THE BOOK--Playing The Percentages In Baseball

There’s an article at The Heater about Maddon’s game plan.  He will start the inning with Balfour, but may leave him in to try for a 1-2-3 inning.  If he brings in a lefty, and the Rays get on base in the seventh, he’ll lose the lefty to a pinch hitter, and he may not want to do that.

If he brings in a lefty, and the Rays get on base in the seventh, he’ll lose the lefty to a pinch hitter, and he may not want to do that.

Hmmm…

If PHI goes 1-2-3 or 1-2-3-4 (leaving Utley or Howard on deck) AND the Rays get someone on base, he loses the lefty but has 2 more lefties in the pen.

If I were him, that would NOT be much of a consideration.

2) I have NO doubt that Maddon will screw something up.  I am willing to lay 2-1 on that if there are any takers...

Didn’t take too long.  Top of the 7th, runner on 1st and one out, he lets Howell hit (bunts him of course).

A 1 out bunt is worthless.  It is equivalent to a bad-hitting pitcher.

There is NOT (as in zero) doubt that a pinch hitter is the correct play, and by a lot. Probably the worst (of many bad ones) decision of the post-season, and it’s not like it is a tough one to make.

Despite the way Maddon speaks (very intelligently) and the media portray him, he is, without a doubt in my mind, completely clueless when it comes to managing an important game.  And I’d say that to his face.  There is no telling how much in WP he has given away this post-season.

And what do you gain by not pinch hitting for Howell?  You let the lefty face a righthanded power hitter.  Then take him out after one batter.

Bradford (or Wheeler) should have started the 8th inning against Burrell regardless of whether the pitcher’s spot came up.  Truly a terrible (non)-move

One move like that never really matters.  For example, take a great pitcher, someone who is a .650 pitcher.  That means he adds some +.150 wins per 9 innings while facing say 37 or 38 batters per 9 IP.

Per batter, that adds +.004 wins.  And we’re talking about an extreme move.

Comparatively, a great hitter is worth some +5 wins per 625 PA over an average hitter, or +.008 wins per PA.

Realistically, the moves being considered is probably worth .001, maybe .002 wins.

A bad pitch call by the ump, turning a ball into a strike or vice-versa, is worth around .15 runs or .015 wins.  In terms of venom toward single reactions, that is the far most costly.  A single bad ball/strike call is 10x worse than a boneheaded roster move.

So, while I agree it is completely stupid to not dip into a bullpen that doesn’t require resting at this point (or other moves), the horribleness shows insight into the manager, rather than that specific move or non-move being so costly.

That said, I agree with you guys.

4. When the pitcher came up to bat/bunt, I yelled at the TV, but quieted down atleast a little bit when he was able to get the sac bunt down.  When the lefty then stayed in to face Burrell, I yelled at the TV to bring in a RHP, Burrell mashes against LHP.  To me, this move may have been the one that cost TB the game.  Second on my list would be the runner trying to score from 2nd baes on the GB to Utley.  He should’ve atleast crashed the catcher.  I know Erstad would’ve.
vr, Xei

Are we talking about the same thing?

If you’re telling me that a random bad strike/ball call has a bigger effect than having the platoon advantage for a hitter, I can buy that, but 15x as much?  Surprising.

And I can’t believe that the difference between a pitcher hitting and Hinske or Aybar as a PH isn’t more than that.

I want to ask about playing the IF in.  Last night’s game provided an intersting result on a play where the Rays had drawn in their infield, only to see Werth bloop one into shallow CF.

Now, Werth is only a career 38% GB batter (39% in ‘08), which appears to be at least ‘below-league-average’ if not a tad lower.  Is the chance at turning a grounder into an out at the plate really worth taking away the chances at other batted balls?  A blooper is just one way in which a drawn-in IF can be beaten on a play it could otherwise make.

Is this just another ‘old baseball’ tactic that will never die, or is there more merit to having the IF drawn in there?  Geoff Jenkins is about as slow a runner as you could have at 3B, to boot.

Looking through the splits on B-Ref, depending on the count, a bad strike/ball call could make a 200-400 OPS difference.  So that’s 5-10 times greater than the impact of having the platoon advantage on Burrell.

