46 Responses to “Ramon Hernandez and a catcher leading the way in WPA”

1. Timmy p Says:
   June 8th, 2011 at 9:03 am

   I looked at Juan Pierre's stats, and then at Maury Wills. I think JP's stats are better than Wills.

2. John Autin Says:
   June 8th, 2011 at 10:37 am

   Looking at it from a seasonal perspective ... In the 61 years of WPA data (back to 1950, with a few games missing), it seems that only 2 catchers have ever led the league -- Joe Mauer in 2008, and Roy Campanella in 1953.

   Two reasons spring to mind: catchers generally play fewer games than other positions, and they usually aren't top hitters. But I'm still surprised that only 2 have led their league in WPA, out of 122 opportunities.

3. Voomo Zanzibar Says:
   June 8th, 2011 at 11:10 am

   Nobody is ever going to break Wills' record of 165 games played in 1962.

   And he had 4 hits and 3sb in the rubber game of that 3-game playoff in a losing effort. Dodgers blew a 2-run lead n the 9th.
   4 walks, a wild pitch, and an error.

   2nd and 3rd, two out, tie game, and they intentionally walk Ed Bailey.
   And then walk in the winning run. Oooof.

4. aweb Says:
   June 8th, 2011 at 11:15 am

   Catchers typically are pinch-run for in close games more often than other positions as well, so they might lose a few chances a year to come up in a close game late or in extra innings. This might also help account for the lack of leading WPA seasonally.

   For instance, Mike Piazza comes up in a tie game in the ninth, leads off with a single, adding to his WPA. He is pinch run for immediately though, and never gets to hit in the 11th and 14th innings of this hypothetical game, when he might have picked up another pig event or two.

5. Voomo Zanzibar Says:
   June 8th, 2011 at 11:32 am

   My breakfast was a pig event today.

6. John Autin Says:
   June 8th, 2011 at 12:17 pm

   @3, Voomo -- Considering that 4 players have logged 164 games in a season without the benefit of a playoff (including Ron Santo and Billy Williams of the '65 Cubs), I'd say that Wills's 165-game record is attainable.

7. BSK Says:
   June 8th, 2011 at 12:30 pm

   I was just thinking about this the other day: what is the highest possible WPA on a single play? My hunch would be that it'd have to be a based loaded, 2-out walk-off grandslam when down by 3. But perhaps there is something I am missing.

   For pitchers, I would assume it'd have to be a triple play with the bases loaded, no outs, in the final inning and a 1-run lead.

   I'd venture to say that the highest possible batters WPA is much higher than the highest possible pitchers WPA, because a pitcher cannot erase a deficit the way a hitter can. Can anyone confirm this?

8. Jimbo Says:
   June 8th, 2011 at 12:51 pm

   How about if a pitcher throws a perfect game with 27 strikeouts and gets the only hit of the game, a home run, himself.

9. Jimbo Says:
   June 8th, 2011 at 12:53 pm

   Oops I wasn't thinking of it being a single play. Then I guess your example is the best possible. What if he caught a liner with the runners moving to start the triple play? Also, the count could be 3-0 to the hitter.

10. BSK Says:
    June 8th, 2011 at 12:54 pm

    JA-

    I think the fact that catchers tend to be bad (or at least slow) base runners is also a factor. Seeing as how position players can only accrue WPA through two means (hitting and base stealing), being largely absent from one category is another hill to climb. Of course, there are catchers who did steal bases and I'm sure plenty of hitters who led in WPA without a stolen base but I would venture to guess this is a third, albeit smaller, factor for their lack of black ink in the category.

11. BSK Says:
    June 8th, 2011 at 12:56 pm

    Jimbo-

    Great point about the count. 3-0 for the pitcher, 0-2 for the batter.

    However, I don't think who is involved in the triple play matters, since the value is added only to the pitcher.

    Another question: would that highest WPA for a pitcher be above 50%? Basically, can a team be in the lead but have less than a 50% chance of winning? My gut says absolutely, and probably in a far less extreme case than the one cited.

12. mccombe35 Says:
    June 8th, 2011 at 1:33 pm

    @7

    Reminded me of the Ventura slam off Gossage back in '91. Except they were only down 2 with bases loaded & 2 out in the 9th.

    http://www.baseball-reference.com/boxes/CHA/CHA199107310.shtml

    Of course, slams were Ventura's specialty.

