Investigating Alex Sanabia’s Pitches

Last night, Alex Sanabia gave up a home run to Domonic Brown in the second inning, and as he was walking back to the mound, the video appears to capture him spitting on the baseball. Sanabia then proceeded to retire 14 more batters without allowing any more runs, and the Marlins beat the Phillies 5-1. Based on the video evidence, it seems as though Sanabia may have been throwing a spitball last night. Based on the fact that he had one of his best starts of the season, by results, it appears as though the spitball might have fairly effective.

Of course, this is all still highly speculative. We only have that one video clip of him spitting that one time, and because the clip is very short, we don’t actually know what happened before he threw the next pitch. Maybe he dried the ball off. Maybe the spit didn’t actually hit the ball, and it’s all a camera angle trick. If we’re going to assume that Sanabia was throwing a loaded baseball, we should investigate a little further.

Thankfully, due to the wonder of PITCHF/x, we can look at how Sanabia’s pitches moved last night relative to how they moved in his other starts the rest of the year. No pitcher is loading up and throwing a spitball on every pitch, and the entire point of putting additional moisture on the ball is to give it irregular movement, so if Sanabia threw a spitball last night, there should theoretically be some evidence of it in the PITCHF/x data.

Here are Sanabia’s game charts from last night.



And here are Sanabia’s same charts for all of his 2013 performances.



Now, let’s overlay the images on top of each other, so you can use your mouseover to see the difference between his seasonal movement charts and the charts from last night.



There’s certainly two weird looking areas on the vertical movement chart, but those are all from the same game – his April 5th start against the Mets. Given that all of those outlier pitches came from the same start, I think we can safely assume that’s just a data error, either with the PITCHF/x calibration that night or the numbers that got stored in the database.

Take away those two blobs, and everything else looks pretty normal. His pitches from last night are in the same general range of movement as they have been in all of his other starts this year.

What if we focus on the pitch immediately after the apparent spitting was recorded? Delmon Young came up to bat against Sanabia after Domonic Brown’s home run, and Sanabia threw him a first pitch fastball, at 88 mph, taken for a ball. Here is the horizontal and vertical movement of that pitch, courtesy of TexasLeaguers.


And here’s all the sinkers that Sanabia threw last night.


And, again, here’s the images overlaid on top of each other.


In terms of vertical movement, it was a totally normal fastball from Sanabia, relative to all the other fastballs he threw last night. In terms of horizontal movement, it is on the edge of all the fastballs he threw, but not dramatically different than the rest.

Of course, that could just mean that Sanabia spit on a few other baseballs, so the lack of an outlier could be due to multiple spitballs making each one look like less of an outlier. Or, it could just be that Sanabia’s spitball didn’t really do much for him after all. While it’s easy to note that he only gave up one run in 6.1 innings, we should also note that he only struck out three batters and allowed a .300 BABIP, so it wasn’t like he was unusually dominating last night.

If we were looking for a smoking gun in the PITCHF/x data, I don’t think there’s one to be found here. MLB might already have the smoking gun considering the video from last night, and in terms of a punishment, it shouldn’t really matter whether Sanabia threw an effective spitball or not, but I do think it’s at least informative to know that his pitches didn’t seem to do anything all that weird last night. If we didn’t have that one second video clip, I don’t think anyone would have thought Sanabia was loading up the baseball otherwise.

But, of course, we do have that video clip, and it sure looks like Sanabia spit on the baseball in that clip. That evidence is probably more compelling than the PITCHF/x data. But, at least from my look at the pitch movement plots, it’s not completely obvious that Sanabia was throwing anything all that different from what he’s thrown all year.

Print This Post

Dave is the Managing Editor of FanGraphs.

21 Responses to “Investigating Alex Sanabia’s Pitches”

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

    Old Hoss Radbourn watched the tape and concluded Sanabia did nothing wrong and several things right.

    +16 Vote -1 Vote +1

  2. Trip says:

    The funny thing about spitballs is that you still have to know how to throw it. You can’t *just* spit on it to get it to move strangely.

    Vote -1 Vote +1

    • BubbaBiscuit says:

      The way he was working it into the ball, he might have been trying to get a better grip instead of a slippery one.

      Vote -1 Vote +1

      • Choo says:

        Probably this. If he worked the spit into the entire ball, and it wasn’t loaded with phlegm, the finished product is no different than wiping the sweat from your forehead or neck and rubbing it into the ball, which just about every pitcher does. And if they don’t do that, they use arm sweat or keep the pocket or a finger of their glove slightly damp – black gloves are good for concealing that.

        However, if Sanabia was working the moisture into one half of the ball while keeping the other half dry, that would have to be viewed as an intent to cheat.

