You Are Here > > Blog >

SITE NEWS: We are moving all of our site and company news into a single blog for We'll tag all B-R content, so you can quickly and easily find the content you want.

Also, our existing B-R blog rss feed will be redirected to the new site's feed. » Sports Reference

For more from Andy and the gang, check out their new site High Heat Stats.

Albert Pujols’ not very good 3-HR game

Posted by Andy on October 23, 2011

Albert Pujols went 5-for-6 last night with 3 homers and 6 RBI. Not bad for a guy who some thought was washed up.

It was a great game, no doubt, but because the Cardinals already had a big lead when he did much of his heavy hitting, it wasn't worth all that much in terms of WPA.

His value of .211 for the game is quite low. Here are all the 3-HR regular season games from 2011:

Rk Player Date Tm Opp Rslt PA AB R H 2B 3B HR RBI BB SO WPA RE24 aLI BOP Pos. Summary
1 Prince Fielder 2011-09-27 MIL PIT W 6-4 4 3 3 3 0 0 3 5 1 0 0.707 4.614 1.320 4 1B
2 Casey McGehee 2011-08-03 MIL STL W 10-5 4 4 3 3 0 0 3 5 0 0 0.424 4.359 .755 5 3B
3 Jason Giambi 2011-05-19 COL PHI W 7-1 5 5 3 3 0 0 3 7 0 2 0.414 5.519 .596 5 1B
4 Carlos Quentin 2011-05-24 CHW TEX W 8-6 5 5 3 3 0 0 3 5 0 2 0.376 3.767 .976 3 RF
5 Aubrey Huff 2011-06-02 SFG STL W 12-7 5 5 3 4 0 0 3 6 0 0 0.284 5.310 .638 4 1B
6 Carlos Beltran 2011-05-12 NYM COL W 9-5 5 5 3 3 0 0 3 6 0 0 0.284 5.006 .682 3 RF
7 Corey Hart 2011-05-23 MIL WSN W 11-3 5 4 3 3 0 0 3 7 1 0 0.283 6.051 .600 2 RF
8 Chris Heisey 2011-06-22 (2) CIN NYY W 10-2 5 5 4 3 0 0 3 5 0 0 0.252 3.552 .634 1 CF LF
9 Jose Bautista 2011-05-15 TOR MIN W 11-3 5 5 3 3 0 0 3 4 0 0 0.040 2.938 .520 3 RF
Provided by View Play Index Tool Used
Generated 10/23/2011.

As you can see, Pujols' performance would rank as one of the "worst" as compared to this group, not that this is his fault (but rather a credit to his teammates.)

In fact, Pujols' game is the lowest-ranked 3-HR game in post-season history:

Rk Player Date Series Gm# Tm Opp Rslt PA AB R H 2B 3B HR RBI BB SO WPA RE24 aLI BOP Pos. Summary
1 Adam Kennedy 2002-10-13 ALCS 5 ANA MIN W 13-5 4 4 3 4 0 0 3 5 0 0 0.634 4.655 1.390 9 2B
2 Reggie Jackson 1977-10-18 WS 6 NYY LAD W 8-4 4 3 4 3 0 0 3 5 1 0 0.386 4.874 .883 4 RF
3 Babe Ruth 1926-10-06 WS 4 NYY STL W 10-5 5 3 4 3 0 0 3 4 2 0 0.337 4.501 .618 3 LF
4 Adrian Beltre 2011-10-04 ALDS 4 TEX TBR W 4-3 4 4 3 3 0 0 3 3 0 0 0.319 2.585 .810 5 3B
5 George Brett 1978-10-06 ALCS 3 KCR NYY L 5-6 5 5 3 3 0 0 3 3 0 0 0.301 2.708 1.220 1 3B
6 Bob Robertson 1971-10-03 NLCS 2 PIT SFG W 9-4 5 5 4 4 1 0 3 5 0 1 0.254 4.736 .716 5 1B
7 Babe Ruth 1928-10-09 WS 4 NYY STL W 7-3 5 5 3 3 0 0 3 3 0 0 0.241 2.119 1.284 3 LF
8 Albert Pujols 2011-10-22 WS 3 STL TEX W 16-7 6 6 4 5 0 0 3 6 0 0 0.211 5.808 .628 3 1B
Provided by View Play Index Tool Used
Generated 10/23/2011.

In fact, even most of the 118 2-homer post-season games resulted in more WPA than Pujols got yesterday.

This entry was posted on Sunday, October 23rd, 2011 at 2:16 pm and is filed under Uncategorized. You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.

115 Responses to “Albert Pujols’ not very good 3-HR game”

  1. Great use of WPA to temper the hyperbole. Awesome game by Albert -- especially that high heater he crushed -- but obviously not a great clutch game, historically speaking. What are the odds that anyone at MLB or Fox will point this out?

  2. This is why some of the stats are silly. Anyone watching the game saw that at least his first home run put the stake in the hearts of the Rangers who kept coming back. Second, this isn't about whether the game was on the line or not--it is about an individual performance that is among the elite players who ever set foot in a stadium during a World Series.

    Now, if you REALLY wanted to make an argument about the homers, the biggest difference is in the stadiums. I can't recall a series where the nature of the game changed so much because of the venue--maybe the '87 series when the Twinkies beat the Cards 4-3, with both teams winning all of their home games. The current Busch Stadium is not as big as the previous, yet we saw tight, well-pitched games. Now that we're in Arlington, it's silly "gorilla-ball". The same was true in the NLCS. Now, ask any viewer who has watched all three WS games, which ones were more exciting? Chicks may dig the long ball, but the game suffers from these ridiculously designed parks.

  3. Slim & none! :-(

  4. Frank Clingenpeel Says:

    I agree with SABRSteve in his response to Shping {#1} -- except that I would probably trade the "Slim" option to my own category: "Is there any less than none?".

  5. Maybe you're just saying "not very good" to get attention, but a comparatively poor WPA doesn't make his individual feat any less "good." If you want to change that to "clutch," that's fine, but don't be obtuse.

    His total bases were higher than any of the other 3-HR games, as was his batting gamescore if I'm not mistaken.

    So how was it "not a very good 3 HR game?"

  6. Greatest context-neutral individual performance ever.

  7. Right on, Susan. Terrible headline. Like the author says, it's not Pujols's fault the WPA isn't very high. It was an awesome 3 HR game; best ever in the World Series by some measurements. Everyone already knows they weren't extremely clutch, so if MLB or Fox don't point out the obvious, that's fine.

  8. If only he'd struck out instead of hitting those two clutch singles earlier in the game, his WPA would've likely been higher!

  9. SocraticGadfly Says:

    David's right. HR No. 1 was important indeed, even if the Cards already had more runs than they would eventually need.

    Plus, add that that and HR No. 2 both forced Washington to work out more of his relievers.

    WPA is, in many ways, a "hindsight is 20-20" stat.

  10. Susan is totally right-on about this issue. Pujols was awesome last night. He can't be blamed that everybody in the state of Texas seemed to get a hit last night.

