THE BOOK--Playing The Percentages In Baseball

Saturday, October 24, 2009

Deconstructing Joe Sheehan’s analysis of Game 6 of the ALCS

You might have missed John Lackey getting hosed on a 3-2 strike to Jorge Posada that Fieldin Culbreth turned into a ball.

He didn’t get “hosed”.  According to some of the pitch f/x guys, Culbreth usually calls that pitch a ball and umpires in general call that pitcher a ball 80% of the time. How is that a “hose?” Disclaimer:  I don’t quite believe the “80% ball” thing for that pitch.  I would have guessed that it was less than 50%.  But a minor point.

It was at this point that Mike Scioscia moved into one of the most puzzling sequences of his long career. With Darren Oliver up in the bullpen and the left-handed Johnny Damon at the plate, Scioscia let Lackey stay in the game.

Joe is 100% correct.  Oliver is an excellent reliever.  Against a LHB, he is even more excellent.  Much better than Lackey even in inning 1.  In inning 7 (when Lackey is on the 4th time through the order), Oliver is infinitely better versus a LHB than Lackey.

Not bringing in Oliver to face Damon is defensible from the standpoint of a typical manager, or even a fan or commentator, but after he gets Damon out he then brings in Oliver, which makes no sense at all of course.  This may be the only thing that Joe gets right in his analysis, although this one is pretty obvious as a head-scratcher.

This wasn’t a problem—Lackey was pitching well up to the Jeter walk, and he’s Scioscia’s best pitcher by any measure.

Can we PLEASE (please, please, please) never use the words, “He was pitching well or not pitching well” in order to explain or justify taking someone out or leaving them in?  Please, please, please.  If you are a “sabermetric writer,” you should never, ever, ever utter those words!  Even if it were true that pitchers who are “pitching well” for a few innings, pitch well for the next few innings and vice versa, the person who utters those words usually has NO IDEA whether the pitcher was “pitching well” or not.  What constitutes pitching well and for what time period?  The score?  The number of K.  The number of hits?  Walks?  HR?  What if a pitcher lets up 3 walks and 7 hits in 5 innings, but no runs?  What if he allows a 3-2 walk on a close pitch, a bloop hit, and then makes ONE mistake that is hit for a HR, all in 4 innings of work?  What if, like Burnette, you allow 4 runs before recording an out in the first inning and then throw 5 shutout innings after that?  Are you “pitching well?” At what point were you “pitching well?” What about after the 4th inning?  Was he “pitching well?” The whole idea of a pitcher “pitching well” or not based on the score is NONSENSE!  And PLEASE don’t tell me that managers look at other things and not the score to determine whether pitchers stay in or come out and to determine whether a pitcher has been “pitching well” or not. They don’t!  If a pitcher gives up 6 line drive outs and 3 warning track fly balls but is pitching a shutout, the manager and everyone else in the world (not me or my cousin in Deluth) considers the pitcher to have been “pitching well.” But if he gives up 3 walks on close pitches, 2 bloop hits, and 2 HR, one on a bad pitch and one on a good pitch, in 4 innings of work, he is considered “pitching badly” because he has given up 5 runs (or whatever), and is likely coming out of the game.

Anyway, enough of that.  It just makes my blood boil and my head spin!

Oh, I forgot about this part:

and he’s Scioscia’s best pitcher by any measure

Yes, Joe, he is probably the Angels best starting pitcher.  He is a very good, but not great, starter.  He is not, by any stretch of the imagination, the Angels best pitcher though!  Is there anyone who knows what the word sabermetrics means that does not yet know that a starting pitcher the 4th time through the order is considerably worse than he is overall or when the game starts?  ALL of the pitchers in Scioscia’s bullpen, Joe, are better pitchers than Lackey in the 7th inning! Please repeat that or write it on a chalkboard 200 times!

Lackey was pitching well save for the walk to Jeter, from which he’d bounced back. He really should have been out of the inning. Oliver has had an effective season, but there was no reason to use him against Mark Teixeira, a switch-hitter with no platoon split, with Alex Rodriguez behind him. Scioscia downgraded as far as the pitcher he’d have on the mound, for no tactical gain, at the biggest moment of the game.

