Baseball

MVP and HOW

So, how about a little thought experiment? The Miguel Cabrera-Mike Trout MVP balloting exposed a pretty deep divide in how people want to look at awards. We can probably all agree on at least that. There will be plenty more over the next few years to say about that divide, what valuable means, and so on. But for now I have a different question:





Which kind of award do you prefer?





That is to say: For you, as a fan, do you prefer an award that goes to the player who has the best narrative -- which is what the MVP award usually represents? Or would you rather have an award that goes to the best player statistically, period, regardless of position or style of play or team performance or your favorite intangible?





I went back to 1995 -- each year I picked the league MVP (Most Valuable Player, of course) and I picked the HOW (the league's Hero of WAR -- that is, the player with the highest Wins Over Replacement based on Baseball Reference WAR, Fangraphs WAR,  Baseball Prospectus WARP and Win Shares).





The question for you then is: Which choice do you as a baseball fan like better?





2012





American League -- MVP: Miguel Cabrera; HOW: Mike Trout (2nd in MVP voting)





National League -- MVP: Buster Posey; HOW: Buster Posey





We're not going to go back and argue Trout and Cabrera -- everyone has staked their position -- but here have been two things people have talked about that are worth mentioning.





1. If Cabrera had not won the Triple Crown would he still have won the MVP?





Personally, I think yes. I think Cabrera won the MVP the last two weeks of the season because several trends came together all at once. One, Trout, while playing well, wasn't as dazzling as he had been in June and July (and his Angels could not quite get back into the playoff picture). Two, the White Sox went into collapse mode allowing the Tigers -- who only played 28-24 baseball the last two or so months of the season -- to take the American League Central rather comfortably. Three, Cabrera challenged for the Triple Crown.





When Cabrera actually won that Triple Crown, I think he clinched a blowout in the voting. But if he had lost by a home run or a batting point, I still think he would have won the MVP. I just think the voting would have been a handful of first-place votes closer.





2. A question from the brilliant Tom Tango: If Cabrera's season had not existed, would Trout have STILL lost the award, only this time to Robbie Cano? That is to say: Would the voters have felt it necessary to vote for a player whose team made the playoffs?





My answer to him: I don't think so. I think with no Cabrera, Trout would have won. I think Cabrera challenging for the Triple Crown -- and eventually winning it -- was the key element in the MVP vote. There are those who would argue ( and WAR suggests this) that Cano actually had a better season than Cabrera when you take defensive stats and base running and other WAR calculations into account. I'll let other people argue about that. But I think the voters would have gone with Trout over Cano -- Trout's season was too good. I will say, though, that I think in a world without Cabrera, the Trout-Cano vote would have been closer than the Cabrera-Trout vote turned out to be.





* * *





2011





American League -- MVP: Justin Verlander; HOW: Jacoby Ellsbury (2nd in voting)





National League -- MVP: Ryan Braun; HOW: Matt Kemp (2nd in voting)





I did not have a particularly elaborate method for picking the HOW. You might add the numbers together and come up with a different HOW than I did. I basically just added together Fangraphs WAR and Baseball Reference WAR, and then used the other methods as tiebreakers, if necessary. In this case -- and this will repeat several times -- the players involved are separated by only a few tenths of a point of WAR. And I don't think even the most strident believer thinks that WAR, as calculated, is so precise that a few tenths of a point mean very much. Ellsbury's edge over Verlander was almost nonexistent (though Verlander's MVP victory over Ellsbury was fairly crushing, 13 first-place votes to 4), and it was the same with Kemp over Braun.





I think this is how it works most years -- WAR is detailed enough to offer a set of good MVP candidates, but not so meticulous that it can really tell you that someone a 7.4 WAR actually had a better year than someone with a 7.2 WAR.





