THE BOOK--Playing The Percentages In Baseball

Allow me to speak on this subject for all Brewer fans who finally got their Yost removed with the game log for the night before the firing.

G Mota J Werth 8 0 ___ Jayson Werth singled to left (Grounder).
B Shouse C Utley 8 0 1__ Chase Utley sacrificed to pitcher (Bunt Grounder). Jayson Werth advanced to 2B.
B Shouse R Howard 8 1 _2_ Ryan Howard was intentionally walked.
B Shouse P Burrell 8 1 12_ Pat Burrell singled to left (Grounder). Jayson Werth scored. Ryan Howard advanced to 2B.
B Shouse S Victorino 8 1 12_ Shane Victorino homered (Fly). Ryan Howard scored. Eric Bruntlett scored.
B Shouse P Feliz 8 1 ___ Pedro Feliz singled to left (Grounder).
E Gagne C Ruiz 8 1 1__ Carlos Ruiz flied out to right (Fly). Pedro Feliz out at second.

To sum that up;

Brings him in to face Utley, great.
Intentionally walks LEFTY Howard of the .293 OBP vs lefties. 
In order to face Burrell of the .400 OBP vs. lefties who, fortunately, this year has not continued his trend of slugging lefties also.
Faces Victorino with men on who slugs .120 points higher vs. lefties.
Finally gets relieved.

Why did he walk Howard?

Also note that whatever Howard’s true talent is against lefties, it’s even lower with Shouse.  And whatever Burrell’s true talent is against righties it is MUCH higher with Shouse.

(Though I suspect that using the sample data from 2008 as you have done will be a close approximation to what I’m describing above.)

So, again, why was Howard walked?  Was it to set up the DP?  Yost said he read The Book. 

p.290:

If all batters have equal ability, intentionally walking a batter to set up a double play, force, or other situation is at best a breakeven move (or insignificantly better than a break-even move).

To walk a worse hitter to face a better hitter would be even more unusual, unless Burrell hits into even more than Jim Rice amount of DPs.

It looks like Yost made an awful move, but I do have one question in the general sense.

Walking a guy to set up the double play (and thus loading the bases) vs. bringing the infield in with runners on 2nd and 3rd and less than two outs. I realize there are lots of factors here: infield defense, the hitter being walked, the next guy up, speed of runners, GB/FB ratios, etc. But what seems to be the better move here if giving up that 1 run is something you can’t afford to do?

All I saw was the game log so I asked a friend who said “I don’t recall the announcers giving any reasoning.

After the game Yost said he wanted Shouse to pitch the full inning and he views shouse as a guy that can get both righties and lefties out.

I think Ned did not want to burn his bullpen on a doubleheader day, especially two days after he had to go deep due to a short outing from parra.

But that is being generous to ned.”

It also doesn’t explain the BB.  He read the Book?  He made strategic errors with regularity.  Not just experimenting and giving guys a chance to succeed and change their stigma.  Stuff you can’t explain away, like Sunday, in a pennant race, against the team behind you.

We talked about Yost a while ago here about how he read The Book:
http://www.insidethebook.com/ee/index.php/site/comments/it_is_official_a_major_league_manager_claims_that_he_has_read_the_book/

Some of Yost’s much-publicized moves this season have puzzled me, especially since he read The Book. I mean, he left Sabathia in for the ninth in a 7-2 (8-2? I forget) game when throwing pretty much any reliever would’ve won the game (I think The Book showed that with a 3-run lead, the team wins the game 96-98% of the time. I think...) and saved Sabathia another inning of work (though it might not have really been much work for Sabathia anyway).

Reading the Book and actually understanding the concepts and implementing some of them are worlds apart.  Worlds.  I have read lots of technical books in my life about lots of things - 99% of them I don’t even remember reading let alone comprehending.

Looks lke Yost didn’t do a very good job with him, but I think most loogys face nearly as many right-handed batters as lefties.

Three reasons:

1) Other managers alternating left-right
2) Pinch hitters
3) Switch hitters

Right, that’s why I brought up Orosco’s last 3 years…

I was about to say the same thing as Rally.  I don’t know whether Yost did a good job with Shouse or not, but I generally use 50/50 as the ratio of RH to LHB that most LOGGY’s face for the reasons Rally mentions (switch hitters, pinch hitters, and when facing L/R/L, as well as “garbage time” when it does not really matter). 