        Vote -1 Vote +1

  3. BubbaBiscuit says:

    This is very good data showing that he didn’t instantly become a better or different pitcher. I kind of wonder about a possible placebo effect, but don’t think it holds up when facing direct competition. The batter doesn’t know, so any increased confidence in himself or his pitches shouldn’t translate into effectiveness, should it?

    Vote -1 Vote +1

  4. All Balls No Brains says:

    I tried spitting on my keyboard to make this a better comment, but it came out about the same as usual. Damn.

    +51 Vote -1 Vote +1

    • David says:

      Incorrect Sir! the large green number below your post indicates that it was far more moving than a normal comment.

      +11 Vote -1 Vote +1

  5. Jeff says:

    The spitball is not only thrown for movement but also in order to “deaden” the ball. Hence the dead ball era where no one could hit it over the wall. A heavier, softer ball is going to travel a shorter distance and contact won’t be as strong if the ball is wet. Sanabia’s PITCHF/x may not vary much(especially if he isn’t a great pitcher), but if he is throwing spitters, the distance the balls travel should be less. This is a much more subtle effect of spitballing and makes it much harder to get caught because the PITCHF/x evidence won’t be there. I would like to see the same overlays with batted ball distance data and see how that correlates. Also would be interesting to see Buchholz’s data over the years as well.

    Vote -1 Vote +1

    • Rob says:

      1) It seems very unlikely to me that his saliva produced enough moisture to “deaden” the ball to any measurable effect.

      2) The “dead ball era” wasn’t a result of pitchers spitting on the baseball. The physical composition of the balls was fundamentally different then than it is today.

      Vote -1 Vote +1

      • Choo says:

        And they would use the same ball for an entire game. Although dunking baseballs in mud, scuffing them, discoloring them with tobacco juice and cramming licorice into the laces didn’t help, either.

        Vote -1 Vote +1

  6. Robin says:

    Do we have any examples of pitchers being caught loading up in the pitch f/x era? It would be interesting to see data similar to what you’ve presented above to see if pitch f/x shows us that spitters move more.

    Vote -1 Vote +1

  7. TKDC says:

    If he wasn’t doing it to cheat, then it is odd, because you can still get in trouble for doing it. The non-cheating reasons don’t seem to add enough value for the chance of getting caught.

    Vote -1 Vote +1

  8. chuckb says:

    Thanks for doing this, Dave. It doesn’t prove that he spit on the ball, nor does it prove that spitting on the ball doesn’t work but it should shut down the argument that the spitting was the reason he only allowed 1 run to the Phillies last night.

    Vote -1 Vote +1

  9. Jason H says:

    Assuming he is a spitballer (as the video suggests), don’t you think he does that every single game? Using his other starts as a negative control makes very little sense.

    Vote -1 Vote +1

    • BubbaBiscuit says:

      This was done to counter the narrative that spitting is what made him hold the Phillies to one run. It shows that the pitches thrown yesterday acted like the same pitches he has thrown all season long to tune of about a 6 FIP. Yes, he spat on a ball. Yes, he held the Phillies to one run. The above data is showing that correlation doesn’t mean causation in this case.

      Vote -1 Vote +1

      • Jason H says:

        The spitter may very well have allowed him to hold the Phillies to one run. …it is quite possible that he is terrible with the spitter (e.g. 6 FIP), but would be even worse without it (e.g. >6 FIP).

        Vote -1 Vote +1

  10. Cus says:

    Spitting on a baseball and throwing a spitball are two very different things. It almost looks like he spat on the ball in disgust. Either way he proceeded to dry the ball using his hands.

    Vote -1 Vote +1

  11. Nathaniel Dawson says:

    As Cus said, a spitball isn’t a ball that a pitcher has spat on. They will use vasoline or some other slick, gooey substance. Spit just doesn’t have enough gooey-ness to do the job properly. I don’t think I’ve ever heard of a pitcher accused of throwing a spitball because they spat on the ball. It’s always something else they’ve put on there.

    I should probably go check out Derek’s book.

    Vote -1 Vote +1

  12. Jason H says:

    I recall reading (in the Yankees book) that Jack Chesbro threw his spitter by licking between his index and middle finger and then holding and throwing the ball like a split fingered fast ball. The lubrication would make the ball leave the hand with no spin and the action would be that of a hard thrown knuckleball. So the movement with this type of pitch has to do with the lack of spin, rather than “loading” up the ball and making it heavy on one side.

    I read this some years ago, and have thought of this as a classic spit ball since. However, the way people write and talk about spitballs does not sound like this pitch at all. Does anyone know how Gaylord Perry threw his spitball, for example? Is it a different pitch? ….or perhaps the author was mistaken about how Chesboro threw the pitch?

    …often, while warming up playing catch, I will throw the Jack Chesbro type spit ball I read about. …it does work, and lead to some startled reactions from unsuspecting catch partners.

    Vote -1 Vote +1

  13. Joe Torre says:

    It’s a judgment call

    Vote -1 Vote +1