More of the same nonsense about Lackey pitching well (and I don’t understand the “save for the walk to Jeter” - how about “save for all the Yankees who got hits or walks in the first 7 innings so far") and more nonsense about Lackey being the best pitcher that the Angels have available at that time.

Four minutes later, Scioscia was presiding over a tied game following a double, an intentional walk, and a single. With Robinson Cano due up and the go-ahead runs on base, Scioscia went to the mound and hooked Oliver. Oliver hadn’t thrown enough pitches, in my opinion, to reach a conclusion on his stuff—and quite frankly, Hideki Matsui hit a pretty good pitch up the middle—and bringing in the righty for Cano, who doesn’t have a big platoon split save for his contact rate, which is much worse against lefties—seemed rash.

OK, Joe is right on here again.  Taking out Oliver for a RHP versus Cano makes no sense at all.  (And neither does Joe’s statement, “Oliver hadn’t thrown enough pitches, in my opinion, to reach a conclusion on his stuff” - does Joe really think that is how a manager should making pitching personnel decisions?)

I’ve seen some criticism of the decision to allow A.J. Burnett to start the seventh inning with a two-run lead. I spend a lot of time criticizing managers, and many decisions really do have a right and a wrong. In this case, I don’t think using Burnett was a problem, nor do I think starting the inning with Phil Hughes would have been a problem. Burnett had been lights-out since Morales’ first-inning single, and there was no tactical reason to take him out the game at the start of the seventh.

Again, with the “Burnett had been pitching well since the first inning.” As if the first inning doesn’t count.  This is a perfect example of the nonsense of pitching well or pitching badly. He was pitching the same all game - first inning, second inning, etc.  Sometimes you get hit, sometimes you don’t. Sometimes you get walks, sometimes you don’t.  Sometimes a batter squares up a mistake pitch and sometimes he doesn’t.  Sometimes a line drive goes in the gap and sometimes it is hit right to a fielder. Sometimes a ground ball sneaks through the hole and sometimes it goes right to the SS for a 6-4-3 DP.  Sometimes a 3-2 pitch nicks the corner and the umpire calls it a ball and sometimes he calls it a strike.  There are hundreds of ways that a pitcher can “pitch well” or “pitch badly”, give up some runs or none at all, and NONE of those hundreds of things have anything to do with the talent of the pitcher.  Trying to find out how “well” a pitcher is pitching based on the score, hits, WHIP, or anything else like that is like trying to find a needle in a haystack - can’t be done.  And even if you could (make a fair and objective evaluation of how a pitcher is pitching - whatever that even means), we still have NO evidence that it has any predictive value over and above the pitcher’s long-term true talent estimate or projection!

Anyway, here is the part that Joe gets completely, 100% wrong, and again should be obvious to any sabermetric writer worth his pen’s weight in salt:

there was no tactical reason to take him out the game at the start of the seventh.

Joe, the reason is that Phil Hughes is a MUCH better pitcher (for one or two innings) than Burnett right off the bat.  He is 1000 times (OK, not literally) better after 6 innings!  Burnett is probably a 4.25 pitcher in the 7th and Hughes is around a 3.25 at worst, maybe a 3.00.  That is a BIG difference folks.  This is the clinching game of an ALCS.  You have almost a week to rest your bullpen.  You bring in Hughes for the 7th, and Mario for the 8th and 9th, or something like that! It is a no-brainer.  You should probably have brought in Joba in the 6th.  Joba in the 7th would have even been fine.  He is a lot better than AJ the 4th time through the order!

I can’t figure what sequence of events Girardi foresaw, up two runs in the seventh, in which he would have preferred Chamberlain to Hughes.

Well, I think Joe was thinking Joba in the 7th, Hughes in the 8th, and then Rivera in the 9th or in the 8th with 1 our 2 outs or if Hughes gets in trouble.  I don’t have any problem with that really.

While Hughes was throwing, Burnett walked Erick Aybar, which is a pretty clear sign that you’re done.

“A pretty clear sign that you are done?” That almost made my head explode.

