THE BOOK--Playing The Percentages In Baseball

Dave has been hammering on Felix’s first inning problems for at least a year now and I’m not sympathetic to the “establish the fastball” dogma. However, Felix keeps getting hit hard throughout the game, well after he starts mixing his pitches:

3rd inning- .321/.356/.429
4th inning- .312/.377/.500
6th inning- .351/.400/.676!

If you just read Cameron’s posts, you’d think Felix was getting destroyed in the first and then cruising the rest of the way. That’s certainly not the case, so clearly things are more complicated than the mere proportion of fastballs he throws to the first 4 or 5 hitters in a game.

Now, he did have a good outing against the A’s after reading Dave’s open letter, but he only struck out two guys so I’m not convinced he’s actually turned a corner. More likely he’s going to continue to disappoint until he further refines his approach or he regains that wicked two-seamer he had for the first two games of the year (before he got hurt).

Of course this and related issues are fascinating and much under-studied and little understood.

First we have to compare Felix’ differential between his overall fastball % and first inning (or first few innings) fastball% with other similar pitchers.  If all or most other pitchers with similar pitches to Felix and similar overall percentages throw a lot more fastballs in the first few innings, then they are all probably doing the right thing (maybe) and Felix is no different.  Just knowing the 59% and the 84% tells us nothing.

That being said, I have wondered why pitchers do in fact throw more fastballs early in the game and if it is “correct.” Generally anything you make more predictable is NOT good.  The only reason it might be correct if it is necessary for the pitcher for some reason (like he does not have command of his other pitches until he throws some pitches) or if batters cannot hit fastballs early in the game as well as late in the game.

The whole idea of adjusting (other than adjusting after you see or get a scouting report on a player OR a player simply learning and improving overall) or setting up batters is bogus, at least to an intelligent batter.  But I don’t want to get into that right now as it is a complicated issue, hard to explain, and it is too late in the evening.  I wil give an example.  We hear all the time about a pitcher coming high and inside with a fastball on a pitcher’s count in order to “set up” the hitter for the low and away off-speed pitch.  That is ridiculous of course.  What good is that if the batter also knows that is what the pitcher is going to do?  ALL pitches have to be thrown ramdomly, tempered by a batter’s weakness and by his lack of an optimal approach.

I feel like “setting up” doesn’t mean getting someone to guess wrong.  I think it means setting up one’s perception.  Just like swinging two bats makes one feel like a toothpick, I thought the idea was that a high and inside pitch, followed by a low and away pitch, will make the low and away pitch seem REALLY low and away, even though it’s not.

If you just read Cameron’s posts, you’d think Felix was getting destroyed in the first and then cruising the rest of the way. That’s certainly not the case, so clearly things are more complicated than the mere proportion of fastballs he throws to the first 4 or 5 hitters in a game.

I’ve never tried to give the impression that pitch selection was a panacaea that would solve all his problems.  In the open letter to Chaves, I explicly acknowledged that he struggles to command his fastball, and ended the post by stating that mixing his pitches wouldn’t turn him into a Cy Young winner.  I just happen to think that pitch selection is a very easy fix, while fastball command is much harder to rectify. 

Now, he did have a good outing against the A’s after reading Dave’s open letter, but he only struck out two guys so I’m not convinced he’s actually turned a corner.

I’m sure you’re aware of the fact that Felix’s problem hasn’t been a lack of strikeouts, but instead a ridiculously high batting average on balls in play and an inflated HR/F rate.  While I’m certainly not saying that the Oakland start proved that Felix is ready to dominate, I do think that a varied pitch selection will help him regress towards more normal performance in these areas.  And, really, if you combine a normal BABIP and HR/F with Felix’s peripherals, you have a Cy Young contender. 

First we have to compare Felix’ differential between his overall fastball % and first inning (or first few innings) fastball% with other similar pitchers.  If all or most other pitchers with similar pitches to Felix and similar overall percentages throw a lot more fastballs in the first few innings, then they are all probably doing the right thing (maybe) and Felix is no different.  Just knowing the 59% and the 84% tells us nothing.

I actually tried to do this, but since the Pitch F/X system is only installed in nine parks to date, it was tough to find a good comparison with a large enough sample to study.  John Lackey, for instance, throws a very high percentage of fastballs the entire game, and I found games where he starts off with a sequence of 10-15 consecutive fastballs before throwing something off-speed, but Lackey’s command is so much better than Felix’s that it doesn’t seem to be a problem. 

