Does the stretch cost a pitcher fastball speed?

Earlier today, Rob Neyer wrote about Ubaldo Jimenez‘s no-hitter at his Sweet Spot blog at ESPN. He linked to an article by Thomas Harding about Jimenez’s switch of pitching deliveries mid-game:

It was after the fifth-inning leadoff walk that Apodaca mentioned the idea that saved the game and set up Jimenez for history.

At the start of innings, Jimenez, like every starting pitcher, was throwing with a full windup. Throughout his career, Jimenez has been far more effective from the windup than from the stretch, the side-to-the-plate stance pitchers use with men on base. But Saturday, he was more effective from the stretch. That meant less of a leg kick and, on Saturday, more control of his body.

“I talked to him between innings and he said he just felt lost,” Apodaca said. “To me, it was night and day the way he was executing pitches. His timing, as far as getting the ball out of his glove, and his delivery to the plate were all sharper out of the stretch.

That’s certainly fascinating to me, but what struck a researcher’s nerve with me was Rob’s comment about this all:

It’s often said that the stretch costs a pitcher 2-3 miles an hour off his fastball, and (considering how easy that is to check) I’ll assume that’s roughly accurate.

It is very easy to check, and it turns out that it’s not accurate at all. A pitcher’s fastball speed turns out to be almost identical with runners on base as compared to his average fastball speed with the bases empty. If anything, the average starting pitcher throws about 0.1 mph harder with runners on base.

fastball speed by baserunner state

Of course it could be that a typical pitcher bears down more and tries to throw harder when there are men on base in order to get the batter out and keep them from scoring. But if pitching from the stretch was a significant hindrance to fastball speed, you’d think we’d see it reflected in the data anyway, even if the pitcher was trying to throw harder. I doubt that the baseball adage that Rob mentioned was merely saying that pitchers end up throwing the same speed from the stretch as from the windup because they’re trying harder.

I realize that runners on base vs. bases empty does not correspond exactly to pitching from the stretch vs. the windup for every pitcher, but it should be close to enough to reveal any major differences between the two. I looked at the time period 2008-2009, first at all pitchers who had thrown at least 1000 fastballs, which would include quite a few relievers, and then those pitchers who had thrown at least 2000 fastballs, which should be mostly starting pitchers. The results were not markedly different between the two groups.

What about Jimenez in particular? He has averaged 95.7 mph with the bases empty and 95.6 mph with runners on base.

The two starting pitchers who really crank it up with men on base? That’s Justin Verlander–94.1 mph with the bases empty and 95.4 mph with runners on–and Ted Lilly–86.2 mph with bases empty and 87.7 mph with runners on. And a virtual cookie (your favorite kind, of course) to the first person who can pick out Jamie Moyer‘s dot on that chart.

Print Friendly
 Share on Facebook2Tweet about this on Twitter1Share on Google+0Share on Reddit0Email this to someone
« Previous: And That Happened
Next: And THat Happened »


  1. Mike Fast said...

    A cookie to Kenny!  What kind would you like?

    The one at the top right is Broxton.  I don’t think Zumaya pitched enough to reach the 1000-fastball cutoff.

  2. Peter Jensen said...

    The major downside of pitching from the stretch is that it’s more stressful on the pitcher’s arm.

    Greg – Do you have a reference to research that confirms this statement?

  3. Mike Fast said...

    Peter, this study from the ASMI guys is the only one I can find.

    Dun S, Kingsley D, Fleisig GS, Loftice J, Andrews JR.  Biomechanical comparison of the fastball from wind-up and the fastball from stretch in professional baseball pitchers. Am J Sports Med 36(1):137-41, 2008.

    Abstract is here:

    Methods: Twenty-eight professional baseball pitchers (aged 22.1 ± 2.8 years) pitched fastballs from both the wind-up and stretch positions in an indoor laboratory setting. Three-dimensional motion-analysis systems were employed to capture the pitching motion. Kinetic variables, kinematic variables at lead-foot contact, and temporal variables of the 2 pitching variations were compared.

    Results: There was no significant difference between the 2 pitch variations for the kinetic, kinematic, or temporal variables. The difference between the ball velocities was statistically significant, but the mean difference was only 0.2 m/s.

    Conclusions: The pitching biomechanics between the wind-up and stretch fastball showed no statistical differences in joint kinetics, kinematics, or timing, and clinically insignificant differences in ball velocity.

  4. Matt Lentzner said...

    Cool little study, Mike. Really liked it.

    My take would be that it’s more tiring to pitch from the stretch. Essentially the pitcher has to do the same amount of work to accelerate the baseball in much less time. Hence more power (work/time) is needed. Physiologically, this is more taxing since a higher level of muscle recruitment is needed.

  5. danmerqury said...