* * *





2010





American League -- MVP: Josh Hamilton; HOW: Josh Hamilton





National League -- MVP: Joey Votto; HOW: Albert Pujols (2nd in voting)





Again, just a few tenths of a point of WAR separate Pujols and Votto. The MVP voting (Votto 31 first-place votes; Pujols 1) was not quite so close.





* * *





2009





American League -- MVP: Joe Mauer; HOW: Zack Greinke (17th in voting)





National League -- MVP: Albert Pujols; HOW: Albert Pujols





Well, here is a good year to ask the question: Which kind of award do you like better? The choice is fairly clear. Mauer had an amazing year for a division winner. He played catcher -- the thinking man's position -- and he won the batting title and he hit with previously unseen power. He was a dynamo.





Greinke, by WAR, was clearly the most valuable player in the American League. He led the league with a 2.16 ERA in 229 innings, struck out 242, threw six complete games and three shutouts. And, of course, his Royals were terrible anyway: They lost 97 games.





So, who do you like? How would you have felt if Greinke had actually won the MVP? Happy? Let down? Infuriated? What does your MVP look like? Is it narrative-based? Stat-based? A combination of the two?





This question goes deeper than Greinke, though. Maybe you believe that WAR, by its nature, overvalues pitchers. That's certainly a viable point of view. You will see, through the years, that the HOW of the league is often a starting pitcher, while the MVP is almost never a starting pitcher.





Thing is, even if you exclude pitchers, Mauer STILL would not have been the HOW of the American League in 2009. It would have been Tampa Bay's Ben Zobrist. This has been a point of derision among some anti-WAR types, who mock any award that would put Ben Zobrist ahead of a catcher who hit .365 and slugged .587.





But, again, let's look a bit more closely. Zobrist played seven different positions in 2009 -- and by general agreement, he appeared to play them all well. He posted a .400 on-base percentage and a .543 slugging percentage, hit 27 homers, walked 91 times, stole 17 bases, scored 91 runs and drove in 91. It was, added together, an extraordinary season. Was it actually better than Mauer's? Was it more worthy of the big award? That's for you to answer -- that's what this whole exercise is about.





Anyway, Zobrist's season was unquestionably better than Kendrys Morales' or Jason Bay's. But those two finished ahead of Zobrist in the MVP voting.





* * *





2008





American League -- MVP: Dustin Pedroia; HOW: Cliff Lee (12th in voting)





National League -- MVP: Albert Pujols; HOW: Albert Pujols





WAR, as mentioned, often goes with great starting pitchers as the MVP. In this case, however, I almost regret even bringing it up because Lee and Pedroia were in a virtual tie. Pedroia led every-day players in WAR.





* * *





2007





American League -- MVP: Alex Rodriguez; HOW: Alex Rodriguez





National League -- MVP: Jimmy Rollins; HOW: Tie between David Wright and Albert Pujols





Rollins didn't finish Top 5 in WAR -- this was a big divide between MVP and HOW (Wright and Pujols finished about two and a half wins better than Rollins).





Rollins, you might remember, had an unusual year. He became the first National Leaguer to get 200 hits with at least 20 doubles, 20 triples and 20 homers. He led the league in runs scored and won a Gold Glove. And the Phillies came back from oblivion to reach the postseason for the first time in 14 years. As far as narratives go, that's awfully good. The other narratives couldn't come close to matching up: Wright's Mets collapsed, and Pujols had already won an MVP.





* * *





2006





American League -- MVP: Justin Morneau; HOW: Grady Sizemore (11th)





National League -- MVP: Ryan Howard; HOW: Albert Pujols (2nd)





Every now and again, something funny happens in the MVP voting: There just isn't an especially interesting candidate. That happened in 2006 in the American League. Yes, Derek Jeter did hit .343, he had a great year and that was a golden opportunity to finally give him the MVP. But, the thing is, it kind of was like EVERY Derek Jeter year. I think his amazing consistency hurt him. Eddie Murray was another guy who never won an MVP, in large part because while every year was great, every year looked the same.