We hear all the time about a pitcher coming high and inside with a fastball on a pitcher’s count in order to “set up” the hitter for the low and away off-speed pitch.  That is ridiculous of course.  What good is that if the batter also knows that is what the pitcher is going to do?  ALL pitches have to be thrown ramdomly, tempered by a batter’s weakness and by his lack of an optimal approach.

While I agree that pitching in any set pattern is a bad idea, such as high-and-in then low-and-away, I’m not so quick to dismiss that there may be a potential benefit for the hitter of seeing similar pitch types/location in repitition.  If you think of going to a batting cage, for instance, I think everyone can understand the feeling of getting more comfortable with the timing and angle of the pitches after getting a few cuts in, knowing that the next pitch is going to be in a similar position to the last one.

That’s essentially what Felix was for the first ten piches of each game - a batting cage.  If you didn’t get a good hack at his first or second fastball, don’t worry, you’ll get another chance at it.

What happens if you take the best pitcher in baseball and give him only one pitch to throw?

Remember the all-star game last year? Brad Penny started for the NL, pitched two innings, and, I believe, threw nothing but his fastball. He struck out the first three hitters (including Ichiro), but allowed a solo homer to Vlad in the 2nd.

I just happen to think that pitch selection is a very easy fix, while fastball command is much harder to rectify.

It’s not simply a matter of how easy a fix it is, you indeed think it’s a major (if not the biggest) problem. I don’t think mixing his pitches will make a big difference, not only for the reasons explained in my first post, but because the data suggest that his main issues have a different root. First, he is getting crushed by lefties (.878 OPS compared to .624 vs. righties). Second, he pitches well when the bases are empty but does much worse once someone gets on (.691 OPS and .829 OPS respectively).

I don’t have a good explanation for the wide platoon split other than some people have argued that his changeup isn’t as good as advertised (I have no idea if that’s true or not). As for his struggles with men on, it’s seem that this would be caused by pitching from the stretch. If that’s the case, then it would be an even easier fix than changing his pitching pattern.

Since the purpose of pitching from the stretch is to discourage baserunners from stealing, but stealing isn’t always possible when guys are on base, I decided to break the numbers down further.

Can’t steal (---,--3,-23,123)
.254/.308/.371

Can steal (1--,-2-,12-,1-3)
.327/.359/.529

The gap is even bigger than the bases empty/runner on split. So if Felix is pitching from the windup in all the “Can’t steal” situations and from the stretch in all the “Can steal” spots, this is further evidence of a potential super simple fix.

It’s not simply a matter of how easy a fix it is, you indeed think it’s a major (if not the biggest) problem.

One of the worst things you can do in a conversation is tell someone else what they think. 

I’ve never stated that a different pitch selection would fix all of Felix’s problems.  I’ve explained, in detail, his problems pitching against lefties, with runners on, and when behind in the count.  I’ve talked about his fastball command, or lack thereof, as well as the noticeable decrease in velocity and movement since lowering his arm angle after coming off the disabled list earlier this year. 

Felix’s early-game pitch selection was a problem.  It wasn’t the only problem, but it was a problem.

One of the worst things you can do in a conversation is tell someone else what they think.

Sorry, then I phrased that badly. Let me restate it: you’re continued harping on Felix’s over-reliance on his fastball to begin each game is highly suggestive of how significant you believe it to be contributing to his overall disappointing performance. Therefore, I believe you were being disingenuous when you claimed the only reason you harp on it is because it is an easy fix.

If the time you spend suggesting different changes in his approach is proportional to how easy it would be to implement each change, then why haven’t you spent more time talking about his issues pitching from the stretch? It’s seems that telling him to use the same motion (his preferred motion at that) for every batter would be a lot simpler than telling him to outsmart the batter, even if the latter is itself really easy.

I’ve never stated that a different pitch selection would fix all of Felix’s problems.

And I never claimed you believed that. My point is simply that you are overstating how helpful improved pitch selection in the first inning will be.

Let me restate it: you’re continued harping on Felix’s over-reliance on his fastball to begin each game is highly suggestive of how significant you believe it to be contributing to his overall disappointing performance.

No, it’s not. 

I’ve spent a lot of time painting my house recently.  A lot of time.  I do not believe that the color of my walls is important to the structural integrity of my house - I just believed that I had the ability to improve the way my house looked. 