OTOH, you should expect that a pitcher like Shouse with an especially large platoon ratio should be on the low end in terms of the percentage of righties faced. For example, if LOOGY’s face anywhere from 40-60% righties (and I don’t know that they do), Shouse should be at 40%, and not 50%.

There’s another reason a LOOGY might face a balance of lefties and righties over the course of a season. Even if a manager tries to maximize his splits by having him face lefties exclusively in high leverage situations, there are still going to be times when he’ll be used in low leverage situations, either to eat innings or simply to get some work in to stay sharp. In those game he’ll almost certainly pitch a full inning regardless of who is due up, and that means going against lots of right-handers.

Yost’s ‘reasoning’ for walking Howard was that Howard was ‘hot’.

But you should also know that Yost is an idiot that explained earlier this year his reasoning for not using a hit and run with runners on first and second(both with above average speed who probably could have executed a double steal successfully) with no outs and down two was fear of running into a triple play.

Yost walked Howard because he’s hot and to set up the double play. He relies very heavily on Shouse to get a ground ball. Here are his comments about walking Howard.

http://blogs.jsonline.com/brewers/archive/2008/09/14/why-yost-stuck-with-shouse.aspx

“At that point, he’s so hot, you’ve got to (walk him),” said Yost. “You bring in ‘Shousie’ to face him but I couldn’t bring myself to do it.”

“I didn’t want him to have to finish the inning and send him back out in the ninth,” said Yost. “I wanted him to have a clean inning.”

A better question than “Why did Yost walk Howard?” is “Why is it important for Yost to get his relievers a ‘clean’ inning?” Because that’s the managing philosophy that leads Yost to make decisions like setting up a double play instead of taking the platoon advantage.

With data from FanGraphs, for Brian Shouse in 2008, in LI > 1.5, with IBB excluded:

57.6% BF = LHB
17.0% BF = RHB (includes switch hitters) PH for LHB or P

Non-PH RHB faced, with a few exceptions, were not the best of hitters.

Google Docs link:
http://spreadsheets.google.com/pub?key=p2bKuc1_nKeUIxggXyxUYRA

Well, that is fantastic stuff!  Funny how looking at the overall numbers, coupled with the Howard fiasco clouded my perception.

FWIW, the “hot” business was also in The Book.

“I didn’t want him to have to finish the inning and send him back out in the ninth,” said Yost. “I wanted him to have a clean inning.”

Who, Shouse?  I don’t get it.  The correct thing to do in that situation is have him pitch to Howard and then bring in a RHP for Burrell.

MGL/#16

Sorry, i clipped the quotes, Yost was talking about Gagne who was up in the pen and reportedly ready to face Burrell.

But, as Yost said, he didn’t want gagne to have to finish the inning and start the next.

Of course not.  It’s more important to get shelled first, and then have a clean inning.

Sometimes, a manager thinks too far ahead, and this is one of those times.  And, in SEPTEMBER of all times, when you must have 15 guys in the pen, there’s no reason to think too far ahead.

Well, I am pretty sure that this one “boner” did not cost much in wp.  It is probably just one example of poor bullpen management by Yost.  I don’t know that for sure (that Yost makes more than his fair share of bullpen mistakes).  I am only going by what other people have said (which certainly could be misguided).  And there are plenty of other managers who make lots of poor bullpen decisions as well.  Probably most of them in fact.

I don’t think it matters one way or another who manages a team for 15 games. No one manager can possibly be more than like a 1/4 or 1/2 of a win different than another manager in only 15 games.  Then again, if a lot of the Brewers did not like Yost, then maybe firing him at any time would be a motivator for the team.  Who knows?  And who knows how much motivation (or lack of it) is worth to a team in wins?  I sure don’t.

mgl, I think you are right - but look at it this way - What are the odds that Yost will cost the team one or more games out of the 15 remaining, plus any possible post-season games, as oppsed to Sveum? 5% for Yost? 1% for Sveum? (for example)

Management sees this as one loss can cost the pennant territory.

Honestly, I doubt that there is any functional difference between the two managers in 15 games, but who knows. I don’t think anyone has any idea.  As far as “costing a game,” that does not really mean anything.  Any decision, good or bad, correct or incorrect, can “cost” a game in retrospect.  Had Shouse gotten Burrell to ground into a DP, he would have been “credited” with “giving them a win” or at least a chance to win, which is just as silly as a manager “costing a win.” All a manager can do is increase or decrease a team’s wp at a certain point in the game, and generally it is on the order of less than .1 wins.  A lot less usually.