    Wow, Broxton’s fastball is that much faster than everyone else that qualified?

    Anyway, very cool, Mike. I’ve been wondering about pitch differences between a normal windup and the stretch for some time now.

  6. Alex said...

    Perhaps pitchers practice more from the stretch than from the wind-up getting them more used to and able to throw at the higher speed. When using a slide step though pitchers probably loose velocity but I know that a former White Sox pitching coach worked on getting a lot of Sox pitchers, and Mussina (when with the O’s) away from using the slide step because of the velocity and control issues along with it.

    It’d be interesting to see if pitchers loose velocity with faster runners on first (those who are likely to steal)

  7. Peter Jensen said...

    Mike – Thanks. I had that ASMI study abstract.  I was interested if Greg had a reference to other studies that had come to different conclusions.  I guess I am wondering whether it would make sense to abandon the windup for all pitchers.  There must be some consistency advantage to be gained from throwing from a single motion rather than two separate motions.  I would also think that switching between the two would be more likely to cause injury.

  8. Mike Fast said...

    Peter, all that my Google search turned up was that paper, plus a message board conversation from 2005 where Fleisig asked if anyone had published research about the stretch vs. the windup.  Kingsley responded to Fleisig’s post that no, there was no such published research, but Kingsley had done such a study but hadn’t published yet. 

    Presumably he was referring to the research for the paper they subsequently published in 2008.

  9. The A Team said...

    I know it’s a little silly to compare my experiences as a D-III college player to pros, but I often went with the stretch over the wind-up for control reasons. I also tended to get a little more movement out of the stretch, but I’m willing to chalk that up to crisper mechanics.

    It is my personal belief that with the bases empty, pitching from the stretch is no more stressful than pitching from the wind-up. That may be confirmation bias or just a flat out incorrect but that’s my personal anecdote for this conversation.

    With runners on, pitching from the stretch did seem a little more stressful on my arm. I imagine that this is something that you really need to look at the individual to measure. Some guys are going to be low stress with runners on base and some guys are going to over-do things a bit. I’d imagine that most major leaguers are going to see pretty low stress results since most of the other guys probably wouldn’t make it that far.

  10. Greg Andrew said...

    The major downside of pitching from the stretch is that it’s more stressful on the pitcher’s arm.  The “loses mph off fastball” is a myth that has proven hard to get rid of.

  11. Mike Fast said...

    Given that pitchers pitch worse with the stretch than with the windup the difference must be captured in one of those two things or its all in the hitters.

    Do they?  Not as a group, where the worst pitchers will put runners on base more often and overweight their influence on the sample, but individually (on average)?  I’m not saying they don’t, I’m just not sure.

  12. Nick Steiner said...


    I had thought it was indeed the case that pitchers did pitch worse (using the delta method) with runners on base.  I can’t remember where I saw the numbers though.  It shouldn’t be that hard for someone to wrestle up.

  13. Mike Fast said...

    Mmmm…peanut butter cookies.  Okay, I’ll make a virtual batch.

    Wakefield didn’t throw enough fastballs to make the chart, but his fastball comes in around 73 mph.

    The next three lowest points on the graph after Moyer are for Doug Davis, Livan Hernandez, and Greg Maddux.

  14. Pull_T said...

    Kenny said… Who is the second-slowest pitcher? Wakefield?

    Can’t be Wakefield.

    1.  His fastball is mid-upper 70s.
    2.  He couldn’t possibly have thrown 1000 fastballs over two seasons.

  15. MGL said...


    It is in The Book, and it is probably because there is less deception.  Of course you have to control for the extra holes in the IF and what have you, which I think we did.


    If indeed on the average pitchers pitches 2-3 mpg worse from the stretch, then yes, you would see it in about 10 minutes (I am making that number up of course) if you watched a game.  Obviously, you might get fooled every once in a while, but not often.

    So what exactly is wrong with that statement?  Is it not true?  I made no reference to anything to do with a study whatsoever.  I was simply making an observation and I was implying that it is odd that Rob (or anyone else) would write something like that when he obviously watches a lot of games, and if it were even close to being true, all he would have to do is watch one game and he would have a very good idea as to whether it were true or not.  That is ALL I was saying.  What are you reading something into my statement which is not there?  Or are you interpreting my “10 minutes” as being literal, which would be preposterous of course?

  16. Nick Steiner said...

    I thought you were criticizing Mike’s article for “needing a study” to prove Rob’s statement wrong.  Obviously I misread your comments.

  17. Alan Nathan said...

    I find it interesting that the pitching mechanics is the same for windup vs. stretch.  It certainly does not look the same to my(admittedly untrained) eye.  I am going to ask Glenn Fleisig of ASMI to post a comment here about his finding.