  11. Brad, you've captured my point---of course Pujols' performance was amazing, as I said, but from a purely stats-perspective, it WOULD have been better if he had struck out earlier in the game. As David says in #2, it's a shortcoming of the stats in this case.

  12. Sometimes the "but did it matter" argument is useless, and this is one of those cases. It was a World Series Game, away, the other team kept coming back and hitting its own home runs and the pitcher on the mound (for No. 1) was supposed to be the difference maker.

    Albert's performance is incontrovertibly the single best box score for a world series game ever. The context of his first home run and the absolute will of his second two (none of which were pull shot squeakers, but rather went deep and high) were amazing.

    Pujols delivered again and again, and helped send a message from the road team in a 1-1 game three to go up 2-1, almost eliminating the home field advantage for the Rangers and now forcing their hand for the remainder of the series.

    The third homer may not have "mattered" from a win/loss perspective in that game, but I bet it will increase the amount of times Ron Washington looks out to his pitcher and simply holds up four fingers when 5 walks into the box.

  13. Lets not swing the pendulum too far in the other direction. That was still more WPA than anyone else had last night.

    The Rangers mounted some comebacks in the 4rd and 5th innings, but they never got closer than 2 runs. To get crazy-high WPA, you need lead-changes.

  14. Seriously? You witness history that you can tell your grandkids about and all you can do is analyze it to death with a totally meaningless crock of a so called stat. Take a moment and forget about how smart you think you are and enjoy the moment. Whether you are a Cardinal fan or a Ranger fan, what happened in Game 3 should make a BASEBALL fan smile. I never saw Babe Ruth play or Ted Williams either. Cobb was way before my time and Musial retired the year I was born. As a kid I saw Aaron and Mays finish their great careers. I love baseball and baseball numbers and stats but like I said it makes me smile to know that I am privileged to be here in this time following the career of one of the best baseball players in history. So just take a breath and enjoy.

  15. This is why sabermetricians irritate mainstream baseball fans. Thank you, thank you very much for telling us that Albert Pujols' single-game performance last night--five hits, three home runs, 14 total bases--wasn't that good, despite actually being the best in World Series game history, because the game wasn't close when he hit the last two. Well done. Sure did advance the field of baseball knowledge.

  16. I'd like to know why Brett's WPA is so low. He kept his team in the game with those HR's. Otherwise, its a blowout.

  17. I still think the stadiums need to be considered. Duke is right that this was a moment to be remembered, but given that Arlington is the best hitters park in baseball, it takes away from what Albert did. Reggie "The Magnificence of Me"Jackson did his in the old Yankee Stadium, and Babe Ruth's games were both in the old Sportsman's Park in St. Louis, which was not exactly a hitters' park (in the 1920s, center field was 430 ft, with the power alleys at 380 for left and 350 for right.

    I would argue that one of the big problems with comparing stats from generation to generation is that the stadiums are not taken into consideration. A home run in the old 1980s Busch Stadium or the Astrodome was a much bigger deal than something hit out of Arlington or Miller Park.

  18. Oh come on David, the Pujols homers traveled 423, 406, and 397 feet. Meanwhile one of Reggie's barely squeaked over that homerun porch in Yankee's right field.

    Look at specifics, not generalities

  19. "Meanwhile one of Reggie's barely squeaked over that homerun porch in Yankee's right field."

    Meanwhile, his last one was 450 feet.

  20. Indeed, but my point isn't that Reggie was a wimp, my point is that David's "Reggie and the Babe were more impressive because of TBIA" point doesn't hold up to close scrutiny

  21. I'm not sure why all the distress at WPA as a stat here. It's just telling us what anybody watching the game knew at the time. The first homer was very important, and WPA tells us that it was the biggest play of the game. We also know it wasn't exactly "game-changing" because the Cardinals did have the better chance to win the game even before Pujols hit it. If you know anything about baseball, these are not surprises. veryone also knows that the last two homers were just icing on the cake in a game that at that point was almost certainly a Cardinals win. All WPA does is articulate in numerical terms these utterly non-controversial observations. There are lots of other stats that can express how great Pujols' game was in context-neutral terms, but there is nothing wrong with having one stat that expresses numerically what we all know about the limited clutchiness especially of those last two homers.

    Of course, Andy's post title was provocative, and perhaps ironic, shorthand headline writing. More accurately it might have said: "Pujols' relatively limited clutch contribution compared to the average three-homer game". That's really what WPA is telling us here, and it is non-controversial. It is, after all, very rare for any player to hit three homers in the late innings after the hitter's team has already taken a lead, and that's why the game produced a relatively low (though still impressive in an absolute sense) WPA for a three-homer game.

  22. WPA? Holding Albert up to scrutiny? In the "bigger pucture", he's SECOND (2nd !) all-time among RH hitters with 6000 career plate appearances in OPS+ behind a guy named Hornsby and ahead of Mays, Aaron, Robinson, Greenberg, Manny, Dick Allen, etc......

  23. @2: "but the game suffers from these ridiculously designed parks"

    By that logic, I bet you are unimpressed by the "Shot Heard 'Round the World."

  24. "Andy's post title was provocative"

    Stupid. That's the word you're looking for: stupid. And the entire entry was generally stupid and pointless: we don't need a stat to tell us that Pujols' second and third homers came after the game was mostly decided. The writer did his best to cheapen and denigrate Pujols' accomplishment, and he fails.

  25. Andy, perhaps you should put "Tongue in Cheek" in the title to prevent further outrage.

  26. SocraticGadfly Says:

    Ding, ding ... Vidor #24 wins the posting! Now, can we all admit Andy stated the obvious, wasted cyber-ink when he could have probably found some other new angle, like doing a game search for anybody else hitting 2-out/9th-inning home runs, etc., on Pujols' performance?

  27. This is so silly.

    Albert Pujols hit 3 home runs in a World Series game that gave his team an 2 to 1 games edge and we are comparing this to 9 regular season games? Come on, lets get real here. Comparing the World Series to a regular season game is a joke. Obviously someone has way too much friggin time on their hands to somehow take away from the accomplishment.

    In regards to the Reggie comment: the guy saw 3 pitches (one each at-bat) and hit 3 homers off of three different pitchers. That is a damn special accomplishment.

  28. #21 Birtelcom, well said, but I disagree on one point. If anybody watching the game knew at the time that the second and third homers didn't mean all that much, then some of them forgot before they posted comments on this thread. It was NOT the greatest World Series performance in history. Not even close. That said, it was awesome and amazing.

    Then Vidor at #24 repeats your mistake, by failing to read that some of the people who commented before him did not know what he says is obvious.

    Some of you seem to be taking it personally. My post is simply pointing out that his overall performance wasn't worth a lot from a WPA standpoint. In my original post I still call his performance amazing and point that that the low-ish WPA is in no way his fault. I'm not trying to rob him or the moment of any of its appeal or appropriate acclaim. I'm simply not going to drop on the FOX bandwagon of naming the latest impressive achievement as the greatest of all time.