The OPS difference between Hinske/Aybar, even with PH penalty, over Howell has got to be more than that.

A pitcher’s wOBA is around .165 or so.  An average hitter, with the PH penalty, will probably be down to .315 or so.  So, we are talking about a .150 difference in wOBA, which is .13 runs per PA, or .013 wins per PA.

So, good point about that.

That impact is the same as one bad ball/strike call.

To translate OPS into wOBA, you divide by about 2.25.  So, a .300 OPS point difference is .133 in wOBA.  You divide that by 1.15 to get the run difference.  So, that’s .116 runs per PA, or .012 wins per PA.

So, your numbers are showing that the impact of a bad ball/strike call is the same as switching between a pitcher-batting and a PH.

Realistically, the moves being considered is probably worth .001, maybe .002 wins.

If you read my analysis on the other thread, the bunting Howell versus a pinch hitter hitting (around an average batter) lowers TB WE by over 7% or .07 wins!  That is huge!

Agreed, but if we assume that umpires are unbiased though human and subject to occasional mistakes, it is something beyond the team’s, or manager’s control.

The decision to bat Howell and use him to pitch to one righty is within his control, and a pretty terrible percentage decision.  In the range of options that Maddon had available to him, that was about the worst he could have made.

Actually, taking Howell out and using Bradford to pitch to Victorino does not look good either.  He’s a switch hitter and I don’t see that much quality difference between Howell and Bradford, I think either is equally likely to get him out.  But if you get him out, with a runner on 2nd and nobody out, what is the worst kind of out you can get? 

Groundball to the right side, which is what happened.  It didn’t have to happen that way but forcing Victorino to bat righty and having a groundball pitcher sure tipped the odds in that favor.  Howell would have been more likely to get a groundball to the left side, and keep the runner on second.

Not a huge mistake compared to the PH/Burrell decisions, but another bad one nonetheless.

Obviously, higher leverage situations will be worth more.  If the LI was 3, then a .013 win difference would become say .04 win difference.

At the same time, a blown ball/strike call with an LI of 3 will give you the same .04 win difference.

.07 seems pretty high.  I’ll have to check it out again.

We’ll almost never be able to pin one particular play as the reason for anything, and bad moves is more an indicator, allowing for inference about the manager, rather than about the outcome of the game itself.

For all I know, a pinch hitter in the 8th might have struck out.  Burrell might have doubled off Bradford anyway.  The Phillies might have gotten the runner in even if the Rays went to Price for the strikeout.  Maddon’s moves may or may not have cost them the game, but on paper those moves certainly decreased his chances of winning it.

No argument here.

Groundball to the right side, which is what happened.  It didn’t have to happen that way but forcing Victorino to bat righty and having a groundball pitcher sure tipped the odds in that favor.  Howell would have been more likely to get a groundball to the left side, and keep the runner on second.

I didn’t even think about that, but that is 100% correct.  You want to make Vitorino hit RH.  That is a big advantage.  Even though RH and LH batters hit about the same number of fly balls to the opposite or pull fields, they hit longer fly balls to the pull side.

Aren’t these the kinds of things that people think the MANAGER thinks of but WE don’t?

We’ll almost never be able to pin one particular play as the reason for anything, and bad moves is more an indicator, allowing for inference about the manager, rather than about the outcome of the game itself.

I NEVER think of things that way, and neither should any good analyst or decision-maker.  One decision gains or loses WE over another.  That’s it.  To decide or determine whether a certain play or plays “caused” the game to be won or lost is an exercise in sheer futility and is completely non-productive.  I learned and practiced that a long time ago while playing blackjack for a living (a tough way make easy money).  Besides that it will drive you crazy. But more than that, it will encourage you to make bad decisions, such as if your manager makes a bad decision that ends up costing you the game, you STILL need to address the fact that he made the wrong decision, otherwise he may do it again, which will, of course, decrease your chances of winning that game.  I am preaching to the choir, of course, but your never know who is reading these.  We have many lurkers (which is great).

makes a bad decision that ends up costing you the game

I meant, “ends up winning you the game,” of course…