  18. RobMac said...

    I recall watching a spring training 1974 telecast from Scottsdale of a night game between the west-trekking Dodgers and the host Giants.  Don Rose threw the straightest fastball you have ever seen and Steve Garvey took it for a strike.  I turned to my elderly friend and said, “If he throws another one of those Garvey will hit it out.”  Sure enough, boom, home run.

    My point is that Rose, like a zillion other major league pitchers, had good velocity.  Power fastballs are sexy and intriguing, yet many major league batters feast upon fat 95+ mph heaters.

    Studies like the above can discuss pure power, yet don’t study the movement (or lack thereof), of pitches.  Last night I saw Kendry Morales whack a 100 mph Joel Zamaya pitch back up the middle for a blistering single.  Last season I saw Carlos Ruiz beat Jonathan Broxton on a 101 mph pitch as he doubled to right center.  A few years back in a battle of steroid/HGH users, Barry Bonds went yard on Eric Gagne in SF on a 98 mph pitch.  I believe that Vernon Wells has taken Josh Beckett deep twice on 98 mph fastballs.  Honestly, these kinds of things happen every week.

    The best pitchers combine control, ball movement, and power, and they keep batters off balance and in wonderment about what and where will be the next pitch.

    I recall Sandy Koufax retorting to a reporter after being lauded for being the best lefty in history with something like, “Me?  How about Warren Spahn?  He’s won 363 games!”

    Also, the reason a smart pitcher would not eschew the windup is that he has more time and opportunity to effectively hide the ball than when in the stretch.  Can you imagine Spahn, Koufax, Louis Tiant, Mickey Lolich, or Juan Marichal being coached to 86 their windups and just go out of the stretch?  Opposing batters sure wished that they would!

    A great windup allows you all the time you need to hide the ball and distract the batter.  Milliseconds are everything and if you can force the hitter to wait until the last possible moment to pick up the flight of the ball then he has a decreased chance of putting a good swing on the pitch.

    Kenny Holtzman’s windup was renowned for hiding the ball very effectively. 

    Though I despise his mechanics and feel strongly that they have led to diminished velocity and arm injury, the Padres’ Chris Young hides the ball effectively so that even his pedestrian fastball is picked up later than normal and thus seems to surprise batters when he is at his best.  He is using his stork-like 6’10’ attributes to hide the ball well, but, to me, does not use his whole body well to be as good and as healthy as he could be with better mechanics.

    Same with Clayton Kershaw.  Get him into a film room for two hours watching nothing but Steve Carlton, Sandy Koufax, and Warren Spahn and he would have to emerge with smart adjustments versus the awkward approach he now features, resulting in a rollercoaster ride of “he’s got it!” to “oh, he’s lost it!”

  19. Dr. Glenn Fleisig said...

    Hi Alan and others,

    Our study measured the biomechanics of 28 professional baseball pitchers each pitching fastballs from the windup and stretch in the ASMI lab.  We found no differences in the kinematics (motions), kinetics (joint forces and torques), and timing between the two pitch styles.  I want to add that we compared the motions starting at the instant of front foot contact with the mound.  There are obvious differences in the beginning of the pitching motion, but the high forces and torques and rapid body motions occur after the instant of foot contact.

    For more on this, please read the entire article.  I don’t check this message board, but you can reach us at the Online Sports Medicine Forum on

    - Glenn S. Fleisig, Ph.D.

  20. Kenny said...

    Peanut butter, thanks.

    It’s surprising that Broxton has that much more velocity than anyone else. It’s amazing what the modern bullpen has done for velocity.

    In “ooh, look at the pretty picture” news, I find it amusing that both of the far left dots are perfectly pierced at a 45 degree angle by the slope line. Who is the second-slowest pitcher? Wakefield?

  21. Nick Steiner said...

    Well this seems pretty conclusive Mike.  Good job.  Maybe you could check pitch movement or location to see if there is an effect on those.  Given that pitchers pitch worse with the stretch than with the windup the difference must be captured in one of those two things or its all in the hitters.

  22. MGL said...

    “It’s often said that the stretch costs a pitcher 2-3 miles an hour off his fastball, and (considering how easy that is to check) I’ll assume that’s roughly accurate.”

    Come on, you wouldn’t need to study the numbers to realize how preposterous that assumption is.  Almost all relievers throw from the stretch and all closers throw from the stretch, even with no runners on base.  And relievers throw harder than starters. If throwing from the stretch cost you significant mph, one, relievers would not throw from the stretch if they didn’t have to, and two, relievers would have to be superhuman if they threw harder overall than starters (which they do) even though they pitch from the stretch almost all the time, and starters pitch from the windup at least half the time (I think).

