Reversal of Fortune

[Edgy MD]

Win probability by event, from the bottom of the eighth inning, NYN@STL, 4/25/2022.



Bottom of Eighth Inning, Cardinals at Bat



　Score is tied, 0-0.



　　　Trevor May pitching for Mets.



　　Mets Win Expectancy: 41.4%



　　　Yadier Molina singles to left.



　　Mets Win Expectancy: 33.8%



　　　Harrison Bader singles to left. Yadier Molina advances to second.



　　Mets Win Expectancy: 24.4%



　　　Brendan Donovan pinch-runs for Yadier Molina.



　　　Tommy Edman grounds out, pitcher to secondbaseman to firstbaseman covering first. Brendan

　　　Donovan advances to third. Harrison Bader advances to second. One out.



　　Mets Win Expectancy: 22.4%



　　　Paul Goldschmidt walks.



　　Mets Win Expectancy: 22.6%



　　　Tyler O'Neill singles to left. Brendan Donovan scores. Harrison Bader scores. Paul Goldschmidt advances to second.



　Cardinals Lead, 2-0.



　　Mets Win Expectancy: 4.3%



　　　Nolan Arrenado strikes out swinging. Two out.



　　Mets Win Expectancy: 5.2%



　　　Corey Dickerson flies out to right. Three out.



　　Mets Win Expectancy: 6.1%



=#002D72]Top of Ninth Inning, Mets at Bat



　Cardinals Lead, 2-0.



　　　Giovanny Gallegos pitching for Cardinals.



　　　Pete Alonso flies out to center. One out.



　　Mets Win Expectancy: 2.8%



　　　Eduardo Escobar singles to center.



　　Mets Win Expectancy: 7.3%



　　　Robinson Canó flies out to left. Two out.



　　Mets Win Expectancy: 3.0%



　　　Eduardo Escobar advances to second on defensive indifference.



　　Mets Win Expectancy: 3.3%



　　　Mark Canha singles to third. (Others might well have scored this an error.) Eduardo Escobar scores on throwing error by Nolan Arrenado.



　Cardinals Lead, 2-1.



　　Mets Win Expectancy: 7.1%



　　　Travis Jankowski pinch-runs for Mark Canha.



　　　Jeff McNeil doubles to right. Travis Jankowski advances to third.



　　Mets Win Expectancy: 19.2%



　　　Dominic Smith singles to first. Travis Jankowski scores. Jeff McNeil scores.



　=#002D72]Mets Lead, 3-2.



　　Mets Win Expectancy: 84.3%



　　　Brandon Nimmo homers to right field. Dominic Smith scores. Brandon Nimmo scores.



　=#002D72]Mets Lead, 5-2.



　　Mets Win Expectancy: 97.0%



　　　Starling Marte strikes out looking.



　　Mets Win Expectancy: 96.9%



Bottom of Ninth Inning, Cardinals at Bat



　=#002D72]Mets Lead, 5-2.



　　　Edwin Díaz pitching.



　　　Dylan Carlson pops out to third.



　　Mets Win Expectancy: 98.7%



　　　Edmundo Sosa strikes out swinging.



　　Mets Win Expectancy: 99.7%



　　　Andrew Knizner walks.



　　Mets Win Expectancy: 98.8%



　　　Andrew Knizner advances to second on defensive indifference.



　　Mets Win Expectancy: 98.7%



　　　Harrison Bader strikes out swinging.



　　Mets Win Expectancy: 100.0%



　=#002D72]Mets Win, 5-2.



https://metsrostercentral.files.wordpress.com/2022/04/screen-shot-2022-04-26-at-11.28.03-am.png>

[roger_that]

Most exciting baseball since Game Six.

[Edgy MD]

With an infield hit in his lone plate appearance, Dom Smith may well earn your PotG vote despite the terrific start by Scherzer.



He gets mine!

[Benjamin Grimm]

Edgy MD wrote: 　　Mets Win Expectancy: 22.4% 　　　Paul Goldschmidt walks. 　　Mets Win Expectancy: 22.6%

Interesting that Goldschmidt walking slightly increased the chances of the Mets winning. Were "they" expecting that Goldschmidt was likely to do something more damaging than drawing a walk?

[metsmarathon]

shoot, i may have to wind up little old m.e.t.b.o.t. after that game!



i believe the drop in win expectancy is because now with a runner on first, you have more ways to get out of the inning with no further damage, including and especially via a GIDP, which are slightly more likely to occur than the ways in which more runs can score. historically. i guess.

