Saturday, January 15, 2011


There you are, on your own 35 yard line.  As a courtesy, they bring out the chains, but you already know the answer -- you didn't make it.  4th and 1.  What do you call?  If you're and NFL coach, you probably punt.  Why?  Not because of the expectations.  On 1st and 10, 80% of running plays garner 1 yard or more.  I can believe that number drops on 4th and 1, but you are probably still more likely than not to get your first down.  It's true that turning the ball over on your own 35 can be costly, but I still think you give up too much.

So why do coaches do it?  Because they can read tomorrow's headlines already -- "Idiot turns over ball on own 35 yard line.  What a buffoon."

The shame of being second guessed outweighs statistical expectations.

Back to baseball

It's 0-2.  You look into the catcher, but you already know what he's looking for -- low and outside.  I don't understand this.  I haven't crunched the numbers, but I'm pretty sure I know what I would find for the Sox.  The majority of 0-2 pitches are low and outside.  Why?  Maybe this is the theory:
  • He probably won't hit it
  • He might strike out
  • You can afford to give up a ball
Well, the 1st and 3rd bullets are probably right, but I can't say that I've seen a lot of strike-outs via low and outside on 0-2 pitches.  Why not?  Here's why not.  That's what the batter is looking for.  If that's where the balls goes, you'll get a weak foul (if it's close to a strike) or a ball (if it's not).

I think that pitchers give up too much here.  By avoiding 90% of the strike zone, they raise the batter's ability to stay alive.  They charge themselves an extra pitch -- a pitch they might wish they had back later in the game.  And for what?  Usually for the chance to do it again.  I can understand "low and outside" being the most likely location for an 0-2 pitch.  But if there was even a 50% chance that the 0-2 was going somewhere else, I think you'd see more strikeouts on 0-2, even low and away.

So how do I rate it?

Enough complaining.  How do I rate the outcomes on 0-2?  Here's how, in descending order from pitcher's best to batter's best:

1) Called strike

There will be some debate here, but my basic theory is this: The batter knows that he has to protect the plate on 0-2.  If he thought he could hold off and gets called out, you got him, but good.

2) Swinging strike

You really earned this on the first two pitches.  Now that the batter has to protect the plate, he's a little more at risk from surprise location, speed, etc.  Still, good on you, mate.

3) Weak hit in play

This is also part of what the pitcher earned.  The injured at-bat leads to an easy out.  I'll bet that good strikeout pitchers get more of these than average.  It's part of why K/9 is a useful metric -- good strikeout pitches get better results on non-strike out at bats.  But see the caveat at the end of this post...

4) Ball

This is really more about penalizing the pitcher than rewarding the batter.  I know, I know -- if it was a close pitch and the batter held off, you probably want to praise the batter's eye.  My guess is that, often, the batter was fooled and got lucky.

5) Foul, not straight back

River City's breaking and everybody's shaking.  I love those sequences where a batter takes 3+ pitches, after an 0-2 count, even when the pitcher ultimately wins.  It's a Pyrrhic victory.  For every 2 batters who manage this feat, the pitcher loses one at-bat at the end of the game.

6) Well hit -- in play or straight back

I regard those fouls (straight back) as almost as good as balls hit well into play (see this post for why).  And as a reminder, once the ball is hit into play, the pitcher has precious little to do with the result.  0-2 should create a low probability of this kind of a hit, so the batter really won this one.


I have a few misgivings about #3 (weak hit) vs #5 (weak foul).  Why is a foul so good for the batter and a weak hit not?  Well, my thinking as follows: the foul was really the batter avoiding the strikeout in order to get another chance.  The ball-in-play ends the at bat (even if the lucky batter gets aboard).  The batter shouldn't be trying to earn that hit on 0-2, he should be trying to work towards a safer pitch count.  Allowing the ball to go in play is a poor outcome in my book.  But maybe I'm wrong.  Stay tuned...


Sunday, January 2, 2011

Revisions and Precision which a moment will reverse

In responding to my 0-0 post, Precision Blogger implies an interesting question that deserves its own post:
"Your ordering might penalize a certain kind of pitcher that does well. I think that baseball commentators want to see more of this kind of pitcher: the guy who believes in his stuff. He doesn't throw a lot of balls because he knows that usually, when a batter makes contact, he'll get the out. These pitchers risk some good, solid hits, but it's not that important when the batter makes good contact, because, overall, the pitcher expects to do well. Some aggressive pitchers of this type throw a lot of fastballs and give up more than an average share of home runs. They do well if their team can score, because they battle hard and don't give up big innings."
 As a reminder, my ranking of outcomes on the first pitch is:

  1. Swinging strike
  2. Weakly hit (in play, or foul ball -- regardless of outcome)
  3. Well hit foul (not straight back, perimeter, or outside of the strike zone)
  4. Called strike (perimeter)
  5. Called strike (center of strike zone)
  6. Well hit foul (not straight back, in the strike zone)
  7. Ball (close)
  8. Ball (not so close)
  9. Well hit foul (straight back)
  10. Well hit, in play (regardless of outcome)
With 1 being the the best a pitcher can do and 10 being the worst.  The implied question is "isn't there another class of pitcher that your rankings will under-rate?"