  29. Gonzo...that "shot heard round the world" which mattered only to NY fans, was only 315 feet.

  30. Well then don't use WPA as your only crutch. WPA is incredibly flawed, even from a clutch standpoint. Take this example from April of last year.

    The Cardinals are losing by 3 in the ninth. Pujols hits a two-out, two-run homer to put the Cardinals within one. Then Holliday hits a solo shot to tie the game.

    Pujols made his team 2% more likely to win the game.
    Holliday made his team 36% more likely.

    Pujols went 3 for 4 that day with 2 HR and 4 RBI.
    Holliday went 1 for 5 with a solo homer.

    Pujols' WPA for the game was .062
    Holliday's WPA was .239

    Would anyone conclude that Pujols' day was "not very good?" Would anyone conclude that Holliday was "better" that day. No one. Except maybe someone who used one statistic to the detriment of others or someone with an axe to grind against Pujols.

    And that's clearly what you're doing here. I can't wait until you follow up by explaining that Pujols' performance in Game 3 was inferior to Kinsler's in Game 2 on the basis of their WPAs

  31. "It was NOT the greatest World Series performance in history. Not even close."

    This guy just doesn't get it. BBRef might want to think about getting a better writer for its blog. Yes, three home runs and 14 total bases IS the greatest single-game WS performance by a hitter in history. That's an objective fact, and anyone who can't figure that out probably shouldn't be writing about baseball in a public forum.

    "I'm not trying to rob him or the moment of any of its appeal or appropriate acclaim."

    And this is simply a lie.

    It is interesting, actually, to see someone who professes to appreciate the game from a sabermetric point of view to cling so desperately to an entirely context-dependent statistic to form an argument. Every other time we read one of these sabermetric articles they tell us that context-driven statistics are deeply flawed as a way of measuring players. They tell us that we shouldn't assume the 20-game winning pitcher is the best, because that's driven in large part by how good his team is; they tell us that Ryan Howard shouldn't have won that MVP because the eight billion RBI he had were driven largely by how many of his teammates reached base ahead of him; they tell us that we shouldn't assume that the guy with the fewest errors is the best fielder because he might have the range of a tree stump and thus never has the chance to commit errors on the balls that are flying by them. But now when we ask ourselves who had the best night for a hitter in World Series history the guy that Baseball Reference is paying (is he getting paid?) to write its blog tells us that the only way to determine who had the best hitting night in World Series history is determined not by what the hitter actually did, but by what the game situation was when he did it.

    No one ever hit more home runs in a World Series game than Pujols did last night and no one ever had more total bases in a World Series game than Pujols did last night. Yeah, I'll take him.

  32. @Vidor: Go fly a kite. If you don't like his writing, don't read it anymore.

    Sorry, Andy. These type of comments make me angry. Saying you should be 'fired' from writing about baseball. This site is so much fun. It's a shame that when someone has an opinion that differs from yours or the stats, they tend to get angry and violent. I love the writers and the commentators and the stats and opinions. I am constantly refreshing to see if a new post is up. Don't ever change.

  33. Just adding to the argument that WPA is flawed is the idea the Adam Kennedy's 3 HR game was that much more valuable than Beltre's. Adrian Beltre hit three solo home runs in a 4-3 win. The other run was Kinsler's home run. 4 run scoring plays in the game, all completely needed to win the game and Beltre produced three of them. Kennedy hit a solo home run while losing 2-0, then hit a solo home run to make the score 3-2, then hit a three run home run to give his team the lead in the midst of a ten run inning. If Kennedy singles or walks, they still score ten runs, if he strikes out, they still score 8 runs and even if he hits into a double play they still manage to tie the game before a third out is recorded.
    And anyone who thinks that Pujols would have had a better game by making outs is just trying to create arguments.

  34. "Your opinions do not agree with mine, and my opinions are obviously never wrong. Therefore, your opinions are wrong and you should be fired for merely allowing them to enter your brain, much less publishing them to the world."

    I love you, Internet commenters. Keep getting your panties in a bunch over baseball, it makes reading stuff like this that much more fun for the rest of us.

    I think Pujols's performance is the greatest in World Series history, but I wouldn't personally attack anyone who disagreed with me. It's better to pick your battles and ignore being so aggressive when it's neither necessary or appropriate.

  35. the last one of reggie jackson's homers in game 6 of the 1977 series was like albert's, mere icing on the cake that made the score 8-3 i believe. one was a 357-footer to right-center field that was a low-liner double in most other parks.
    i think his 2nd homer was also not in a close contest, i could be wrong. but what made jackson's HRs so unique was that they were on 3 straight pitches and swings. and being done in NY in the series-clincher against the other major market and traditional rival added to the lore.
    regardless, the cardinals scored enough runs without albert's home run barrage to win. he gets his hits in bunches, usually has the last few years. that is part of what makes his batting "average" misleading.

  36. Off subject a little, but did anyone catch Tim Tebow in Miami today? The most exciting pro football game I've ever seen. The key with Tebow is he believes he can be successful, and he feeds off the haters.

  37. Anyone who uses WPA as a reference of the quality of an individual performance is never, ever, ever allowed to complain about the overemphasis on RBI ever again.

    Both statistics emphasis context/situation over individual performance. The only advantage WPA has is that it's more complicated.

    Albert Pujols had the worst game ever by a player who had the best game in World Series history.

  38. @Timmy P:
    I did, and thanked God (Tebow himself) that I don't have a rooting interest in any of those teams. Easily the worst football game I've seen in recent memory. Two teams entirely inept for 75%+ of the game. Not sure what you find exciting in that, I truly hope that's sarcasm, otherwise we were watching two very different versions of the same game.

  39. Patrick Greene Says:

    Seriously? Not a very good 3 homer game. Is there such a thing? Is this the Onion without the humor part?

  40. This is a little crazy. Pujols had a great game-there's no reason to minimize his accomplishment. But let's get real, if this were a playoff basketball game, and he scored 25 of his 50 points in the 4th quarter, after his team had a huge lead, we'd be calling some of those points "garbage time points".
    If you read this blog, what you find is every different type of opinion-on steroids, on who is a hall-worthy, who should get the various post season awards, the individual decisions managers and even players make in a game, poor Adam Dunn, worst contracts, whether Paul Konerko and Michael Young are the greatest players in history (sorry, Timmy P.), etc etc. This is a terrific site (and my opinions have been taken to the woodshed more than once). Why not stimulate discussion with something a little off the beaten track?

  41. @38 The last 5 minutes of the game was very exciting Max, you're right it was a snoozer for 3.5 quarters, but if you're a Broncos fan or a Tebow fan it was exciting. It's funny the TV analysts keep saying Carson Palmer is just rusty, but Tebow's bad start was because he's terrible. Double standard?

  42. @40 some good points, but garbage time in basketball and baseball are very different. Texas can score runs in bunches, ask Detroit about the 9 run inning they had.

  43. @40

    Because, as someone who loves statistics, we're always fighting an uphill battle. If you use a statistic to say something out of the mainstream, that's awesome - it stimulates discussion, and has the potential to change the way people think. Sure, often it will provoke powerful negative responses, but that's always part of challenging the mainstream opinion.