[Edgy MD]

Yeah, a lot of things Goldschmidt could have done there would have been more damaging than drawing a walk, including a modestly deep groundout or a modestly deep flyout. That's why the intentional walk was on the table, and it's certainly what I would have done.



Buck opted not to go in that direction, as per his post-game comments, because he respects Tyler O'Neill's batting skillz and he thinks O'Neill is a particularly difficult guy to double up, neither of which would be particularly intuited by the win expectancy measurement.



I still would have intentionally walked him, but O'Neill's base hit supports Buck's position to a meaningful extent.

[metsmarathon]

btw, m.e.t.b.o.t. woulda shaeffered as follows:



smith 5.10

scherzer 3.21

mcneil 0.71

canha 0.56

diaz 0.24

escobar 0.15

nimmo 0.02

[Frayed Knot]

Benjamin Grimm wrote: Edgy MD wrote: 　　Mets Win Expectancy: 22.4% 　　　Paul Goldschmidt walks. 　　Mets Win Expectancy: 22.6% Interesting that Goldschmidt walking slightly increased the chances of the Mets winning. Were "they" expecting that Goldschmidt was likely to do something more damaging than drawing a walk?

No, it's the force play factor.

Like we were discussing the other day, these probabilities are generic so the situation is factored in but not specifically Goldschmidt and his Goldschmidtedness. It's just Joe Batter up there as far as the odds are concerned.

[batmagadanleadoff]

=metsmarathon post_id=90415 time=1650992362 user_id=83]
btw, m.e.t.b.o.t. woulda shaeffered as follows:



smith 5.10

scherzer 3.21

mcneil 0.71

canha 0.56

diaz 0.24

escobar 0.15

nimmo 0.02

[Edgy MD]

It does, but RBI does the same, and ERA credits the pitcher for quality defense that has little to do with him. Dom is also getting credit for Gallego being slow to get off the mound, and he's losing credit because Goldschmidt makes a terrific stop, neither of which have much to do with him.



It's a team sport and statistically extracting the performance of one guy from the team's outcome is always going to be a challenge. But we're moving in the right direction.

[batmagadanleadoff]

I'm not a big RBI fan either. At least with the RBI producing at-bat, though, someone else is getting credit for a run scored. And now there's FIP.

[ashie62]

That's why the play the games, right?

[batmagadanleadoff]

=ashie62 post_id=90432 time=1650998419 user_id=90]
That's why the play the games, right?

[Chad ochoseis]

Frayed Knot wrote: No, it's the force play factor. Like we were discussing the other day, these probabilities are generic so the situation is factored in but not specifically Goldschmidt and his Goldschmidtedness. It's just Joe Batter up there as far as the odds are concerned.

That's why the baseball gods invented the intentional walk. Buck should have just held up 4 fingers there. Even more so because it was one of the best hitters in the game at bat.

[Frayed Knot]

I would have IW'd him too. And for Ronnie to be any more adamant about the free pass he would have had to actually be Adam Ant.



The result of Buck's method was ultimately the same, just with more stress on his pitcher ... not to mention on us.

[roger_that]

Frayed Knot wrote: I would have IW'd him too. And for Ronnie to be any more adamant about the free pass he would have had to actually be Adam Ant. The result of Buck's method was ultimately the same, just with more stress on his pitcher ... not to mention on us.

Wonder if the odds bear you out on the IBB.



Let's say the greatest hitter in the game is up there. Willie "Babe" Trout is a .350 hitter with50 HR power when you're trying to get him out, i.e., throwing some pitches into the part of the strike zone that he favors least.



What happens to those .350/50 numbers if you're determined to pitch him where he doesn't like the ball, only you're trying to put it six inches off the plate?



Probably, 70% of the time you end up walking him anyway, because Willie "Babe" Trout knows the strike zone. Which is the same result as an IBB. But the other 30% of the time, you've reduced those gaudy numbers to mere human levels, .275/20 or something like that.

[Edgy MD]

That's mostly all there, but I would disagree that the end result is the same as an intentional walk. For one, the pitcher is spending his bullets — and perhaps some of his best bullets — on a batter he's 70% (in your model) likely to walk anyhow. For two, pitching around a guy can lead to wild pitches and passed balls, and with the go-ahead run on third, that's some powerful poison you're messing with. Those aren't really factored into the data, but I'm sure feed into the thinking of most managers facing the decision.



While I'm on Team Intentional Walk here, the downside is that a pitcher loses a lot of his margin for error. A walk or hit-by pitch plates the runner on third.