In 2006, in addition to Jeter, a bunch of guys -- Sizemore, Joe Mauer, Vernon Wells, Travis Hafner, Johan Santana, David Ortiz -- offered very similar value by the WAR stats. Nobody really jumped out, and I would argue that when that happens we will often get a weak MVP.





Morneau was a weak MVP. He finished about 20th in the league in WAR, his .375 on-base percentage and .559 slugging percentage were not in the Top 5, his 34 homers were not in the Top 10, he couldn't run and his first-base defense was subpar. He literally won the award because he led the league in RBIs and because he is a good guy. It's probably the worst choice of the decade.





Ryan Howard had an amazing offensive season -- but when you consider that Pujols hit for a higher average, had better on-base and slugging percentages, was a much better base runner and defender AND killed it the last month of the season (which seemed so interesting to voters this year), it's kind of hard to square this choice with reality. But Pujols had won the MVP the year before. And Howard hit a lot of home runs (58) and drove in a lot of runs (157), which traditionally is where MVP voters start their search.





And, by the way, one quick connection between Howard-Pujols and Miggy-Trout -- Pujols' team made the playoffs (and went on to win the World Series) while Howard's did not reach the postseason. But -- and I knew I had heard this somewhere before -- Howard's Phillies actually had a better record than Pujols' Cardinals. So, and I remember hearing this, people said: "Hey, Howard's team had the better record. It's not his fault that he was in a tough division."





In other words, that argument is perfectly fine for BBWAA voters to use, but only if the voters choose to use it.





* * *





2005





American League -- MVP: Alex Rodriguez; HOW: Alex Rodriguez





National League -- MVP: Albert Pujols; HOW: Albert Pujols





Voters and stats unite! Andruw Jones actually led Pujols by a tenth of a point in Fangraphs WAR. He finished second in the voting. More on him in a minute.





* * *





2004





American League -- MVP: Vlad Guerrero; HOW: Ichiro Suzuki (7th)





National League -- MVP: Barry Bonds; HOW: Barry Bonds





Here's something that seems to happen a lot -- the right guy wins the award, but he wins it in the wrong year. Ichiro won the MVP in 2001; by WAR, he was an appreciably better player in 2004. Vlad won the award this year, but he was an appreciably better candidate in 1998 and 2002.





But, the MVP generally comes down to three things: (1) success of team; (2) homers and RBIs -- with batting average thrown in as a tiebreaker; (3) late-season heroics or general leadership points.





Ichiro hit .372 this year, set the major league record with 262 hits, stole 36 bases and played masterful defense. He was a force of nature. But the Mariners stunk to high heaven -- lost 99 games, finished dead last in runs scored, tried to prolong the careers of Bret Boone and Edgar Martinez and so on. Ichiro finished seventh in the voting. The Angels, meanwhile, were good after Guerrero signed a free-agent deal. Vlad was one of my favorite players. This was probably his fourth or fifth best season.





* * *





2003





American League -- MVP: Alex Rodriguez; HOW: Alex Rodriguez





National League -- MVP: Barry Bonds; HOW: Barry Bonds





Both too good to ignore. This was A-Rod's first MVP award. Based on WAR, it should have been his third at minimum.





* * *





2002





American League -- MVP: Miguel Tejada; HOW: Alex Rodriguez (2nd)





National League -- MVP: Barry Bonds; HOW: Barry Bonds





A-Rod hit 57 home runs and won the Gold Glove at shortstop; it's hard to imagine that guy on the most basic level is not the MVP, especially when the voters chose a different shortstop in the same division who was not nearly as good.





Bonds in all four of these years -- 2001 to 2004 -- was the no-brainer choice, and that showed up in the voting. There were 128 first-place votes available those four years. Bonds got 114 of them.