    However, when statistics are misused, to explain a narrative that wasn't there, those powerful negative responses are magnified, because they also happen to be correct. Trying to take away from what Albert Pujols did last night is going to be silly. The low comparative WPA is not a reflection on what Albert Pujols did at all. If Cardinals pitchers had given up 14 runs, his WPA would have been higher. That's not something Pujols has any control over, and therefore, using it to measure his individual achievement is wrong.

    Look at it this way. Would it be acceptable to talk down Babe Ruth's 1928 game because he had only 3 RBI in his 3 home run game? Of course not - that's stupid. Well, RBI is a reflection of context and opportunity, right? So why is WPA an acceptable measure of individual performance, but RBI isn't?

    Using WPA to say Albert Pujols didn't have a great game last night might promote conversation among the posters on this board, but it makes those who are critical of statistical metrics have another argument why they're no good. I can see it now, something like "those myopic statheads, they don't even think Albert Pujols's Game 3 was impressive because he had a poor WPA." And they'd be right.

  44. @43 I was going to post my thoughts in this thread but you stated them perfectly, so I will simply add a "well said".

  45. yeah-hah, bad stuff last night, man, 3 taters in one WS game, not so speical

  46. "I think Pujols's performance is the greatest in World Series history, but I wouldn't personally attack anyone who disagreed with me."

    We shouldn't get to disagree about facts. People are not entitled to their own facts.

    Otherwise, Jim Dunne @43 did a great job of explaining just how poor this blog entry was.

  47. @Vidor: That's true, but people interpret the facts in very different ways. For example, just as Pujols's WPA was very "pedestrian," if you will, because of a variety of other factors, he also had the highest RE24 in World Series history, at 5.808. In fact, the only other RE24 of 5 or more was Ryan Howard's Game 4 of the 2008 World Series. If you only look at WPA, it might not seem impressive, but looking at other stats paints a fuller picture. You look at the .211 WPA compared to all World Series performances and it's nothing great, but look at it compared to everyone else last night and no one was higher.

  48. @46, Not to get my head handed to me twice. I'm not arguing that Pujols didn't have a terrific game, and it is objective fact he had the most total bases in a World Series game. Does that objectively make it the greatest hitting (WS) game of all time? I don't know. We contextualize everything-there are people out there who claim Schilling's "bloody sock" game was the greatest pitching performance. Would I be more impressed if the Cardinals won the game 6-5? Yeah, maybe. That's just a human reaction to a single game in which the value of each individual hit could be perceived as somewhat less than a hit in a pitchers duel.

  49. Part of the delight of baseball for me is its complexity. Words like "greatest" or "not very good" don't summarize all aspects of a performance. If you are interested in complexity it is helpful to have different stats to describe different aspects of a performance. WPA describes something different than RBI, something different than counting runs created, it counts something that I find very interesting to count -- the degree to which a hitter improved his team's chances of winning.

    WPA doesn't claim to do anything else, but it does do that particular thing pretty well (better than RBI, for example). As compared to most other games by a hitter with the 14 total bases, this particular performance increased the hitter's team's chances of winning by a relatively small margin. Compared to everyone else in Game 3, Pujols had a high WPA -- that is, his PAs in total improved his team's chances of winning more than anyone else's total PAs in the game. But compared to other games in which a batter has 14 total bases, the degree to which he improved his team's chances of winning was, relatively, lower.

    That is all that WPA is telling us and all that it purports to tell us. If you don't find that particular subtlety in understanding Albert's performance interesting, then ignore it. I happen to find that understanding and being able to measure different subtleties of a particular performance, such as this particular one, adds an extra element to my enjoyment of the game. It is not a yes or no answer to a question "was this performance 'great'". It just tells us that in this particular aspect, Albert's 14 TB performance was different than most others in a specifc, interesting way.

  50. @41/Timmy P:
    By the point, I just considered it a foregone conclusion the Dolphins would find some spectacular way to blow it, and they didn't disappoint. It's funny, they needed everything to go wrong for them, and everything did. Between Sparano's decision to go for 2 and failing (an XP would've required two 2-point conversions), the fumbled onside kick recovery, and the complete defensive ineptitude in no one covering the QB sneak on the tie. It was fascinating watching two teams who apparently were both trying to Suck for Luck and seeing who would win the crap-fest. If that wasn't either team's plan, Tebow's play through 3 quarters was disgraceful, but the Dolphins defense allowing him to just turn it on like that was even more so. They had a lot of sacks, I'll give them that, but a competent defense should've obliterated Tebow. Several times I saw him escape linemen rushing in to sack him - most notably when they probably would've gotten a safety deep in Broncos territory - and he evaded capture for a significant gain, as opposed to a loss that would've buried the offense even more. To me, that says far more about poor defense than Tebow's skill. I still need to see him have prolonged success - a whole game, or even just a half. He was overthrowing receivers so much the first 45 minutes not even the defense on the worst possible coverage could have possibly intercepted them. Then in overtime, Denver barely inside the Miami 40, and yet they hand the ball off all 3 downs for gains of no more than one yard, setting up Prater for a field goal of 52 yards, despite his missing two shorter field goals earlier in the game. Even though he made it, that's incredibly suspect play-calling. Granted, I thought Tebow was going to throw a pick, but I'm very skeptical. The Denver coaching staff obviously isn't, or they would have put Orton in - far and away the better QB, in my mind.

    Overall, it wasn't a few minutes of Tebow that made the game awful, but everything surrounding his play then - defense, play-calling, QB protection, routes, coverage, efficiency, etc., by both teams.

  51. Let's admit it. The real hero of yesterday's game was the 1st base umpire, not Pujols.

  52. @50 We'll see Max, but Kyle Orton is not better than Tebow. Kyle Orton has adult-cartoon-watching-loser written all over him. He's droopy and depressed looking , he might throw a prettier spiral than Timmy, but his record speaks for itself. GO TEBOW!

  53. @49

    I agree and disagree. WPA isn't the problem. It isn't, on it's own, a "bad" statistic, but it is an extremely easy one to misinterpret. As you said, it measures "the degree to which a hitter improved his team's chances of winning." That is a far cry from measure who had the best individual performance.

    The parallel I was drawing with RBI wasn't meant to say that they do the same thing, but to draw the comparison between a traditional statistic that is often misused, and a modern one being taken out of context. RBI has been used for determining the "most valuable" baseball player, which is the wrong way to use RBI. WPA was used in the original post to denigrate Pujols's individual performance in Game 3, which is an incorrect use of the statistic. Like you said, "WPA doesn't claim to do anything else." This post used WPA to measure something that it doesn't. That's not WPA's fault.

  54. Victor @43: Whether something is "the greatest" is not an objective fact. The concept of "the greatest" in a baseball sense is filled with ambiguity and subjectivity. Only when one uses the term "the greatest" in a mathematical sense ("Seven is the greatest number in the sequence of whole numbers one through seven") is an objective fact in question. In a baseball sense, "the greatest" is like "the most beautiful" -- subject to many different meanings and interpretations.