13. BSK Says:
    June 8th, 2011 at 2:39 pm

    Ventura's WPA for that play was .83. Less than Ramon Hernandez. Has a walk-off grand-slam ever been hit when trailing by 3? I know Giambi hit one for the Yanks during a rain storm his first year there, but I don't think they were down 3.

14. topper009 Says:
    June 8th, 2011 at 2:47 pm

    BSK,

    here you play around with this to determine WPAs

    http://www.hardballtimes.com/thtstats/other/wpa_inquirer.php

15. Thomas Says:
    June 8th, 2011 at 3:11 pm

    @13, I know it's happened because I remember reading about it in an old Phillies Media Guide.. they called in an Ultimate Grand Slam. Down by three, bottom of the ninth, two outs. I'm searching for it/one now in PI...

16. Thomas Says:
    June 8th, 2011 at 3:17 pm

    there's 9 instances since 1973... heres the most recent:

    http://www.baseball-reference.com/boxes/CIN/CIN200606300.shtml

    This is the only one with a full count:

    http://www.baseball-reference.com/boxes/DET/DET198806210.shtml

    So, these are 2 of the 9 games, since 1973 that are bottom of the 9th, 2 outs, bases loaded, down 3 runs.

17. Thomas Says:
    June 8th, 2011 at 3:20 pm

    Also, it's only happened once in extra innings (again since 1973):

    http://www.baseball-reference.com/boxes/SLN/SLN197905010.shtml

18. BSK Says:
    June 8th, 2011 at 3:25 pm

    A few notes...

    A) The boxscore indicates that Hernandez's HR had a WPA of .91, not .93. Not sure why there is a discrepancy.
    B) I'm not exactly sure how to properly set the run environment.
    C) Depending on the run environment, there seem to be instances where the team down by 3 with the bases loaded has a higher chance of winning than a team down by 2 with runners on 1st and 3rd. I suppose that, theoretically, these might be the same, because they both require getting the man from first all the way around to tie the game. But I can't figure out why the greater deficit would be a more likely win. However, if WPA is calculated simply based on a historical results, it is possible this is simply a "glitch" in the data, wherein the historical record doesn't necessarily fit witht he predictive odds.
    D) For the pitcher, it looks like a WPA of about .75 is the most, presuming the bases-loaded triple-play scenario is the highest possible result. Teams with a 1-run lead but bases loaded and no outs on their opponent win only approximately 75% of the time.

    Can someone explain to me how to set the run environment properly? And how the hell does C) happen?

19. BSK Says:
    June 8th, 2011 at 3:27 pm

    Thanks, Thomas. Those three all had WPAs of .90 or .91 in the boxscore. Still not sure why the box score WPAs and the ones appearing in searches are different, but it looks like .91 is around the upper limit. It's cool to see how steep the graph of odds of winning is at the far right tail.

20. Thomas Says:
    June 8th, 2011 at 3:29 pm

    I think the discrepancy (your A) is that the .932 is his game total, not the at bat total.

    Although, he did recieve .91 for his 3 run walk off, which is the same as a few grand slam walk offs I posted above, and .01 higher then Adam Dunn's walk off grand slam, down 3. So... still a bit of confusion...

21. John Autin Says:
    June 8th, 2011 at 3:34 pm

    Re: highest possible WPA on one play:

    (1) I'm pretty sure that B-R does not have WPA figures that take the count into consideration. I don't know if anyone does.

    (2) I do think that a team with a lead can have a less than 50% chance of winning, according to WPA methods. I don't know quite how to convert run expectancy into win expectancy, but ... a team with the bases loaded and no outs has a run expectancy of around 2.3 runs in that inning., so a team on defense with a 1-run lead in that situation would not be sitting pretty.
    -- Run expectancy matrix: http://www.insidethebook.com/ee/index.php/category/Run_Win_Expectancy/

    (3) I don't know how to calculate the maximum possible WPA on a single play. But based on what has actually happened in the WPA-searchable past, it would seem that a hitter can amass far more WPA than a pitcher on a single play. The highest WPA by a batter who had 1 PA or less is 0.918, while the highest WPA by a pitcher who faced 1 batter or less is a mere 0.557.