I don’t know how to fix Felix’s command, so, naturally, I’m not going to spend a lot of time pointing out that I don’t know how to fix Felix’s command.  I believe I found a problem with Felix, and through a lot of work, I think I also found a potential solution. 

Naturally, I spent a lot of effort and attention on the thing I thought I could help. 

Therefore, I believe you were being disingenuous when you claimed the only reason you harp on it is because it is an easy fix.

Thanks for the personal shot at my character, guy-who-I’ve-never-talked-to-before.  It’s always nice to know when it’s no longer worth having a conversation with someone.

I’ve spent a lot of time painting my house recently.  A lot of time.  I do not believe that the color of my walls is important to the structural integrity of my house - I just believed that I had the ability to improve the way my house looked.

Since you didn’t paint your home to improve it’s strength but to improve it’s looks, your analogy proves my point. You wanted your house to look better, and the quality and color of the paint job is important in achieving this, so the time invested in the new coat of paint reflects that importance.

I don’t know how to fix Felix’s command, so, naturally, I’m not going to spend a lot of time pointing out that I don’t know how to fix Felix’s command.

In your first post you only mention pitch selection and general command as a problem. When I pointed out that there’s more to it than that, you said that you were aware of a whole bunch of problems. Now you’ve gone back to only talking about those first two things as if they are the only issues Felix faces. Again, his windup/stretch splits suggest not only another issue that can be readily solved, but hint at his command problems being more situational than universal. If you are aware of this, and agree that it would be easy to fix, why don’t you harp on it on a seemingly weekly basis and write an open letter to Rafael Chaves about it? If you don’t think it’s an easy fix, just say so.

Thanks for the personal shot at my character, guy-who-I’ve-never-talked-to-before.

Pointing out a specific case of dodginess is not an attack on your character. If someone told you that you got a single fact wrong in a post, would you take that as an accusation that you get all your facts wrong and are not trustworthy? I hope not.

It’s always nice to know when it’s no longer worth having a conversation with someone.

Good day to you too, sir.

So Brad Penny once again did his fastballs-only thing in the All-Star game this year.

Here’s the poster child for 1st inning problems (click on my name).

Telling someone that “I believe you were being disingenuous” can be construed as a personal attack.  Unless you really know the guy (MGL has called me disingenuous) and the context permits it, it’s best not to call someone that.

I appreciate Aaron’s viewpoint, especially on the stretch/windup issue (which Andy researched in The Book as a true skill).  Rather than focusing on Dave’s blog entry as if it were a thesis, it’s best to focus on the points made, rather than trying to infer Dave’s position on his points not made.

Dave’s point is legitimate in-and-of itself, and deserves further research. Another example of a pitcher who tries to establish something (strike zone in the case I’m about to point out) are the Braves pitchers (Glavine, Maddux, Smoltz).  They have tons of walks in the first inning (more than expected considering you are facing top players in the first inning).  I looked at other top pitchers (Clemens, Schilling, Pedro, Mussina, RJ, Pettitte), and they don’t show those kinds of splits.

I think it’s fascinating discussion when it comes to pitchers, regarding pitch selection, pitch location, and pitching mechanics.

Let’s trick to stick to the merits, rather than the motivation behind the merits.

I called Tango disingenuous?? 

Anyway, as is the case with so many words, many people misconstrue its meaning.  It is not really used to indicate that someone is “lying” which is often what people take it to mean especially when it is directed at them.  It usually means “lacking in sincerity” such as a “disingenuous apology.”

Anyway, I agree that these are complex and fascinating issues and are to some extent the next frontier in sabermetrics.

http://www.baseballthinkfactory.org/files/primate_studies/discussion/eric_enders_2002-10-01_0/

24. MGL Posted: October 06, 2002 at 08:53 PM (#606595)

Tango, your arguments in favor of making sure that runs add up is disengenuous! Because people like it? Because it gives them confidence that your methodology is sound? So I should purposely bastardize my work so people will “like” my methodology? Please!

I think the reason that I remembered it is not that I took offense, but that I like the word itself!

On further thought, I realize I was misinterpreting Dave’s reason for consistently posting on the subject of Felix’s pitch patterns. Once someone notices a flaw somewhere, the tendency is to keep focusing in on it. And if that flaw could be easily fixed but goes uncorrected, it can become maddening even if the problem isn’t a huge deal in the big picture. That’s what Dave was trying to say and he was being sincere so my impression was incorrect. Therefore, I was wrong to call him disingenuous, and for that I apologize.