  55. I love baseball.

  56. i think the game 3 1st base umpire (ron mea kulpa) did a nice job of more than evening things up behind home plate in game 4. he gave holland both sides of the plate and encouraged the cardinals to extend their strike zone. tone was set early with several ?able strike calls up and in to right handed stl batters. even in the 9th inning, strike 2 was clearly outside to pujols, making it 0-2 instead of 1-1 and helpin gmake a comeback less likely. i am sure FOX likes it being 2-2 instead of 3-1 as well...hey can the cardinals get don denkinger behind home plate for game 5 to make up for the 1985 Series he blew too?

  57. @49 "... As compared to most other games by a hitter with the 14 total bases, this particular performance increased the hitter's team's chances of winning by a relatively small margin." " just tells us that in this particular aspect, Albert's 14 TB performance was different than most others in a specifc, interesting way."
    As compared to what other 14 total bases in a World Series game?
    In any aspect Alberts performance was different in that it is the only one that has ever happened.

  58. This sparked an argument on a different message board I post to. I'm right in line with comments #14, 31, and 43. And "Albert Pujols had the worst game ever by a player who had the best game in World Series history" is great.

  59. @52/Timmy P:
    Well, QB wins are even less meaningful than pitcher wins. I point to Orton's history of success in the NFL as evidence that he's better, and I think he coulld have done all of the things Tebow has done, or more, for example, not putting them in a 15-0 hole 90% into the game against one of the three winless teams in the league. But like basically everything else in sports, this is subject to much subjectivity. I suppose we'll see how he does against a much better Lions team. Meanwhile, we're in a spirited discussion on football on a baseball blog. Weird. GO ORTON!

  60. There is a good consensus here that the context of what others do should not effect how good a game they had. I am surprised another point has not been introduced.

    That WPA overwhelmingly only measures "clutch" performance in the sense of how key to victory a particular play is. besides the major caveat that everything else done adds to the likelihood that either team can win, there is very little "clutch" play in the true sense. True meaning what the words INTENDS to communicate.

    Which has always been about the ability to produce better in stressful & crucial situations. But most know that the variation occurs overwhelmingly in a random way. As if you measured what a guy did in the 4th inning on Tuesdays, there will accidentally always seem to be clutch & choke performances. Like how streaks, shooting, hitting-are almost entirely a product of BA, BB (makin hits & having opportunities) mixed in with luck.

    Most know that even many things measured over a whole season can be overwhelmingly luck, like BBIP. We could run an algorithm about how much things like this vary, & how long it takes stats to return to the mean. Sometimes a guy may do something worse technically to produce weak contact/poor results, but usually it is chance if a ball is fielded.

    I recall reading an article on Dugout Central about clutch performance-over a career. Using frequently enough occurring categories, like RISP & late & close, that you get enough chances so that random variations are minimized. In that case, almost nobody had a statistically significant variation from their own mean!

    Mantle did something over 10% better in these situations In the WS he did only 93% of his career OPS +: likely about what most did w/the better pitching. But most all had minimal variation from their usual performances. Nobody hit 15% better than their usual performances.

    So if WPA is to be more than a curiosity, at least in the sense most assume it is meaningful, that is, reflecting REAL, full bodied, can ratchet up production in stressful situations, it needs to be considered over an adequate # of chances to reflect any of this.

    Even if used to show who did the best in terms of having the biggest effect on winning, unless you show that the guy normally did that, or that a player usually does worse when the outcome is not in serious doubt (also rarely true over a large enough sample size-

    It is always misleading if it is not understood that these varying performances are almost always due to chance, not reactions to pressure. Ironically, the context dependent stat's core assumed meaning is unseated by...a larger context.

  61. Some get too worked up about the conflation of indivi8dual performance with group & context dependent efforts, even if they are right.

    But Andy: I read the link where you wrote "some thought Pujols was washed up". Nobody said anything REMOTELY like that.

    One guy said it would be ironic IF he started declining in a contract year. Sean thought he was a year or two away from any significant decline. Nobody else even referenced someone suggesting that Phat Albert was anything like washed up.

    Memory has been shown to be far less reliable than we think it is. Even eyewitness reports vary tremendously, when folks are surveyed right after they occur! let alone in criminal cases.

  62. shaqfearsyao Says:

    Am I the only one who still can't believe Adam Kennedy hit 3 hr's in the same game?

    Anyways, sometimes stats don't tell the whole story about how someone is doing. Those hr's that Pujols hit demoralized the Ranger's team.

  63. 56) Umpires who messed up a crucial call should be chased out of town immediately, and should not be in the game with the team which was wronged for at least one full (not calender) year.

    They think they can get away with anything with a cheapo apology. What I can't understand is why the managers cower down to them.

  64. shaqfearsyao Says:

    to 63:

    Ron Washington should've appealed to the home plate umpire because he obviously had the better view and could've called Holliday out. I don't know why Ron didn't appeal and it cost them the game. Baseball needs instant replay because you hate that the costly error by the umpire played a big role in the Rangers lost. Bud Selig is clueless as a commissioner.

  65. For a bit more context, since 1950 there have been 112 regular season 3-HR games (among those games with PBP data) where the batter compiled a 0.211 WPA or worse.

    The lowest WPA is -0.123 by Dave Kingman in the Cubs memorable 23-22 loss to the Phillies at Wrigley on May 17, 1979. Kong went 3 for 6, with 4 runs and 6 RBI.

    Among players whose team won the game, Jeff Bagwell managed a -0.002 WPA in Houston's 13-4 pasting of the White Sox on June 9, 1999. Bagwell also went 3 for 6, with 6 RBI.

  66. Ortons record with Denver is 12-21!

  67. Mark in Sydney Says:

    This is all kind of funny to me. Having played a little competitive non-professional ball, the ability to hit a home-run in any context is a major achievement. To be able to hit one in a major league is an amazing skill, given to very few. To hit three in a World Series game? Let's see. There are now eight people who have done it. Ever. Superman status I think.

  68. You are getting more moderate Cab, that is good. Now maybe express some baseball sentiments unrelated to an umpire obsession man!

  69. Well, I appreciate the poster for writing this article. The sportscasters were all saying that Pujols' game topped those of Reggie and the Babe, but I don't think it did, and this is why. Yes, Pujols racked up more hits and more total bases, but the idea is not simply to count up how many homers a guy hits in a game. Hitting three homers in a World Series romp is awesome (in the original sense of the word), but a player's performance is more meaningful to the extent that his contributions impact the game. I would take Reggie's game over this one - three homers in three at-bats, two of which were impactful, and a leadoff walk (and run scored) in his only other plate appearance. The chart above suggests that if you consider earlier postseason series as well, Adam Kennedy's performance 25 years later easily supplanted Reggie's as the most impactful 3-homer game ever.

  70. Thomas Court Says:

    Boy Andy is sure taking it on the chin for this one... Undeservedly so I believe. I agree with the poster who said don't change a thing Andy.