    BTW, here's that 0.918 WPA in 1 AB. In 1973, Cincinnati's Hal King pinch-hit a 3-run HR with 2 out in the bottom of the 9th, erasing a 2-run deficit.
    -- http://www.baseball-reference.com/boxes/CIN/CIN197307011.shtml

    Believe it or not, right before King batted, Don Sutton gave an IBB to Johnny Bench, representing the tying run. On the surface, it looks like one of the dumbest moves imaginable, since that situation is all about HR rates, and while Hal King had a poor BA, he had a good HR rate, not that far below Bench's HR rate -- and King would have the platoon advantage. How can you choose to put the tying run on base there?

    But consider:
    -- To that point in time, Bench had 5 HRs in 62 AB vs. Sutton, including a 9th-inning, 2-out, 2-run, game-tying blast just 2 weeks earlier.
    -- Bench would finish his career with 12 HRs in 146 AB vs. Sutton -- 50% more HRs than any other hitter had against Sutton, and the most HRs Bench had against any pitcher.

    The 0.557 WPA by a pitcher facing just 1 batter was done by Chris Holt of Houston in a 1999 interleague game against KC. The Astros began the inning with a 3-run lead, but the Royals had pulled within a run and had the sacks loaded with 1 out. Holt came in and got Mike Sweeney to hit into a game-ending DP.
    -- http://www.baseball-reference.com/boxes/KCA/KCA199907090.shtml

22. Thomas Says:
    June 8th, 2011 at 3:46 pm

    I guess I'm a little confused then on how hitting a 3 run homer down 2 in the bottom of the 9th has a higher WPA then hitting a grand slam down 3 in the bottom of the 9th. Is it just that the odds of hitting any home run down, in the bottom of the 9th is that small?

23. John Autin Says:
    June 8th, 2011 at 3:46 pm

    I may be dense, but ... It seems to me that as long as your team is behind with 2 outs in the bottom of the last inning, all baserunner/deficit situations in which the tying run is on 1st base will have the exact same WPA.

    So there's no special WPA magic to a grand slam. After all, WPA doesn't "care" how much you may lose the game by, only whether you win or lose.

    I was wondering why my search results for highest WPA with just 1 PA did not feature many grand slams (just 2 of the top 26). I think I just explained it.

    Am I missing something?

24. John Autin Says:
    June 8th, 2011 at 3:50 pm

    @22, Thomas -- I think my @23 answers your question. It's all about the tying and winning runs. Whether you're down 1 with a man on 1st, or down 3 with the bases loaded, all that matters for WPA is getting those tying and winning runs in.

25. John Autin Says:
    June 8th, 2011 at 3:53 pm

    [incredibly sheepish look]
    OK, so, I just realized that B-R's play-by-play include the count. Dunno how I've missed that all these years; do I have to turn in my propeller beanie?

    Still not sure that the count impacts the WPA figures.

26. Thomas Says:
    June 8th, 2011 at 3:57 pm

    @23... that makes sense... I hadn't really thought of it that way....

27. John Autin Says:
    June 8th, 2011 at 4:01 pm

    I'm 99.9% certain that the count plays no role in B-R's WPA figures.

    B-R's description of Win Expectancy and WPA does not mention the count:
    -- http://www.baseball-reference.com/about/wpa.shtml

    B-R also says the underlying charts were produced by Tom Tango, and links to this article:
    -- http://www.hardballtimes.com/main/article/crucial-situations
    ... in which Tango writes: "It is important to note that a game state is anything you want it to describe. While I will describe it, for the purposes of this article, as the state of the game given the half-inning, score differential, base, and out situations, you most certainly can expand the game state to include the identities of all the players in the game, the park, the count, and the climate." [emphasis added]

28. topper009 Says:
    June 8th, 2011 at 4:09 pm

    Remember, WPA is not theoretical, it is empirical based on previous results. Now there is 90+ years of data, but that is across different run environments. All WPA says is giving the current state of that game how have previous games ended up.

    In theory you should never get a 100% win probability until the game is over because there is no time limit and any comeback is possible, but there are some deficits that a team has never comeback from so if you found yourself in that situation you would have a win % of 0. I tried to find some of these and I think leading by like 7 in extra innings is one example.

    Also, whatever was the largest comeback in MLB history, (not sure what it is) had a 0 % of happening since it had never occurred before, however now it has a 1% chance of happening.