The short answer is yes.  The longer answer is no.  But the deeper answer is maybe. Allow me to explain.


I have admitted this before, but I really can't admit it enough:  I'm not really a student of baseball.  I don't have the time.  Since I became a serious pitch counter, I have only had time to watch Red Sox games (plus playoff games).  It is quite possible that there are classes of pitchers that I just don't know that well because Theo Epstein doesn't like them.  So, I definitely allow for the possibility that I am unfairly decrementing those guys.

I will also add that PB's assessment probably describes a significant percentage of closers who get by on "looks" (velocity) alone.  Both Papelbon and Jenks might well fit that category.  I'll have to think about that more.


The idea that a pitcher can be effective while giving up on strike outs has appeal, but -- on average -- the numbers don't back that up.  In 2010 there were:
  • 165,353 at bats
  • 42,554 of which resulted in hits
  • 34,306 of which resulted in strike outs
So, the average batter batted .257.  But when the average batter didn't strike out, he batted .325.  How good is that?  There are only a handful of guys who batted .325 and didn't make it into the Hall of Fame.  Allowing batters to hit the ball turns the average batter into a Hall of Famer.


PB asks a question with which I have long been fascinated.  Are there any great starting pitchers who live by much quicker at bats but fewer strike outs?  There are two optics I like to use for starting pitchers that are in opposition:
  • Strikeouts per nine innings
  • innings per start.  

Take two pitchers with identical batting average against.  It would seem that the one who relied more on strikeouts would throw more pitches and fewer innings.  I feel like there ought to be some great pitchers who have fewer strike outs per nine innings. 

If you look at the 50 pitchers who put up the highest inning counts in 2010, 29 of them struck out 7 or more per inning. Only 10 struck out 6 or fewer.  The guy at the basement of this list was Mark Buerhle, best known for his perfect game in 2009 (he also came pretty close, with another no-hitter in 2007).  A quick look at his lifetime stats show that Buehrle has never been a huge strikeout guy.  In his abbreviated rookie season he approached 6.5 but only topped 6.0 once in the 10 years subsequent.  Maybe he's that guy.

I'll be watching this more closely.  Strikeouts per 9 is clearly a good indicator of pitcher longevity, but that doesn't mean it's the correct way to evaluate every kind of pitcher.


Saturday, January 1, 2011


Count me among those pleased to see Okajima back in the fold.  Please note -- this is not because I have unrealistic expectations about how he will do next year -- quite the opposite.  Rather, the reasons are twofold.

First of all, if you ever chanced to ask Red Sox management about their plans for left-handed relief pitching in 2011, they would have simply said: "We're really looking forward to seeing what Felix Doubront can do."  Now, while his 2010 debut wasn't terrible (5 earned runs, 11 base runners and 13 strikeouts in 9.2 innings of relief) it wasn't exactly Joba-like (1, 18, 34 in 24).  While I hope to see Doubront continue to blossom, he's hardly the sure thing they need in the bullpen.

Second, just as you shouldn't get too bent out of shape about non-credible awesomeness, you shouldn't get bent out of shape about non-credible stinkage.  It's true that it's hard to find any stretch last year during which Okie looked good -- except maybe September (2, 13, 8, in 13) -- one whole season just isn't credible.  Just as you can't count on a reliever looking good next year when he looked good last year, the opposite is also true.

That said, I do have a major concern.  Okajima doesn't throw very hard -- his fastball rarely hits 90mph.  He does have decent control and movement, but I always kind of believed that his success hinged largely on his crazy delivery, which has to be seen to be believed.  [The video commentary points out that his head is in a funny position.  I always thought the bigger oddity is that looks like his had is curved around the ball to hide it...]

But  my point is this -- if his 2007 success was really just a parlor trick, then you'd expect the league to catch on.  Did they?  Let's look at WHIP, one of my favorite stats for a reliever:

2007: 0.971
2008: 1.161
2009: 1.262
2010: 1.717

Ruh ro.  How about K/9 innings?

2007: 8.2
2008: 8.7
2009: 7.8
2010: 6.5

They might be on to us.

Regardless, I'm glad to see that not-proven-useless lefty is back on the payroll.  I'm not asking Okajima for any guarantees, but I think that he deserves the chance.