I go for the walk based on:



1. it setting up the double play,



2. it setting up the force at home (or third, to a lesser extent),



3. having one less high quality hitter to face.



I respect Buck's reasoning, and in fairness, May ended up coming pretty close to getting Arrenado.

[roger_that]

Malcolm Gladwell has a great podcast episode on the dumb stuff sports teams do systematically and reflexively that studies have proven are counter-productive. The IBB may be one of those. There are those who simply counsel "Never* IBB anyone ever cuz more times than not it bites you on your ass." That is, the number of runs that result after IBBs is much greater than the number of runs if you pitch to guys in IBB situations.



Gladwell doesn't use it as an example, but his NFL and NBA examples of dumb persistent counterproductive policy decisions are wonderful: https://www.pushkin.fm/episode/the-big-man-cant-shoot/https://www.pushkin.fm/episode/the-big-man-cant-shoot/







*not literally. With Babe Ruth at the plate, and Dean Chance on deck with no pinchhitters available for Chance, and men on second and third, okay, go fer it!

Edited April 26, 2022 by Guest

[Johnny Lunchbucket]

I'm glad Godschmidt's great stop came up. That could very easily been double for Dom and no drama. You got to think that Gallegos musta figured base hit off the bat contributing to the slow response

[roger_that]

Johnny Lunchbucket wrote: Gallegos musta figured base hit off the bat contributing to the slow response

No excuse, really. What, was he doing something more important than running his ass off to 1B at that moment?

[Frayed Knot]

Probably, 70% of the time you end up walking him anyway, because Willie "Babe" Trout knows the strike zone. Which is the same result as an IBB. But the other 30% of the time, you've reduced those gaudy numbers to mere human levels, .275/20 or something like that.

Providing said pitcher can put his pitches where he wants, which if he could he probably wouldn't be in that predicament in the first place.

I'm on team-IW for the same reason Edgy mentions: sets up the all-around force and it's Goldschmidt.

Of course there's a downside to it, there always is.



In a NYM board from another era, some of us here frequently heard cries of 'Walk him every time up!!' whenever some Met-killer-du-jour stepped up to the plate.

Explaining the pure 'dumpth' of that reasoning was exasperating (this board was essentially created to avoid dopes like that) but I'm also not going to go to the other

extreme and perch myself on planet 'Never IW' either. I won't advocate the 'Four Fingers of Fate' too often but I would have there.

[Johnny Lunchbucket]

Johnny Lunchbucket wrote: Gallegos musta figured base hit off the bat contributing to the slow response No excuse, really. What, was he doing something more important than running his ass off to 1B at that moment?

Of course not. I'm just trying to ascribe some motive for his laziness. I would be surprised if he'd have lollygagged against a more tradional first-baseman-off-tge-bag kinda play

[Edgy MD]

And hey, Malcolm Gladwell is no dummy, but a world of data crunched together still suggests the intentional walk gave the Mets a marginally better chance of winning, without considering the batters.



I could talk about the plusses and minuses of the walk in that situation all night. And I was up late last night thinking about. It was part of the reason I launched this thread, and I'm delighted that Grimm noticed that Arrenado's walk was listed as a disadvantage to his team.



But staying up all night talking about it is one thing. The manager only has a few moments to make the call.

[roger_that]

Edgy MD wrote: I'm delighted that Grimm noticed that Arrenado's walk was listed as a disadvantage to his team. But staying up all night talking about it is one thing. The manager only has a few moments to make the call.

Definitely interesting to have Arenado's walk so noted.



But the "manager only has a few moments to make the call" is both true and, in Gladwell's larger sense, completely false. In his sense, managers have had over a century to come up with the answer and they (may have) come up with the wrong one. I'm just speculating here, because he never addressed this particular issue of IBB, but if you take the half hour to listen to that episode, it's amazing the things that very smart people in sports do all the time that are completely contrary to their own aims, and IBB may be among them. Wilt Chamberlain, for example, may have won several more championships, and certainly many more games, if he had been willing to do one small thing to improve his game, and NFL teams may be managing their draft picks all wrong. But they persist in doing so, not because they're dumb, but because that's how they've always done it.



Anyway, I don't mean to hijack this thread so maybe it's best to raise it elsewhere, and discuss the larger issue on its own merits, or lack thereof.