29. Jahiegel Says:
    June 8th, 2011 at 4:48 pm

    An observation that is, upon one's considering that almost all of Hernández’s WPA came from a single plate appearance, unsurprising: John Axford's Opening Day WPA, -.963, remains the lowest total compiled by a pitcher this season.

    That loss to the Reds, one on which even more than two months on we Brewers fans look with grave disappointment, is worthy of note for another, almost incredible reason: it was the only of the 18 games in which he has appeared this year in which Wil Nieves had a positive WPA; his futility has been astonishing even for a player of his limited offensive capabilities (worst-hitting catcher [by wRC+] of the last 25 years [minimum 800 PA], better [by wRC+] than just nine of the 1416 position players to have made at least 500 PA since 1986).

30. Jahiegel Says:
    June 8th, 2011 at 4:53 pm

    Lest one should think me to have committed an non-parallelism, the most serious non-genocidal crime of which one can be guilty, I should say that the second paragraph should be recast to read, in pertinent part, "his futility has been astonishing even for that ofa player of his...", although even that locution is sub-optimal. I can think only that I was so overwhelmed by the revelation of how bad Nieves has been that I wrote before properly constructing my thoughts.

31. Johnny Twisto Says:
    June 8th, 2011 at 4:58 pm

    No, WPA is theoretical.

32. BSK Says:
    June 8th, 2011 at 5:43 pm

    JT-

    Can you elaborate? How would one go about calculating that theoretically? You could only do so based on historical data. Hitting a ball is not flipping a coin; there is no theoretical odds to it.

33. Johnny Twisto Says:
    June 8th, 2011 at 5:51 pm

    Well it is based on historical data in the sense that we know how often certain events occur and how many runs they are worth. But the win expectancy for a game in Yankee Stadium in 1972 while down by 2 runs with 1 out in the top of the 5th is not empirically based on only the results of games which happened to be at that specific game-state in the past. The sample would be too small to be useful.

34. Voomo Zanzibar Says:
    June 8th, 2011 at 5:53 pm

    @6
    John, yes I see.
    Had not considered the other ways to log extra games.
    I was just assuming that a tie at the end of the regular season will not be handled with a three-game series ever again.

35. BSK Says:
    June 8th, 2011 at 6:57 pm

    JT-

    So what else is it based on? More general historical data, such as how often a team scores 2 runs in an inning? Etc, etc, etc?

    Please note, I'm not trying to be snarky, just trying to better understand.

36. thom-13 Says:
    June 8th, 2011 at 9:36 pm

    @17

    I saw that some others gave the count when the game-ending blow was hit. Is there an easy way to check the count when Roger Freed pinch-hit the game-ending grand slam in that game?

37. Neil L. Says:
    June 8th, 2011 at 11:13 pm

    @36
    Thom-13, the count is not available in BRef for the Roger Freed game.

    In a gamelog the count is shown in the fifth column from the left. You can see that for the Freed game that column is blank.

38. Johnny Twisto Says:
    June 8th, 2011 at 11:41 pm

    BSK, I suppose it's more accurate to say it's theoretical based on empirical. (Well, maybe that's true of everything theoretical....)

    Win expectancy is based on run expectancy. Run expectancy is based the observed runs scored from each of the 24 possible base-out states in an inning. But it is not based solely on those actual observations, it is tweaked a bit. This is done both to account for insufficient sample size and to account for different run environments.

    I suggest reading the first chapter of The Book. (I see you can read a piece of it at Amazon. Anyone know if there's a way to read more? I feel like Tango has often referred people to some free site to read the book, but I don't remember if that was Amazon or something else that has even more of it available.)
    http://www.amazon.com/Book-Playing-Percentages-Baseball/dp/1597971294/ref=sr_1_1?ie=UTF8&qid=1307590203&sr=8-1

    And I know you're not being snarky. If you have more questions, I'll do my best to answer them, though we may be butting up against the limits of my abilities here....

39. BSK Says:
    June 9th, 2011 at 12:00 am

    JT-

    I actually picked up "The Book" from the library recently and have been working my way through it (as "math brained" as I am, stats was never my strong suit). I should plug B-R for inspiring me to do so. It was my understanding the the Win Expectancy and Run Expectancy were entirely empirical, though I'd probably still consider them such even allowing for minor tweaks. It actually makes a lot of sense that Win Expectancy would not be based solely on W-L records after certain base-out-run-inning situations have occurred. But it is still pulled from historical data, namely Run Expectancy.

