The American League MVP will be announced on Tuesday, and like its NL counterpart (won yesterday by the Cardinals‘ Albert Pujols), there’s no strong favorite to win, a situation likely to incite controversy. As the BBWAA ballot explicitly acknowledges, “There is no clear-cut definition of what Most Valuable means. It is up to the individual voter to decide who was the Most Valuable Player in each league to his team. The MVP need not come from a division winner or other playoff qualifier.” According to the instructions, offense, defense, playing time, and character are all fair game.
Value is a subjective term to voters, who tend to emphasize offensive production as denoted by the traditional “Triple Crown” categories (batting average, home runs, and RBI). Fielding ability, if addressed at all, enters the argument via vague positional adjustment-a shortstop need not lead the league in a Triple Crown category, for example. Voters tend to screen for players whose teams were in contention; since 1969, just three MVPs have come from sub-.500 teams, and in the AL, only one out of the last 16 (Alex Rodriguez, 2003) has come from a non-playoff team. Narratives become attached to the statistics, with exceptionally strong clutch performances-with runners in scoring position or late in the season-informing the voting.
Statheads view player valuation more objectively. There’s no shortage of metrics to quantify a player’s offensive, defensive, and/or pitching contributions in runs or wins relative to the league average or a theoretical replacement level, and the context of ballpark and league scoring environments can be incorporated. BP’s preferred route is via Wins Above Replacement Player (WARP1), which enables measuring every player-pitchers as well as hitters-on the same scale.
WARP illustrates the disconnect between objective and subjective notions of MVPs. Since 2000, NL voters have made “correct” calls in six out of nine years (Jeff Kent in 2000, Barry Bonds in 2001-2004, and Pujols this year). AL voters, on the other hand, have matched the winner to the WARP leader just twice in that span via Rodriguez’s 2005 and 2007 wins. Justin Morneau, the 2006 winner, ranked 20th in the league (fourth on his team!) with 7.8 WARP, both lows for the era.
Here’s a thumbnail guide to the AL’s top 10 finishers in WARP and their MVP cases, along with those of two popular lower-ranked candidates:
Cliff Lee, Indians (10.4 WARP)
Working For Him: Lee went 22-3 with a 2.54 ERA and ran away with the Cy Young award, and he has the narrow WARP lead here.
Working Against Him: No pitcher has won the AL MVP since Dennis Eckersley in 1992, and no starting pitcher has done so since Roger Clemens in 1986.
Mariano Rivera, Yankees (10.3)
Working For Him: The 38-year-old future Hall of Famer put together one of his best years, with a microscopic 1.40 ERA and an astounding 77/6 strikeout-to-walk ratio.
Working Against Him: Although three of the last four pitchers to win the MVP have been relievers (Eckersley, Willie Hernandez in 1984, and Rollie Fingers in 1981), their teams all made the playoffs, while the Yankees did not. Also, Rivera’s 39 saves look pedestrian next to Francisco Rodriguez’s record-setting 62.
Roy Halladay, Blue Jays (9.8)
Working For Him: In going 20-11 with a 2.78 ERA, Halladay may have been even better than Lee given the quality of competition.
Working Against Him: Lee’s raw numbers, both traditional and sabermetric, outdo Halladay’s, and the Cy Young voting suggests that argument is settled.
Dustin Pedroia, Red Sox (9.8)
Working For Him: In a year that saw the mid-season trade of Manny Ramirez and subpar performances from David Ortiz and Jason Varitek, Pedroia was the glue that held the AL Wild Card-winning Red Sox’s offense together. He hit .326/.376/.493 with 17 homers, led the league with 54 doubles and 213 hits, and played excellent defense.
Working Against Him: Teammate Kevin Youkilis could siphon votes away thanks to a higher OBP (.390) and SLG (.569).
Joe Mauer, Twins (9.6)
Working For Him: Mauer won his second batting title in three years (.328 AVG) and finished second in OBP (.413) while playing above-average defense (+10 FRAA, 36 percent caught stealing) and shepherding a relatively inexperienced and unheralded pitching staff all the way to a Game 163 play-in.
Working Against Him: Mauer hit just nine homers and drove in 85 runs, numbers likely to be held against him despite his virtues.
Alex Rodriguez, Yankees (8.9)
Working For Him: A-Rod led the league in SLG (.573) while finishing third in homers (35) and fifth in OBP (.392) and runs scored (104).
Working Against Him: The Yankees’ failure to make the postseason aside, it was a down year for Rodriguez relative to his three previous MVP seasons, especially his 54-homer 2007. He missed 17 games due to a quad strain, and a .271 average with runners in scoring position stoked the notion that he was unclutch.
Mike Mussina, Yankees (8.6)
Working For Him: Mussina bounced back from four rather ordinary seasons to win 20 games for the first time while finishing sixth with a 3.37 ERA.