[roger_that]

Johnny Lunchbucket wrote: Johnny Lunchbucket wrote: Gallegos musta figured base hit off the bat contributing to the slow response No excuse, really. What, was he doing something more important than running his ass off to 1B at that moment? Of course not. I'm just trying to ascribe some motive for his laziness. I would be surprised if he'd have lollygagged against a more tradional first-baseman-off-tge-bag kinda play

Okay, noted. But this is strangely the same issue that Edgy MD just raised with me, that of a lack of time to make the wisest choice. It's often unnoted how vital split-second thinking is to sports in general. We never think of this as an athletic quality, but so many games depend on instantaneous thinking rather than strength, or speed, or physical agility, and it's certainly much harder to notice, as a fan. We say "Wow what a great throw to third base from the right-field corner" but we don't say, "In addition to the strength and accuracy of his throwing arm, how the hell did he figure out which base to throw to in the quarter-second between catching the ball and fishing it out of his mitt?"

[Frayed Knot]

But the "manager only has a few moments to make the call" is both true and, in Gladwell's larger sense, completely false. In his sense, managers have had over a century to come up with the answer and they (may have) come up with the wrong one.

Except that each situation is different so neither the 'Always IBB in that case' or 'Never IBB' is the correct call. One may be the better option today but the

other tomorrow based on a whole host of factors. The Bill James-led statistical revolution correctly pointed out the poor results from the near-automatic

use of the man-on-1st/no-out Sac Bunt that was in vogue for most of the 20th century but that's not the same as advocating it be scrapped forever.



As related by Gary last night, Buck cited the speed on O'Neil hitting behind Goldschmidt (reducing GiDP odds) and moving one batter closer to Arenado

as his primary reasons for Not walking Goldschmidt intentionally. And, as we all know, they pitched to him carefully and walked him anyway.

[roger_that]

Frayed Knot wrote: But the "manager only has a few moments to make the call" is both true and, in Gladwell's larger sense, completely false. In his sense, managers have had over a century to come up with the answer and they (may have) come up with the wrong one. Except that each situation is different so neither the 'Always IBB in that case' or 'Never IBB' is the correct call. One may be the better option today but the other tomorrow based on a whole host of factors. The Bill James-led statistical revolution correctly pointed out the poor results from the near-automatic use of the man-on-1st/no-out Sac Bunt that was in vogue for most of the 20th century but that's not the same as advocating it be scrapped forever. As related by Gary last night, Buck cited the speed on O'Neil hitting behind Goldschmidt (reducing GiDP odds) and moving one batter closer to Arenado as his primary reasons for Not walking Goldschmidt intentionally. And, as we all know, they pitched to him carefully and walked him anyway.

I'm not saying anything so absolute as "Never" or "Always."



What I am saying is that, maybe, the IBB is called for ten times, or a hundred times, more than it should be.



What I'm saying is that in 100 "IBB?" situations, if you go for the IBB all 100, 200 runs will score in that inning from that point forward, and if you go for it 0 times, 75 runs will score from that point forward, so you need to be FAR more selective than managers have been.



But managers are reluctant to look at the overall numbers, because 1) tradition and 2) if you buck the odds, the blame attaches to you. "Why didn't you walk him? With men on second and third, he hit a 3-run HR and won the game" makes you look stupid, whereas if you'd put him on first and they ended up scoring 10 runs in that inning, it's still defensible.



What I suspect is that managers issue IBBs for many different reasons:



to gain a platoon advantage

to avoid a powerful hitter

to force the other manager to pinchhit for the guy batting after the IBB

to set up the DP



and that all four, or at least three, need to apply, to reach the breakeven point, but managers go for the IBB if only one or two apply. Also other factors: late in the game, with a very small lead, or none at all, or at home/on the road. There might be ten factors, and nine of them are needed for the IBB to make sense.



Food for thought anyway. This might be an example of an exploitable strategic inequity.

[Edgy MD]

Fun Fact: The Mets have issued but two intentional walks* so far this young season. More interesting is that puts them in the top half of the league. Twelve of 16 National League teams have issued zero or one intentional walks.



* Both occurred in the https://www.baseball-reference.com/boxes/NYN/NYN202204160.shtmlsame game. The Mets ended up getting out of both situations with no runs scoring. In both situations, there were two out.

[metsmarathon]

=batmagadanleadoff post_id=90428 time=1650996622 user_id=68]
=metsmarathon post_id=90415 time=1650992362 user_id=83]
btw, m.e.t.b.o.t. woulda shaeffered as follows:



smith 5.10

scherzer 3.21

mcneil 0.71

canha 0.56

diaz 0.24

escobar 0.15

nimmo 0.02

[Edgy MD]

=metsmarathon post_id=90415 time=1650992362 user_id=83]but he always brought a unique perspective to the game, that may or may not have merit.