Damaso Marte came in to face Figgins, who greeted him warmly with a ridiculous sacrifice bunt on the first pitch. I’m pretty sure that giving Marte an out is tax-deductible; doing it from the right side may qualify you to have your student loans canceled. Marte got Abreu to ground out, with a run scoring, at which point Girardi finally got Hughes into the game.

As most of you know, I am an advocate of sometimes bunting and sometimes not to keep the defense from playing too far in or too far back.  I am also an advocate of making the defense pay if they are not playing in an optimal position, given the batter and the situation.  I don’t know why Joe thinks it is “ridiculous” to bunt here.  I don’t know how often it would be correct to bunt or not (my guess would be 50/50 unless the defense were playing sub-optimally).  And I have no idea what Joe is talking about with Figgins batting RH.  You would be MUCH more likely to bunt with a batter batting RH, and Figgins batting RH especially.  Figgins is a worse hitter RH I think.  A RH hitter hits into more DP.  And a RH hitter moves the runners over less often when he is not bunting.  It is the left-handed hitter who should be less likely to bunt.  They will move the runners over more often on a non-bunt grounder to first or second or a deep fly ball to RF or RCF.  And they will stay out of the GDP more often.

o in the seventh inning of a playoff game, with Vladimir Guerrero at the plate, what I don’t think we should be seeing is Hughes shaking off Posada.

Where does he come up with that?  Pitchers can and should throw what they want to throw. All pitchers do that, young and old. And how does he know what Posada wanted and what Hughes wanted?  He doesn’t.  You didn’t see them arguing when Posada went to the mound.

Guerrero beat Andy Pettitte’s fastball—a cutter, fine—Tuesday, and he beat Burnett’s in the first inning Thursday, and he’d just swung and missed at a big curveball 15 seconds prior to fall behind 1-2, and he may have displayed an occasional tendency to swing and miss at balls way out of the strike zone. Hughes decided to announce his presence with authority at the wrong time against the wrong hitter in most definitely the wrong location.

Yeah, we heard McCarver go on for 5 minutes about how that was the wrong pitch to throw - that he should have thrown the curve ball again.  Well, McCarver is an idiot also.  Hughes has to mix up his pitches, like all pitchers at all times, so we cannot say what single pitch he SHOULD have thrown!  We can only guess what percentage of the time he should throw each pitch, and then he and Posada flip a mental coin.  If McCarver or Joe Sheehan thinks that the correct strategy is to throw a curve ball 100% of the time, I definitely don’t want them catching or pitching for my team.  Do they think it would be OK for them to also say out loud, “Hey Vladdy, I’m going to throw you a curve ball in the dirt!  That OK with you?”

In that situation, Hughes was trying to throw a shoulder high fastball a little inside which was a perfectly acceptable pitch choice.  He just didn’t execute it very well, needless to say. He could have just as easily thrown another curve ball. Or a fastball away. Or whatever other pitch he has in his arsenal and given the batter and the situation.  And what if he hung a curve ball and Vladdy got a single or a HR?  Would McCarver and Sheehan find a way to criticize him for that pitch? And what if he threw a curve ball in the dirt for a wild pitch?  That possibility was obviously part of the equation.

About Fuentes:

He’s not an effective pitcher right now, and if he’s asked to protect a small lead against big hitters this weekend, it may not go well for the Angels.

My head is spinning again.

There might be an argument for running Brett Gardner out there for a day, or maybe Jerry Hairston Jr., given that it’s Joe Saunders starting Game Six. Guys like Swisher—Three True Outcome players—bring a lot to the table, but sometimes you just need a single. Hairston was a better choice for that at-bat last night, and might be the better choice for six innings or so this weekend.

Hairston is not a bad player, especially with a LHP on the mound, so I am not going to argue who is the better choice here - I don’t really know off the top of my head.  But Swisher is the same player he was 3 weeks ago or 2 months agao.  We show that in The Book and we write about that ALL THE TIME.  Joe, are you listening?  Have you even read The Book?

Not to blow my own horn or anything, but I can’t imagine in a million years a sabermetric writer not reading The Book and practically memorizing it. Seriously.

OK, my blood pressure is already through the roof…