* * *





2001





American League -- MVP: Ichiro Suzuki; HOW: Jason Giambi (2nd)





National League -- MVP: Barry Bonds; HOW: Barry Bonds





The first-place votes in the American League were divided among four players, Ichiro (11), Giambi (8), Bret Boone (7) and Robbie Alomar (2). It really was close by WAR too, though Ichiro was a clear fourth behind Giambi (.477 on-base percentage and .677 slugging percentage in Oakland); Boone (.331 average, 37 homers, 141 RBIs, 118 runs scored ) and A-Rod (led league with 52 homers and 133 runs).





* * *





2000





American League -- MVP: Jason Giambi; HOW: Pedro Martinez (5th)





National League -- MVP: Jeff Kent; HOW: Randy Johnson by an eyelash over Todd Helton (Unit was 17th in voting)





Another situation of "right guy, wrong year." Giambi was better in 2001 than 2000 by WAR. Meanwhile, Pedro was otherworldly. A-Rod was, again, the best position player in the league based on WAR.





Two things I found fascinating looking at all this. One is that Johnson was the best player in the league two or three times. He legitimately could have won three MVP awards (though he never came close to winning any). I have written that I think Greg Maddux has a chance -- a chance, mind you -- of setting the record for highest Hall of Fame voting percentage. Well, what I have written is that Maddux could become the first unanimous Hall of Fame choice, but I'm backing off that because someone will undoubtedly send in a blank ballot or a ballot covered in mud or something like that to protest, you know, whatever they're protesting. Anyway, I think Maddux could be a 99 percenter. And I think Unit is right there too. I have no idea who would vote against him.





Second: I'd say that three or four times in his career, Andruw Jones was a strong MVP candidate. He finished second in the voting in 2005. He was even better by WAR in 1998, 1999 and 2000. Obviously, it depends on how much value you give his defense, but if you see Jones as a breathtaking defensive center fielder in the realm of Willie Mays (as some people do) and you throw in his 434 homers, he's at least an interesting Hall of Fame candidate. Winning an MVP award might have helped him.





* * *





1999





American League -- MVP: Ivan Rodriguez; HOW: Pedro Martinez (2nd in voting)





National League -- MVP: Chipper Jones; HOW: Randy Johnson (15th in voting)





A famously close vote in the AL; Martinez got one more first-place vote than I-Rod but lost out when he did not appear at all on two ballots. Well, that's the MVP voting -- narrative rules. In 2011 the narrative shifted enough for Justin Verlander to not only contend for MVP but to win it outright. In 1999, though, there was some real doubt about whether a pitcher could actually be the best player in the league. Rafael Palmeiro, a glorified DH* playing half his games in an absurd hitters' park, got four first-place MVP votes. It was like that.





*Though he famously won the Gold Glove at first base, despite playing only 28 games there.



Randy Johnson threw 271 innings, 12 complete games, struck out 364 batters and posted a league-leading 2.48 ERA. Chipper was fabulous, the best every-day player in the league, I believe. This time around, though, I don't think there's a particularly good argument for anyone except Unit as the best player in the league.





* * *





1998





American League -- MVP: Juan Gonzalez; HOW: Alex Rodriguez (9th in voting)





National League -- MVP: Sammy Sosa; HOW: Kevin Brown (16th in voting)





Two thoughts: One, you would have to imagine that members of the BBWAA -- I know many of these people, and know how seriously they take their votes -- absolutely CRINGE when they look back at the two MVP awards they gave Juan Gonzalez. Baseball writers, myself included, have mea-culpa'd everyone to death about not covering the steroid scandal better when it was going on. It might be time for official apologies on the Juan Gone MVPs too.





Second: Could you even IMAGINE what would have happened in 1998, after that amazing home run summer, if the voters had chosen Kevin Brown as the league MVP? People might have taken to the streets. Brown was incredibly good that year, though. Take a look at these two years:





Player 1: 251 innings, 174 hits, 24 homers, 250 strikeouts, 57 walks, 4 complete games, 2 shutouts, 2.40 ERA





Player 2: 257 innings, 225 hits, 8 homers, 257 strikeouts, 49 walks, 7 complete games, 3 shutouts, 2.38 ERA





You probably figured it out. Player 2 is Kevin Brown in 1998. Player 1 is Justin Verlander in 2011.