    I have a new stat. It's called, "Watch-ability Probability Added" - or WPA for short. It simply indicates in percentage form how much more watchable any individual play makes a game.

    I saw the first home run - a titanic drive to be sure. And the second one looked just as majestic. The beauty of Albert Pujols when he connects is something to see. And a lot of Texas fans were treated in person to what National League fans have been spoiled to for the last eleven years.

    The third home run? I missed it. Want to know why? Because at the same time Michigan State was winning their game against the undefeated #6 ranked Wisconsin Badgers on a HAIL MARY pass with no time remaining. At the same time, the Texas Tech Red Raiders were storming out to a 31-7 lead against the #3 ranked Oklahoma Sooners in a game that would eventually end up 41-38. Those two games had much higher WPA scores than a World Series game that ironically was ending with a football score.

    Why don't we credit the player who swipes second in a 5-1 game in the 9th inning? How do you explain that to a viewer who is just understanding what a stolen base is? "Well, that's called Defensive Indifference... because the defense doesn't care that the runner is stealing... so the runner does not get credit for a stolen base." Well, the third home run falls under Viewer Indifference due mainly to a 15-7 score and much more "Watchable" games going on in the world of sports.

    I love baseball. My library is chock full of baseball books that I have read and re-read ceaselessly. I often put on Ken Burn's Baseball as background noise when I lay down to sleep the way some people put on sounds of ocean (I am not kidding about this). But that game had lost all of the drama. The Watchability Probability had shrunk to zero, while a couple of HUGE college football games were dramatically ending and beginning at the same time.

    WPA is not a super nerd stat. It doesn't roll off the tongue the way batting average or ERA does - but it is just as easily explained if the person you are explaining it to understands - Hmm... I don't know... MATH.

  71. @70

    So, for the record, you weren't watching Albert Pujols dominate late in the game because he had dominated so much early in the game that it was barely a game by the end.

    I'm not sure what "WPA is not a super nerd stat" means. I'm something of a stat nerd, and I do like WPA. I just don't like using it to measure who had the "best" individual performance in the context of one game. That's not what it's meant to measure. Allen Craig going 1 for 1 in Game 2 nearly equaled Albert Pujols going 5 for 6 with 3 home runs in Game 3. Does that mean WPA is "flawed" on its own? No - but it does mean that it doesn't necessarily measure "best."

    Also, it's not as easy to explain to a casual fan as batting average or ERA, because it can't be calculated easily. (Aside - Is there ANYTHING as easy to explain as batting average? People who don't even follow baseball will understand references to "batting 1.000". ). By following WPA, you're taking sort of a leap of faith. If a player goes 1 for 4 with a double, we KNOW that his batting average is .250 and his slugging percentage is .500. It's not arguable. What's his WPA? There's no way to know. Reasonable people can disagree about win probability, which makes it less concrete and therefore less easy to explain. Don't be condescending toward people who either don't understand or don't agree with it by inferring that they don't understand math.

    Again, I like WPA. I think it's cool to look at the graphs to see the most impactful plays of the game. But that's what the tool is for - looking at individual occurrences that changed the direction of games, and also for looking at "trends" - i.e. is a players long term WPA a lot higher or lower than one would expect based on his production, a measure of "clutchiness" as my old friend Dan Smith used to call it. As #60 Mike Felber pointed out, that's only useful over a statistcally significant period of time. It's not a measure of "good" or "bad." It's simply not.

  72. @2...still feel the same after last night's 4-0 by Holland?

  73. need to watch more pro football.

  74. I've got a solution to this WPA crises. Let's bring in Bud to decide who wins the argument and what compensation they are due. That is, to say, if we can't resolve this by November 1.

  75. @ David #17:
    Re Sportsman Park, is this the same version of the Sportsman Park where George Sisler in 1920 hit .473 with an OPS of 1.279 versus .341/.880 on the road and Hornsby in 1924 went .469/1.333 versus .370/1.055 on the road?
    I'm just asking, but it seems several Browns (Ken Williams/Jack Tobin) really preferred the home field. Almost seems like a Coors Field type of differential.

  76. I loved this comment, posted during last night's game on my other message board (alluding to this thread): "Can anyone tell me if that home run was important?"

  77. Since the Rangers only scored 7 runs. You "could" make a case that none of the runs for the Cardinals were important after they scored their 8th run In the 5th inning.

  78. shaqfearsyao Says:

    When Pujols retires from baseball and people look back on his postseason career, they're going to think of two things.

    1. His hr against Brad Lidge
    2. 3 hr's in the world series.

    Nuff said.

  79. I'm with #24 and a couple of the other earlier posts.

    This is a stat orientated site with a stat orientated audience, so the author is clearly 'writing" to his audience, but some things are better left unsaid.

    Pujols' WPA would have been higher if he had struck out earlier?


  80. @70 - Just because you decided to miss out on history doesn't lessen the accomplishment. And just because the third home run didn't have much to do with the outcome of the game doesn't lessen the accomplishment. Using WPA to determine the worth of the accomplishment is like measuring someone's weight with a yardstick. Andy's contention from the start was that if Albert Pujols had made outs instead of singles in his early at-bats, meaning the game was closer when he hit the home runs, he would have had a better statistical game. This logic is counterintuitive to the game and so it is flawed. Your idea that it just wasn't worth watching anymore is just a matter of personal opinion.
    I do however believe WPA is flawed and here is why: Luis Castillo hits a foul ball into the stands and the WPA doesn't change. But everyone on the field, in the stands and watching at home knew that moment was key.

  81. @78, not to mention getting a hit in four consecutive WS innings. Obviously you need a lot of help from your teammates to even have a chance to do that, but it's still pretty cool.

  82. @80
    No. Nobody is claiming that high WPA equals better "statistical game". At least I hope not.

    WPA measures something very specific... the running in-game win probability. It is what it is. Nobody has ever proposed using it as an "uber-stat".

    You're all beating a dead horse and the dead horse is a straw man.

  83. @82

    At least three times this year, Andy has made a post referencing "best" or "worst" in the title, and then used WPA as a statistic. He's obviously doing it to be provocative, but it's not a straw man we're beating on.


    That moment was key in hindsight, but there were several other measurable things that happened on the field of play during that chain of events. No statistic can properly measure those outside influences. That doesn't mean those statistics are flawed - it just means that every statistic in the world isn't a substitute for the joy of watching the game.

  84. You know that last home run by Pujols, the one in the ninth inning?

    It wasn't at all important, because it turned 15-7 into 16-7. The game was already over.

    It was very important, because it was only the fourth time that anyone's ever had three home runs in a World Series game.

    Hey, if the physicists can cope with light being both a particle and a wave, then we can cope with WPA and non-WPA. They both give valid insights into the same situation.

  85. @65

    It's gratifying to note that Dave Kingman also holds the record for the highest WPA for a 3 HR/6RBI performance, giving +110% against the Dodgers in 1978.

  86. @80, Pauley: "And just because the third home run didn't have much to do with the outcome of the game doesn't lessen the accomplishment."

