An infinite amount of monkeys on an infinite number of typewriters will eventually type William Shakesphere’s works in correct order.

    Just sit back and enjoy it for what it is — a fun, amazing, lucky ride for a nice guy who deserves it. But he will finish with an ERA of around 3.90. Of course then I’m sure fangraphs will post how it’s not his fault because his xFIP or something.

    Comment by Fatbot — June 22, 2012 @ 1:00 pm

  33. The links that mention movement pretty much focus on Dickey’s movement on pitches this year, in particular the recent game against Baltimore.

    I haven’t seen any evidence yet that his movement is going down any with the uptick in speed. The reason that his average speed is up this year is that he is throwing fewer pitches in the 72-74 MPH range.

    Unless Dickey’s 72-74 MPH pitches were the pitches with the most movement historically, I don’t see any reason to expect his average pitch movement to decrease significantly this year.

    Comment by JKB — June 22, 2012 @ 1:13 pm

  34. I wish I would have read your post before replying above. Is there an ideal speed for Dickey’s pitch movement (like 72 MPH) or is it pretty consistent in the 70-75 MPH range?

    Comment by JKB — June 22, 2012 @ 1:20 pm

  35. Most MLB hitters can time a great fastball. A great knuckleball is different, even if you know it’s coming you still can’t hit it.

    Comment by Hurtlockertwo — June 22, 2012 @ 1:50 pm

  36. What I think Dickey’s success has revealed is how vulnerable even major league hitters are to unpredictability. The hitters simply don’t know what to expect in terms of location, movement or velocity when Dickey delivers. And if the hard knuckler allows Dickey to throw more strikes and relatively less time to adjust the ‘knuckling’ of the ball, that puts the hitters backs against the wall even more.

    What will be interesting to see is if Dickey does indeed have patterns in velocity and location, if not movement, particularly on certain counts, whether any hitters are able to make an adjustment and feel more confident in anticipating where and how soon to swing.

    Because for the moment most of them look totally clueless. Merely reacting without being able to anticipate.

    Comment by Samuel — June 22, 2012 @ 6:58 pm

  37. His velocity and prowess on the mound can not be questioned this year yet his honesty in this book can. I personally think that he is just using the subject of mokestation which i think isnt true to make some cash and to get his wife’s forgiveness and i wrote why that is right here

    Comment by Eddie — June 22, 2012 @ 9:10 pm

  38. You’ll do anything for website hits, huh? What a douche.

    Comment by vivalajeter — June 23, 2012 @ 1:00 am

  39. Seriously, can the mods delete – at least – the link?

    Comment by Dandy Salderson — June 23, 2012 @ 8:00 am

  40. While, there is less overall movement on Dickey’s harder thrown knuckleballs, it looks like most of the movement comes right as the knuckleball goes over the plate. So, overall it’s easier for Dickey to locate the 79+ knucklers since it comes out of his hand more straight versus his low seventies knuckler which sometimes starts to flutter the second it leaves his hand.

    The game plan versus most knuckleball throwers was simply to make him throw strikes. Then, if it’s low let it go, if it’s high let it fly. Dickey has turned that theory on it’s head due to his ability to throw strikes and his harder thrown knuckleballs having a late elevating movement.

    I would give RA Dickey a better chance than almost anyone else at finishing in the top 3 for Cy Young votes for three reasons. Number one, he plays in a very, very pitcher friendly park. Number two, Dickey never gets worn out. So, he never has to turn the game over to the Mets horrible pen and while a lot of the other Cy Young candidates begin to wear down this season he will be floating along. This is Lynn’s first year as a starter so I imagine there is some sort of innings limit on him. McDonald threw 170 last year which was way more than his previous career high of 64IP. I’m not sure if there will be an innings limit on him or not. Then, Strasburg definitely has an innings cap. Plus, with Beachey being done for the year I would have to place Cain, Dickey, Kershaw, and Greinke as the top four contenders. Finally, number three is that Dickey’s knuckleball is only going to get more nasty as it gets more hot and humid.

    Comment by Ryan — June 23, 2012 @ 8:20 am

  41. “First of all Dickey has been in the league for 10 years now and up until this season he has been terrible”

    WRONG. He’s actually been quality since 2010. I thought this site was for the eggheads who know enough to look past wins and losses

    Comment by SRT120 — June 23, 2012 @ 8:45 am

  42. he’s cy young material. i don’t see the harm in making that projection this early. it’s a projection. whether this early or not, either case can prove wrong. but i think it’s a fair prediction. he’s playing lights out. if he keeps it up… why not?

    Comment by musclemenace — June 29, 2012 @ 11:09 pm

  43. “But these velocity graphs don’t know counts. They show that regardless of count, Dickey is focusing on the faster knuckleball.”

    Know they don’t. Dickey could simply be involved in more counts where he (or his catcher) feels the need to throw strikes.

    Comment by Jack Strawb — December 2, 2012 @ 3:51 am

  44. Listen to Eno he is right

    Comment by Matt — May 11, 2013 @ 12:19 am

  45. Phil Neikro finished second in NL Cy Young voting in 1969.

    Comment by Spit Ball — September 15, 2014 @ 9:09 am

  46. MY nephew is a professional pitcher who throws a knuckleball 84 mph and can not get anyone to even look at him.

    Comment by Paul Jannis — March 17, 2015 @ 7:08 pm

Leave a comment

Line and paragraph breaks automatic, e-mail address never displayed, HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Close this window.

0.206 Powered by WordPress