And, in retrospect, knowing what we now know about PEDs: Kevin Brown allowing just 8 home runs in almost 260 innings is absolutely mind-blowing.





* * *





1997





American League -- MVP: Ken Griffey; HOW: Roger Clemens





National League -- MVP: Larry Walker; HOW: Larry Walker and Craig Biggio in virtual tie





Once more, based on WAR, the MVP went to the right guy in the wrong year. Griffey was almost certainly the best every-day player in the American League in 1997. But this was an otherworldly season for Clemens. He apparently pitched angry all year because of the way the Red Sox treated him. Let's also compare him to Verlander's 2011:





Verlander: 251 innings, 174 hits, 24 homers, 250 strikeouts, 57 walks, 4 complete games, 2 shutouts, 2.40 ERA





Clemens: 264 innings, 204 hits, 9 homers, 292 strikeouts, 68 walks, 9 complete games, 3 shutouts, 2.05 ERA





Bill James, in his New Historical Abstract, went through just how remarkable Craig Biggio's 1997 season was. He hit .309, walked 84 times and was hit by a league-leading 34 pitches. He had 67 extra-base hits, scored a league-leading 147 runs, stole 47 bases and, incredibly, did not hit into a double play all year. He could not come close to matching Walker's absurdly gaudy numbers (.366, 49 homers, 46 doubles, 130 RBIs, .720 slugging) but those numbers were unquestionably inflated by the altitude at Coors Field. If you contextualize everything, those two players were REALLY close in value, I believe. But Biggio did not get a single first-place MVP vote.





* * *





1996





American League -- MVP: Juan Gonzalez; HOW: Ken Griffey (4th)





National League -- MVP: Ken Caminiti; HOW: Barry Bonds (5th)





This was the worse of the two Juan Gone MVP votes.





Here's a thought: I wonder if every year we here at the blog should give out what I would call the Tommy Harper award -- that is the award to the player (usually relatively unnoticed) who does the most things well. I would name it after Harper because the guy was awesome and, more to the point, in 1970, in Milwaukee, Harper played five different positions, hit .296, walked 77 times, hit 31 homers, stole 38 bases, scored 104 runs and accumulated 315 bases.





Obviously, Biggio was the winner of the Tommy Harper Award in 1997. Chuck Knoblauch would have won it in 1996. He hit .341, walked 98 times, was hit by 19 pitches, led the league with 14 triples, added 13 homers, stole 45 bases, scored 140 runs, accumulated 299 total bases and seemed to play an excellent second base (before the throwing thing happened).





* * *





1995





American League -- MVP: Mo Vaughn; HOW: Randy Johnson (6th)





National League -- MVP: Barry Larkin; HOW: Greg Maddux (3rd)





From 1987 through 1996, 10 seasons, the BBWAA voted Mo Vaughn (4.1 WAR); Juan Gonzalez (3.5 WAR); Andre Dawson (3.7 WAR) and Dennis Eckersley (2.8 WAR) as MVPs. I would have to believe that, even as people yelp about the inequities of WAR, the voters will never again choose anyone quite that low in statistical value again.





Perhaps the most interesting part of the 1995 voting in the American League is actually a trivia question: Between 1990 and 2012, which Red Sox every-day player recorded the highest WAR in a single season?





I suspect Red Sox fans know this right away. And I suspect that, because I asked the question, you know that it isn't David Ortiz or Kevin Youkilis or Dustin Pedroia or Nomar Garciaparra or anyone obvious.