    Without referencing WPA at all, I think the circumstances of the game do lessen the accomplishment somewhat.

    I assume all would agree that, if Albert's last HR had come off a position player, that would lessen the accomplishment.

    Now, of course, Mike Gonzalez and Darren Oliver aren't position players; they're good pitchers. But they're lefties, and I don't think either one would have pitched to Albert had the game been competitive.

    Surely Ron Washington would not have let Mike Gonzalez face Albert after pitching to 5 other batters over 2 innings. Gonzalez allowed a .900 OPS to RHBs this year.

    Look back to Game 1: 7th inning, Cards batting and leading 3-2, Gonzalez comes in to face Jon Jay, but then Scott Feldman comes on for Pujols and Holliday.

  87. Phil Haberkorn in Indiana Says:

    Roaming through this debate, I wonder if we need to consider that at times, the greatness of this game, its players, and its moments are sometimes beyond stats.

    Reggie Jackson hits a home run with his team trailing by one run, and puts them ahead by one run.
    Reggie Jackson hits a home run and puts his team ahead by four: the opponents could tie the game with a grand slam.
    Reggie Jackson hits a home run and puts his team ahead by five: the opponent now needs a grand slam plus another home run to tie. Without home runs, the opponent needs to get even more hits in order to catch up.

    Reggie Jackson did this in a World Series-clinching game, on three consecutive pitches facing a different pitcher each time.

    Nothing tops that.
    Nothing even comes close.

  88. Casey McGehee hit 3 home runs in a game this season. Wow. Just think how bad his stats were in the other 154 games he played in.

  89. @86 John Autin,

    By the same token, if, by the time Albert came up for the sixth time, he wasn't already 4 for 5 with two homers, 3 runs scored and 5 RBI, maybe the game is in hand and Ron Washington is still managing for that day. The fact that he was dominant throughout the game, making his later-in-game performance have a lower WPA, is what is off-putting, to some, about the statistic.

    In short, the biggest reason that Albert Pujols's ninth inning at bat didn't impact the game was the way he played in innings 1-8. If we're going to say that "lessens" the accomplishment, no wonder people who aren't into statistics think we're crazy.

  90. What the stat needs, then, is some kind of adjustment for the fact that your team is far ahead because you got them there--something like the idea that unearned runs don't impact a pitcher's E.R.A. unless the error was made by the pitcher himself. (And I can't even remember if that's the real rule, or if it was an idea of Bill James'.)

  91. There are just a lot of sensitive people these days. Seriously, an internet posting causes your appreciation of Pujols' feat to be diminished? You can't handle it? Is your whole world-view crushed, you're raison d'etre shattered?

    The title is actually quite great. You see it and go 'huh' instantly, click the link to read the story and then go 'aha', I get it. Albert Pujols had an all-time performance...but it wasn't really that clutch. And in all the aftermath of his great game, you know what, I didn't know that. And so I get to read about something I wouldn't have known or even thought about. It is what it is.

    You must not criticize the Pujols. Love the Pujols. Exhalt the Pujols.

  92. No offense intended, but you needed this thread to tell you that home runs hit when the score was 12-6 and then 15-7 weren't exactly clutch? I think one of the points that some of us are trying to make is that we don't need an abstruse statistic to quantify this very obvious point to us.

  93. I'm just going to bring up a point that may have already been brought up, to be honest, I'm to busy to read through everything now, but even though he has the lowest WPA of everyone on the 3 home runs in a post season game list he had the following: Most hits, RBI and Runs.

    I think that he still had the best single game post-season performance for a hitter regardless of his WPA.

  94. On Saturday night, I got to thinking that someone was going to try and sound smart by using the single most trumped up "stat" in baseball. Do you HONESTLY believe that Pujols didn't help them win? Really?

    Aside from that if you wanted to use a sebermetric stat to evaluate the game, why not RE24? Oh, I know, because that would be the highest in WS history, and second highest in postseason history. And just in case you were wondering, he also had the fourth highest.

    I like B-R and the blog, but this was really a silly, silly, article. This is not the sort of thing that should make us question the value of the game played; it should make us question the value of the formula making the stat. If the most productive game in the history of baseball is only the 300th+ in the history of the series, then you have to think it's the stat, not the game that's off.

  95. #79,

    Chuck, of course it's a stat oriented site, but the use of the stats is misleading. Hypothetically say a QB literally leads his team to a touchdown every time they have the ball. He does so for an entire season, putting together massive numbers along the way and a perfect QB rating.

    Then someone said it wasn't that impressive because there weren't any fourth quarter comebacks or game winning drives.

    The point is that the REASON the QB didn't have any fourth quarter comebacks or game winning drives because of his excellence. You can't use that against him.

    That's tantamount to what is being done here. Pujols was so potent that it literally brought DOWN his WPA.

    That's why, as a stat geek myself, and subscriber to this site, it bothers me that so little thought went into this "number crunching." I expect MORE as a part of the target audience.

  96. Phil Haberkorn in Indiana Says:

    Billy Williams would occasionally strike out with nobody on base in the first inning, to set up the pitcher for later in the game, to get the same pitch back when there are runners on base and a hit can be more productive. As I recall, this worked best with rookie pitchers.
    I recall sitting in the grandstands a couple of times, and we all KNEW what pitch was coming in the late-inning at-bat, and sure enough, Billy socked it.
    Is there a stat to account for that?

  97. Somehow Brett lost his game, but his homers contributed more to his team than Pujols's did? So much for SABRmetrics

  98. 47


    There is a distinction though in the wording. "Not very good" implies that in order to be "good" it needs to be clutch. There are implications that it is the "worst" and that there are a good many two homerun games that are 'better".

    Using words like "clutch" would engender a lot less controversy. Language is essential in composing an argument. Better and worse mean different things that "clutch" and 'not clutch."

    The reason people are bothered isn't so much the use of WPA, it's the use of a stat that doesn't support the argument he's making based on the language he's using.

    Some call "provocative." Others call it "stupid." I call it fallacious. Particularly that of affirming the consequent.

  99. Would we be arguing this hard if Andy had entitled his post something like "Interesting Statistical Quirk to Albert Pujols' Great Game"

  100. Mike,

    If he had title it "Albert Pujols not so Clutch 3 HR Game" it would have gotten the message across.

    I think there is some case made that until then it was a "anything you can do I can do better" game with both sides putting up 3+ runs in 5 straight half innings. The Rangers only got one more run after his first HR.

    Sometimes HRs aren't measured in distance, they are measured in how much they take out of another team. That was a MIGHTY shot, (a heck of a lot further than 423 feet I assure you, one ESPN writer suggested they forgot to measure the other half) and it pretty much crushed the Rangers as much as the ball.

    Still, the second and third home runs were far from clutch. That doesn't take away from the magnitude of the performance though.

    Honestly, the article comes across as someone who is either a "hater" or someone just trying to garner attention by mud-raking.

  101. @89: Great hitting game by Pujols, but (all things being equal) if he was issued 5 intentional base on balls the Cardinals win the game 13-7.

