Were the All Stars really throwing that hard?

If you watched the All-Star Game on Tuesday night, you may have noticed a lot high-nineties fastballs zipping from pitcher to catcher. At one point Fox put up a graphic listing the fastest fastballs of the night:

Pitcher    Speed (mph)
Price        100
Verlander    99
Johnson      99

Some people wondered, particularly with David Price, whose average fastball speed this year is 95 mph, whether the Fox speed gun was “hot”. I believe that Fox was actually getting its pitch speed data from PITCHf/x rather than a radar gun since the speeds that Fox reported match up very well with the speeds in the PITCHf/x data for the game.

Rather than looking at peak speeds, let’s look at the average fastball speed for every pitcher who threw in the game.

Pitcher    Speed (mph)
Verlander    99
Price        98
Wilson       98
Johnson      98
Thornton     97
Broxton      97
Jimenez      97
Valverde     96
Wainwright   95
Kuo          95
Hughes       95
Bailey       95
Capps        94
Lester       94
Bell         94
Halladay     93
Soriano      93
Lee          92
Pettitte     91

That’s a lot of mid and upper-nineties fastballs! Of the 272 pitches thrown, 190, or 70 percent, were fastballs. The average fastball speed in the game was 96 mph. Wow.

Was the PITCHf/x system reporting pitch speeds accurately in the All-Star Game? The simple answer is, as far as I can tell, yes.

Many of the pitchers were definitely recording faster speeds than they had throughout the season. In addition to the aforementioned Price, Justin Verlander, Josh Johnson, and Adam Wainwright were each measured as bringing their heat two to three mph faster than during the season. The average fastball for all pitchers in the game was measured at one mph faster than the same pitchers threw during the season.

The initial inclination would be to say that the PITCHf/x camera system in Angel Stadium was out of calibration such that it was measuring pitch speeds about one mph too fast. That sort of error is not unheard of. However, we should also not be surprised if starting pitchers threw their fastballs harder than usual in short one or two inning stints or if the emotion of the confrontation with All-Star batters on a national TV stage was enough to give well-rested pitchers a little extra zip.

I don’t know a simple way to determine whether either of those things are true, although we do see that the pitchers with the biggest boosts to their fastball speed were all starting pitchers. The average fastball speed boost relative to the regular season was 1.5 mph for starters and 0.4 mph for relievers.

Moreover, there are a couple things that I usually check when making adjustments to pitchers’ fastball speed data in PITCHf/x. The first is the drag coefficient calculated from the PITCHf/x data. This is a physical constant that is dependent on things like the physical characteristics of the baseball but independent of a lot of other things, like the ballpark. The drag coefficient measured by the PITCHf/x system on Tuesday night was very close to the average expected value of 0.36.

The other thing I check is the average fastball speed of pitchers in a given game relative to their season average. But instead of simply doing it for a single game, as reported above, I do it for multiple games and look for trends. The average fastball speed recorded by PITCHf/x in Angel Stadium this year has actually been 0.8 mph less than the average fastball speed for the same set of pitchers pitching in other ballparks.

So, if anything, the “gun” in Angel Stadium has been a little cold in 2010. (I don’t know whether the PITCHf/x system was recalibrated immediately prior to the All-Star Game, such that any bias in the data from earlier in the season would be irrelevant to the data collected Tuesday. We’ll know more about that after we get data from the Angels’ first homestand after the break.)

Fastball speed data is interesting, but we don’t always know how accurate it is. In this case, although I can’t give an official Good Housekeeping Seal of Approval, I don’t see any convincing reason to believe that the speeds that Fox (and PITCHf/x) reported were juiced.

Print Friendly
 Share on Facebook0Tweet about this on Twitter0Share on Google+0Share on Reddit0Email this to someone
« Previous: And That Happened: Classic
Next: Heat from the Futures Game »


  1. Nick Steiner said...

    Good stuff Mike.  It was incredible to see Wainwright throwing in the mid 90’s along with his incredible curveball (and what looked like a low 90’s cutter, which apparently he only throws in relief), but I thought it was too be good to be true given how much sense it would make for MLB to juice up the guns a little bit. 

    One thing I’ve heard was that the guns were juiced for the first inning with Price only and then they fixed them.  Is there any evidence for that?

  2. Harry Pavlidis said...

    And I just posted a run-down of the Futures game speeds, without the Seal of Approval. Perhaps a Donkey of Admonishment.