It's John Valentin in 1995. He posted an 8.1 WAR, which is actually higher than Jim Rice's legendary 1978 season (higher, in fact, than any Jim Rice season). It was the highest in the American League in 1995 among every-day players.





Now, let's just say this: Valentin was crazy good in 1995. He hit .298, walked 81 times, hit 27 homers, stole 20 bases, scored 108 runs, drove in 102 runs and, by the numbers, played exceptional defense at shortstop.





But was he really the best every-day player in the league? Better than Albert Belle or Edgar Martinez? Was his year really better than Rice's in '78? Or is this just a statistical blip based on him being rated way too high defensively or the numbers not being properly adjusted for Fenway Park (he hit 40 points higher and slugged 40 points better at home)? Is this merely a sign that WAR is just hopelessly flawed -- "an overrated bauble" in the words of the great Tom Verducci?





Or do these stats -- because they do not even attempt to measure leadership and aura, because they do not care if the player is named Valentin or Clemente, Zobrist or DiMaggio -- do a better job of calculating who contributed most on the field?





* * *





Bonus breakdown: Cy Young 1993





There's a 

cool interview on Fangraphs with David Cone

, and he brings up the 1993 Cy Young Award. His point is that, take away pitcher wins -- as we try to do here -- and it would be very hard to separate Jack McDowell's Cy Young Award season from David Cone's season (though Cone did not receive even a third-place Cy Young vote).





McDowell: 256 2/3 innings, 261 hits, 20 homers, 158 strikeouts, 69 walks, 10 complete games, 4 shutouts, 3.37 ERA, 1.286 WHIP





Cone: 254 innings, 204 hits, 20 homers, 191 strikeouts, 114 walks (yikes!), 6 complete games, 1 shutout, 3.33 ERA, 1.256 WHIP





Wow, Coney, that's a lot of walks, brother. I had no idea. But, other than that, Cone is absolutely right. The numbers are eerily similar -- almost same number of innings, almost same ERA, same homers, almost same WHIP, McDowell gave up more base runners by hit, Cone by walk. Look at it this way:





Runs allowed by start





Cone





0 runs -- 4 times



1 run -- 1



2 runs -- 9



3 runs -- 9



4 runs -- 4



5-plus -- 7 times





McDowell





0 runs -- 4 times



1 run -- 3



2 runs -- 8



3 runs -- 4



4 runs -- 8



5-plus -- 7 times





Pretty close to identical, no? McDowell had a couple more one-run starts, Cone had an advantage in two- and three-run starts. Pretty close to identical. And yet, McDowell got 21 first-place Cy Young votes and Cone got none, and I imagine that nobody even thought of giving Cone a first-place vote. Why? You know why. Wins. McDowell went 22-10. Cone went 11-14. That was the whole story.





But, amazingly enough, it wasn't Cone but ANOTHER Royals pitcher who absolutely, positively, no questions asked should have won the Cy Young that year. Look at one more comparison:





McDowell: 256 2/3 innings, 261 hits, 20 homers, 158 strikeouts, 69 walks, 10 complete games, 4 shutouts, 3.37 ERA, 1.286 WHIP





Royals pitcher: 238 2/3, 183 hits, 8 homers, 186 strikeouts, 81 walks, 5 complete games, 1 shutouts, 2.56 ERA, 1.106 WHIP





That's a whole lot more run prevention going on. It's even more striking when you break it down by start:





McDowell





0 runs -- 4 times



1 run -- 3



2 runs -- 8



3 runs -- 4



4 runs -- 8



5-plus -- 7 times





Royals starter





0 runs -- 5 times



1 run -- 5



2 runs -- 11



3 runs -- 8



4 runs -- 1



5-plus -- 3 times





Not close, is it? The Royals starter, of course, was Kevin Appier and his 18-8 won-loss record just didn't impress the voters quite like McDowell's 22-win season did. It's a shame. Appier was a great pitcher from 1990 through 1997, and almost nobody noticed.

  Comments