Working Against Him: Two Yankees had higher WARPs in a year in which they missed the postseason for the first time since 1993.
Justin Morneau, Twins (8.6)
Working For Him: Morneau hit .300/.374/.499 and finished second in the league in RBI with 129, one short of the total that propelled him to the 2006 award in spite of his weaknesses.
Working Against Him: Apart from his big counting stats, he wasn’t as good as Mauer offensively or defensively.
John Danks, White Sox (8.5)
Working For Him: Danks enjoyed a breakout year, tossing 195 innings with a 3.32 ERA (fifth in the league) and pitching the White Sox into the postseason with eight innings of shutout ball in the play-in.
Working Against Him: Danks was credited with just 12 wins, and the popular perception is that teammate Carlos Quentin (7.7 WARP) was more valuable before getting hurt.
Kevin Youkilis, Red Sox (8.4)
Working For Him: Youkilis helped pick up the slack in Boston’s remade offense, leading the Sox with 29 home runs while hitting .312/.390/.569 and filling in for injured Mike Lowell at third base.
Working Against Him: While Youkilis had better rate stats than Pedroia, the latter had 105 more plate appearances and played the tougher defensive position.
Carlos Quentin, White Sox (7.7)
Working For Him: Quentin was leading the league in homers (36) and the White Sox in OBP (.394) and SLG (.571) when he broke his wrist on September 1.
Working Against Him: Missing a month is generally fatal to a player’s MVP case, and he wound up ranked 27th in WARP. The Sox were tied with the Twins when he went down, and edged ahead via the play-in, so there’s no “if only” counterargument to his absence.
Francisco Rodriguez, Angels (7.6)
Working For Him: With 62 saves, K-Rod demolished Bobby Thigpen‘s 1990 single-season record of 57. He did so while pitching for a team that finished the year with 100 wins, the AL’s best record, and the largest Pythagorean overachievement of all time.
Working Against Him: None of his saves came in outings longer than one inning, and the Reliever Expected Wins Added (WXRL) metric, which is more sensitive to inning and score context than WARP (which ranked him 28th), shows him about half a win behind Rivera.
As to how this will all pan out, put your nickel on a Pedroia victory. Starting pitchers have a long history of being overlooked in the voting, and Pedroia, as the top-ranked hitter on a playoff team, thus has the edge as well as a nice little “stepping up” narrative that should satisfy the storytelling set. On the other hand, given the travesty of the 2006 vote and the absence of a slam-dunk candidate, don’t be surprised if K-Rod’s record carries the day.
Thank you for reading
This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.
Subscribe now
The fact that quite often the \"best hitter\" (with respect to positional and context adjustments) has a high WARP score is directly related to the fact that almost the entirety of a position player\'s value stems from his offensive contributions. You can\'t just ignore the fact that ARod\'s value in 2003 was higher than any other SS. We can measure that his performance contributed N Wins more to his team\'s win total than the next best guy. That\'s the name of the game, getting the most wins. So if ARod is worth a full win more than Quentin, then ARod is more valuable to my team. End of story. That the best players are generally on the best teams should be obvious simply because THAT\'S WHY THEY ARE THE BEST TEAMS. But that doesn\'t mean they would be less valuable if you surrounded them with dross (or a terrible pitching staff in Texas). Punishing a player because his teammates are not as good as another player is dumb.
Frankly, if you want to get into the marginal value of a Win to a team, you\'ll probably end up punishing the front running teams _more_ than the \"good players on bad teams\" contingent as you get into diminishing returns for the value of a Win once a team has already assured themselves of a playoff spot.
The distinctions between leagues aren\'t nearly so tidy. Both leagues have been vulnerable to the Most Valuable Narrative in recent years - Jimmy Rollins winning in 2007 despite ranking eighth in the league and second on the team to Utley (9.3 to 8.8), Morneau in 2006, Miguel Tejada in 2002 (10th, and nearly a win behind teammate Barry Zito but a hair ahead of Tim Hudson - 9.8, 8.9, 8.7 - in the year after Giambi left those poor little A\'s without their big, big slugger). Etc.
If you\'re serious that you want the award to be for \"greatest actual contribution toward winning, even though that depends on how well his teammates played\", then you should look at the measures like WXRL that calculate exactly how much that individual player\'s contributions increased the team\'s expected win total, given the situations the player hit and pitched in. That will reward players for getting lots of PAs with men on base, pitching well in close games, etc. Baseball Prospectus doesn\'t calculate any such \"value added\" stat for offense, but other people do. But beware: it might still tell you that the biggest contribution to winning came from a guy on a last-place team.
Once you go beyond that, to say that contributions only matter if the team succeeded in the end, you\'ve stopped handing out an individual award. We already have an award for teams; it\'s called \"making the playoffs\".
I don\'t have the info in front of me, but has a player ever won the MVP with a team mate in the top 3 or top 5?