  102. Scott-Well, first, that's probably not true, but yes, they probably would have happened. But so what?

  103. @102: The post is about win probability.

    "It was a great game,no doubt, but because the Cardinals already had a
    big lead..."

  104. @92: No, I didn't need this post to tell me that a homerun at 12-7 isn't clutch ordinarily, but, what this post did do is point out to me that I wasn't acknowledging (for whatever reason) this in this particular instance. And that to me had value, even if it pointed out something that regularly would be self-evident.

    No, my comment (while perhaps upon second-review errs more toward being caustic then my intended tongue-and-cheek) was a response to some of the earlier posts attacking the OP, where they took the title as a literal attempt to bash Pujols, instead of the tongue-and-cheek 'look, it must not be good because it was only xxx WPA, despite the fact that his performance was other-worldly'. That they missed the point that the poster agreed with them and acknowledged his performance was great by his very choice of the post title they were criticizing!

  105. @82- Andy, who wrote the article, makes that very statement in post 11- "...of course Pujols' performance was amazing, as I said, but from a purely stats-perspective, it WOULD have been better if he had struck out earlier in the game."

  106. Uncle Jimmy Says:

    I guess my interpretation is more generic, but I have always thought that the moment tends to define the achievement. It magnifies the accomplishment and also the act of the mistake. I always felt bad for Buckner because of the way he had to "pay" for the sins of the Boston bullpen in 1986, but we are always defined by our actions in a defining moment. In that perspective, Albert Pujols's game was outstanding, but there was no "defining moment" when any of the home runs was hit. I thought the shot off the second deck facing was massive. I guess I can only trust my eyes and what I saw to define what I thought was an individual game performance which will be talked about forever. It is very rare when the performance of one individual decides the outcome of a game (excepting a shut down pitcher, such as Bob Gibson). So, I don't know stats as well as you do, but I know beauty when I see it and I know when a performance becomes unique.

  107. @101

    You're assuming too much. If Pujols doesn't hit that first home run, the game is still very close, and may take a very different turn. Maybe everything else happens equally, maybe not.


    Exactly! To put this hypothetical into baseball terms, suppose a player goes 5 for 5, with 5 solo home runs, and his team wins 7-1. Each individual home run won't produce a significant add to WPA - the first ones because they are so early that they're not changing the complexion of the game (the other team still has significant chance to come back), and the later ones add little because their teams chance of winning is so large. He's being penalized on that 5th home run because of the first four. To say that it's "not very good" completely mischaracterizes the achievement.

    Then suppose the next night he goes 1 for 5, with four K's and a solo home run in the bottom of the 9th, to win the game 1-0. The second game will have an enormous WPA, but it's hard to classify that as "better." More clutch? Maybe. But saying that it's "better" that he didn't hit home runs instead of striking out in those at-bats is kind of daffy, don't you think?

    So let's stop attaching value words like good, bad, better, worse, etc to WPA. That's not what it measures.

  108. @107: I agree 100%. Things would have little chance of happening equally.

    Still, a visiting team with an 8-6 lead after 5 innings will win the game about 77%
    of the time. Also, you can't run out the clock in baseball.

    Win probability is my personal favorite advanced stat. It's a telling stat in
    my opinion.

  109. @108 : Didn't word that answer very well.

    You can't run out the clock in baseball making the first home run an important one.

  110. In all the arguments about WPA, it dawns upon me that nobody has noticed a stat Pujols accomplished that could be unique. When we think of hitting for the cycle, it's rarely done, and the usual comment afterwards is: "missed gettting the cycle by a homer", or however else it came up short. Well, I have always wondered if anyone had ever hit a HOMER cycle: a solo, 2-run, 3-run and slam in the same game. From what everyone has said, Pujols certainly came closer to that than anyone else post-season; he missed a homer cycle only by a slam. Not to trivialize a grand slam, of course, but even getting 3/4 of a homer slam is nothing to be sneezed at!

  111. Sorry, though I had read my post before submitting, this little glitch still snuck in; should be: "even getting 3/4 of a homer CYCLE is nothing to be sneezed at!"

  112. Game 3 was historic even before Albert. From the top of the 4th through the top of the 6th, each team scored 3+ runs for a record total of 5 frames. In fact, only 4 times previously, in 1922, 1959, 1978 & 1997, had 2 teams each scored 3+ runs in the same inning in the WS. In Game 3, it happened in consecutive (4th & 5th) innings.

  113. To all who are using the logic that Pujols' performance was the best ever, based on his 3 HR and 14 total bases, something no one else has topped in postseason history:

    Would you also say Shawn Green's 4 HR / 19 total bases performance was the best hitting performance ever? You'd have to, right?

    But that's ridiculous, as anyone watching that game would attest; the fact is basing "the greatest" on number of HR and total bases is just as ridiculous as basing mediocrity on a low WPA (which I'm quite positive Andy was NOT doing in the first place).

  114. Why wouldn't you think of Green's game as the best hitting performance ever? Why would that be ridiculous?

    You are conflating two terms here, "greatest" and "best." They are slightly distinct. "Best" doesn't depend on story line etc. "Greatest" does. For instance, right now Jeter is NOT the "best" shortstop in the game today, but most would agree he is the "greatest." People often mix up these two words.

    Green's performance might very well be the best. You would certainly make a case for it, though Whitten's record RE24 game would be in the conversation.

    Basing the BEST game on hard numbers like the most number of bases etc. IS pretty logical. It is certainly not "ridiculous." What would be ridiculous is to dismiss doing so out of hand. What's your defense for that position other than what it would mean to Shawn Green? You really don't defend your argument, you just sort of scoff Green's game and then pretend you scored a point.

    As far as Andy goes, he should have worded his article differently. If just a few people were confused, it would be on them. If you confuse virtually everyone, it's on you. Communication is measured by how well you communicated your thought, and if you didn't do that then its on you.

  115. My point (hastily and poorly made, to be sure) re: Green was that I've never heard anyone talk about it as the pinnacle of achievement, though in terms of HR and runs scored, it has never been surpassed (Guy Hecker's 8/15/1886 effort notwithstanding). Maybe it's because it was a May game and every pitcher Milwaukee trotted out was a tomato can throwing BP to an otherwise yawn-worthy Dodgers offense. You'd argue those are strikes against the performance's position as greatest, not best; I would argue that very few people make such pedantic distinctions.

    But whatever, I didn't mean to distract and I'm not trying to break into Green's house to steal his sense of accomplishment, any more than I think Andy was in the first place. I just re-read the original article and, aside from the tongue-in-cheek headline (which everyone is taking way, way too seriously) I can see no conceivable problem whatsoever with anything Andy wrote. Factual, concise, and giving the full picture of WPA (i.e., not denigrating Albert's individual performance based on the stat, but rather recognizing his position among others according to that particular stat).

    I see we're playing the "rewrite the headline" game as well. "Albert Pujols' not very good (as measured by WPA against similar individual achievements as defined by key accumulation statistics) 3-HR game" Would that have been a better headline? Is there just no room for hyperbole in writing on this blog?