    I got into it a bit with Andy once regarding this same issue with CoolStandings (I think it was over the infamous Phillies post). It is something that has generally always bothered me about such approaches to predictions. Saying something has "never" happened before doesn't mean it can't happen or won't happen; only that it is probably something very unlikely to happen. Now, when a data sample gets large enough, we probably can make some generalizations, but there will always be noise in the data.

    As I noted before, we can look at a coin toss and *know* (assuming a fair coin) the likelihood of a given flip coming up heads or tails. From there, we can determine the likelihood of any combination of flips. We can compare this to the historical data, but the theoretical odds will *never* change. This is not true of baseball. As you note, run environments play a huge factor. Overcoming a 7-run deficit in the mid-90's is very different than doing so in the ead Ball era. How different? We can probably approximate, but never know for sure.

    In the end, we're arguing (if we even are arguing!) about numbers a place or two too far to the right of the decimal point to be truly meaningful. What matters most is that we understand what the stats are attempting to tell us and whether or not they are successful in those endeavors (or, really, the people behind the stats... the stats are simply what they are).

    I'll keep going through "The Book" and hopefully can add more to the conversation.

40. Doug Says:
    June 9th, 2011 at 12:29 am

    @6, @34.

    Another way a player could end up playing more than 162 is if he is traded in mid-season from a team that had played more games than the team he is traded to.

    My hunch, though, is this is pretty unlikely to happen since you don't usually trade a player who's been in your lineup every game. Probably, he'd lose some playing time first before you traded him. Could be possible, perhaps, if it was a salary dump or an unsignable free agent. But, even then, if a deal was close to being consummated, that player would probably be sat down just prior to the deal going down, to avoid risk of injury.

41. John Autin Says:
    June 9th, 2011 at 10:26 am

    @40, Doug -- That situation happened in reverse to Matt Holliday in 2009. With Oakland, he played 93 of 94 games through July 23, then was traded to St. Louis. He played with the Cards the next day, their 99th game -- so he "lost" 4 potential games. Holliday missed just 1 game with Oakland and 1 with St. Louis, but played just 156 games. Had the trade gone the other way, he might have played 164 games.

    These days, with free-agents-to-be often dealt at the deadline, I don't think it's uncommon for an everyday player to be traded during the season.

42. kds Says:
    June 9th, 2011 at 11:49 pm

    JT and BSK, I think it is likely that a Markov chain simulator is used to generate the WPA figures.
    Home team down by a run loads the bases in the bottom of the 9th with none out, in a 5 runs/gm environment they have a 74.92% chance of winning. So a triple play would get the pitcher .7492 WPA. I think this is the highest possible for one play by the pitcher/defense. (Given the run environment.) Thanks to Topper009 for that link, lots of fun.

43. Johnny Twisto Says:
    June 10th, 2011 at 12:55 am

    I think a Markov is used, that is referred to in The Book.

44. nesnhab Says:
    June 11th, 2011 at 8:14 am

    Sorry if it's been asked before...does the WPA account for fielding events of significance...for example, Joe Rudi's leaping catch in 9th, G2, 1972 WS? And, what was that catch (out) considered of any significance by the WPA compared to other events that day?

45. nesnhab Says:
    June 11th, 2011 at 8:16 am

    The batter who hit Rudi's catch was Denis Menke...IIRC

46. Johnny Twisto Says:
    June 11th, 2011 at 9:47 pm

    Sorry if it's been asked before...does the WPA account for fielding events of significance

    It does not. All credit is given to the pitcher. It will be a great step when someone can devise an objective, consistent way to divide credit with the fielders too. We brainstormed a bit a couple months ago on how that might be done.

    Rudi's catch was worth about 0.07 WPA, improving Oakland's chance of winning from 85 to 91% (rounding errors, apparently). OAK was up 2-0 with 1 out and none on in the bottom of the 9th. I guess the real value of the catch is what the win expectancy would be if he doesn't make it. CIN has a runner on 2nd(?), 1 out, tying run up. No idea what OAK's WE would be there -- 75%? So the swing is between ~75 and 91.