  23. MGL said...

    Not to mention the fact that if you watch a game or two (I realize that analysts typically don’t watch games), you would see in about 10 minutes that there is no discernible difference between when a pitcher throws from the stretch and when he throws from the windup.

    Seriously, sometimes I think that people who write about and comment on baseball DON’T ever watch games, or at least they don’t think or pay much attention when they do.

    On Primer the other day, there were about 50 comments or more about the “pitcher fakes to third and throws to first – Steve Busby” move.  People were going on about how the pitcher should actually throw to third occasionally so that the runner on first knows that it is not always a fake.  Not one person, even the regulars (not that I have much respect for most of the regulars on Primer – I don’t), mentioned the reality of the “fake to third and throw to first” move.  Not one.

    The runner on first base does not think you are throwing to third and thus it is not really a “fake” to third and you don’t need to occasionally throw over there (to keep the runner in first “honest”).  You are trying to deceive the runner on first into thinking that you are delivering the ball home. You can’t fake to home of course, so you have to step toward third. If the runner on first thought you were going to throw to third, he would NOT try and steal!  In fairness to the Primer people, ONE person did mention this even though everyone else ignored him and went on about the runner at first thinking that the pitcher was throwing to third…

  24. Nick Steiner said...

    I’m very surprised those last two comments came from MGL. 

    Not to mention the fact that if you watch a game or two (I realize that analysts typically don’t watch games), you would see in about 10 minutes that there is no discernible difference between when a pitcher throws from the stretch and when he throws from the windup.

    Yeah, that makes for a good study.  Some guy says pitcher’s lose 2-3 MPH from the windup to the stretch and Mike writes an article saying, “no, of course not, anyone who’s watched a game knows that there is no discernible difference between fastball velocity from the windup and from the stretch”.  If only there was a certain field of study dedicated towards quantitatively and objectively investigating such questions…

  25. MattG said...

    I find it hard to understand how a wind-up can aid a pitcher in hiding the ball, since a pitcher’s release point is not likely to change from the wind-up to the stretch. The wind-up is, for the most part, something that happens before the pitcher gets into his throwing mechanics. If he can hide it in the wind-up, he’s going to hide it in the stretch, too.

  26. RobMac said...

    A windup can simply take more time in distracting the hitter with a deliberate backswing, Satchel Paige’s famous hesitation pitch, Juan Marichal’s extremely high leg kick, Luis Tiant’s turning showing his number to the hitter as he faces his centerfielder, Mickey Lolich’s unique movements prior to heading toward the plate, etc.

    From the stretch the main point is to minimize your time in getting the ball to your catcher.

    With an elaborate windup you can take your sweet time while the batter has to keep his cool and concentration.  A smart pitcher wants to keep him in a tense physical condition as long as possible.

    From the stretch you need to hide the ball effectively, as well, yet the reality is that you have much less time to do it.

    Make sense?

  27. MattG said...

    Yes and no…

    I understand what you write, and I understand how all that would be distracting—which is not really the same as hiding the ball. Hiding the ball, to me, would be a product of how the ball gets to the release point. It can come from a very traditional angle, or it can come out from behind a pitcher’s head, or curled in his hand, and so forth. And that would remain consistent in wind-up or stretch.

  28. RobMac said...

    Gotcha.  The superior hiding of the ball out of the stretch may take a tick or two longer than just hurrying the ball to the plate to make that ideal 1.3 second goal.  I say this because the glove arm is probably the best way to stall the ball being seen by the batter.  Taking the trouble to do this well may take a tad longer, but my main point here is that one can be quick to the plate and throw a superior fastball all he wants, but I’d rather have a guy on the mound who makes a superior pitch with location, movement and (icing on the cake but not the most important element), velocity after hiding the ball well.

    Thus, even with runners on third, second and third, and loaded, many pitchers feel that they toss their best stuff out of a windup and they are willing to have the runners get good leads off of his windup.

    I’d say that more modern pitchers prefer the stretch.  Heck, I’m helping with a Mustang age team (9 & 10), now, and only one kid on the whole team likes to start out with a windup.  Being young and new to the game, their main stress is having a compact approach that maximizes strikes.  Hopefully, they will learn the full value of an excellent windup.

  29. MattG said...

    OK, now I see why a pitcher can hide the ball better in the wind-up than the stretch.

    But do pitchers vary their glove positioning from wind-up to stretch? I always believed that the mechanics from one to the other were very much the same. It never occurred to me that pitchers might purposely use two different throwing motions, even if the only difference is something as minor as the position of their glove hand.

  30. RobMac said...

    That’s a great question to study.  How much does a pitcher alter his approach from the windup to throwing from the stretch?

    I bet it varies widely pitcher to pitcher.  My guess would be that starters have much better stats when pitching out of the windup…but that is just my unscientific hunch…

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>