  3. Mike Fast said...

    Thanks, Nick.  Price’s fastball speed did drop from 98.5 mph in the first inning to 97.3 mph in the second inning, but Ubaldo’s speed jumped from 96.4 in the first to 97.8 in the second. 

    I suppose it’s possible they made a minor re-calibration in between half innings in the first.  I have no way of knowing without asking the Sportvision folks.  But even if they did, it wouldn’t explain the speed readings from the rest of the game.  Price’s first inning was the biggest anomaly, but it was far from the only one.  Wainwright’s speed was another notable one, as you mentioned, and that occurred in the 7th inning.

    I wonder if people didn’t make that speculation based on Price’s first inning velocity being so high, and touching 100, which catches everyone’s attention, and then slightly lower in the 2nd inning, rather than based on any factual knowledge.

  4. Mike Fast said...

    Ha.  If the seal won’t bark its approval, maybe a Pigeon of Palatability?

    Re the Futures Game, we have three pitchers with a previous PITCHf/x record: Mike Minor, who was throwing 0.3 mph faster in the Futures Game as opposed to previously; Anthony Slama, who was throwing 0.2 mph faster; and Simon Castro, who was throwing 0.4 mph slower.

  5. RZ said...

    Only two of the 22 fastballs he threw had a higher velocity than any other regular season fastball he threw since he came up. He did maintain his velocity around 97-98 in the second inning knowing it was his last inning so he kept the hard fastball going also. So I don’t think the camera system was juiced. Also his curve was just about the same velocity as one he threw two weeks ago against the Twins which was the highest ever recorded curveball from Price by pitch f/x.

  6. Mike Fast said...

    Yeah, Harry, you’re right, I did miss Scheppers. He’s there on my list right underneath Minor.  I’m not sure why my brain skipped over him.

    I have him as 0.5 mph faster in Anaheim than in the AFL + spring training.

  7. Harry Pavlidis said...

    I forgot to split ST from his ASG ….
    Winter 2009 96.9
    Spring 2010 97.1
    ASFG 2010 98.7

    I have him throwing some two-seamers in the Spring/Winter games (96.8 and 96.9)

  8. Mike Fast said...

    Your numbers for Scheppers are probably more accurate than mine, Harry.  For the purposes of my calculation, I lumped four-seamers and two-seamers together in my query.  If you looked individually at a pitcher’s repertoire, you’ll get a more accurate number for that pitcher than I will.

    I also backed out the effect of the drag coefficient, though Surprise generally seems fairly normal to a little low on Cd, which would increase the speed delta a little, if anything.  Maybe not so for the specific games that Scheppers pitched in?

    My numbers have Scheppers throwing a little harder on November 19 and March 4 than he did on Sunday, for example.

  9. Jim C said...

    I think you’re overanalyzing this. I put the extra velocity down to the extra adrenaline involved in pitching in an All-Star game.

  10. Mike Fast said...

    Jim, it’s a given that I’m overanalyzing this.  That’s what fun. 

    There also happen to be practical things we can learn from this, like how much harder a starter can throw in relief if they want to, and whether or not we can trust the PITCHf/x velocities for the young prospects from the Futures Game on Sunday.

    But I think I came to the basically same conclusion you did: most likely the extra speed was due to shorter stints for the starters and extra adrenaline and not due mostly to measurement error.

  11. Harry Pavlidis said...

    Scheppers’ curveballs:
    AFL 81.7
    ST 82.1
    Anaheim 81.5

    Wainwright’s curves were up a good 3 mph (77.4 vs 74.something year by year)

  12. Kyle said...

    Well… Verlander has thrown harder.  The dude has hit 100 more than once in the ninth inning before.  I’ve seen Valverde, Wilson, Jiminez all hit triple digits before.  Price and Johnson have hit 98 that I’ve seen.

    The averages seem a little absurd, but I think it’s like Jim C said.  Adrenaline… Verlander started game one in the World Series back in 2005.  He was horrible, but was throwing 99-100.  He just couldn’t field groundballs(neither could the other Tiger pitchers other than Kenny Rogers), and he wasn’t throwing strikes. 

    I do think they lied about Price, because his pitches didn’t look amped up like Verlander or Johnson.  It showed him at 100, and I thought the pitch was about 96.  Fox one time in the same postseason had Verlander throwing 91-93 in a playoff game(but when they showed whether or not the pitch was a strike, the mph would say 96-98.)  When Zumaya came in the game and his fastest pitch was 98 mph which looked like he blew out his arm after hitting 95 a few times instead of 100, he said they were intentionally slowing down the gun after the game.

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>