No worries. I sure like the stats you have and understand them when used in articles. It just appears that I need to have some sort of knowledge of Excel or something in order to make it work.
Don\'t feel bad, though.
Question: Rivera\'s WARP1 is 10.3, but his WXRL is only 6.2. Is WXRL relative to league average, rather than to some estimated replacement level? Is a 6.2 WXRL worth more than 10 wins above replacement level? Given Rivera\'s high leverage score, I\'d expect his context-dependent stats to be better than context-neutral stats like WARP.
1. It\'s a misnomer to say that WARP informs MVP decisions because I doubt there\'s a significant fraction of the electorate who even thinks to use it as a guide when deciding their vote. To the non-voting population who cares what went into the voting, WARP does provide some illumination regarding the decisions, and it might inform our Internet Baseball Awards votes.
2. In general, I do think choosing from among the top handful of WARP leaders is defensible because of the varying contexts of those players, not only along the lines you suggest but also because of other evidence being incorporated into the argument. If I have two candidates whose FRAAs are contradicted by another system I respect like Plus/Minus scores, I should consider that, and (more for pitchers) I\'d consult SNLVAR and WXRL to get a better read on Win Expectancy added.
3. WXRL is play-by-play based, and yes, it\'s relative to replacement level, but it\'s the replacement level from the VORP universe, not the WARP one. In WARP, reliever WARPs are in part a function of the number of decisons (including saves) which are used as a proxy for leverage. If I were going to consider a reliever for my MVP vote, I\'d make sure his WXRL supported that decision even if the two figures don\'t come that close to matching.
Rich Campbell of the Fredericksburg (Va.) Free Lance Star is taking a lot of heat for leaving Ryan Howard off his ballot completely, the only voter to do so. Campbell\'s top 10 are all in your top 11 in VORP (depending how you want to count Manny Ramirez. And that\'s including Santana and Lincecum. Could it be there\'s another BBWAA writer paying attention? (And that he couldn\'t figure out how to get it to sort by WARP either?)
As for myself, I think that since the awards have no actual repercussions other than perhaps salary, I am content for people to weight stretch-drive drama against season-long value accumulation however they like.
Is it the same people voting for the two awards? If Cliff Lee was the best pitcher in the AL, isn\'t he the \"Most Valuable Player Who Happens To Be A Pitcher\"; and, therefore, a better MVP candidate than K-Rod?
For me I look at the CY Young award as who is the pitcher I\'d most fear the other team having, informed by the last year\'s pitching. While the MVP is who was most valuable last year. So for me the CY should be more about peak value and include a little more context past the current year and more prediction of future skill (in the next game), while the MVP should be a backwards looking value question (covering only
what happened).
So the two may not be perfectly aligned. So if one pitcher only started 28 games and had worse BABIP than another pitcher who pitched 35 games and had a worse perERA but better BABIP and slightly more value than the first would be my CY and the second would be my MVP. Still most of the time they are aligned for me.
Anyway, with the respect to the question of what logic would explain K-Rod finishing ahead of Cliff Lee in the MVP voting when Lee blew him away in the Cy Young race, remember that team performance is taken into account far more for the MVP than the Cy. K-Rod\'s team won 100 games, Lee\'s team was below .500.
Whether you agree that it should be factored in at all is one thing that can be debated (and is frequently =P), but the instructions given to the voters for the MVP is much very ambiguous on this point, whereas the Cy voters are *explicitly* instructed *not* to factor the team\'s overall record (though some of them have admitted they still can\'t help themselves and use it there too).
I\'m wondering if anyone could crunch these numbers for me:
When Mike Lowell went down, Youk slid over to third and played very good third base defense, allowing the Red Sox to play Sean Casey (and later Mark Kotsay) everyday. In most cases, if your third baseman goes down, you can\'t plug in a first base-type. If Youk had been a normal third baseman, the Red Sox would have had to play Alex Cora full-time. If you gave Youkilis the credit for the difference between half a season of Cora\'s at bats and half a season of Casey/Kotsay, anyone have any idea how that would affect his numbers?
Positional flexibility is nice, and Youkilis gets credit for it in his elevated FRAR/FRAA numbers, but the net offensive equation still boiled down to Lowell versus Cora/Casey/Bailey/Kotsay, and that\'s a loss.
(That person needs to sit with Joe Morgan in the \"time out\" corner)
He\'s a great player and a handsome devil, but I have a hard time seeing the argument for him being elevated above this crowd. Among CFs, Upton and Hamilton both had basically equal seasons, WARP-wise. Among the Indians, Lee clearly had the more valuable season. And the Tribe drastically underachieved this year.
So: not clearly the best at his position in his league, very clearly not the best on his team, no boost for his team\'s accomplishments... I\'m struggling to see how one could justify ranking him first among this lot.
WARP does factor in baserunning, right?