When it comes to the MVP award, I prefer to set aside my scouting hat and focus just on performance measurement -- and we have plenty of tools available to us to help guide us to the right answer to the question: "Who produced the most value for his team this year?" And the stat that best gets at this question, in my opinion, is Wins Above Replacement, better known as WAR. Here's a table of the top candidates for the American League MVP award showing their WAR totals to date using two methods, one from FanGraphs (fWAR) and one from Baseball-Reference (rWAR), both sites I typically keep open in browser tabs from the moment I turn on the computer in the morning until I shut it off at night.
[h4]MVP Candidates[/h4]
A look at top candidates based on Wins Above Replacement from FanGraphs (fWAR) and Baseball Reference (rWAR):
| | |
Jose Bautista | 7.2 | 7.2 |
Dustin Pedroia | 6.8 | 6.1 |
Jacoby Ellsbury | 6.5 | 5.7 |
Justin Verlander | 6.2 | 6.9 |
Ben Zobrist | 6.2 | 5.1 |
Curtis Granderson | 5.6 | 4.2 |
Adrian Gonzalez | 5.0 | 5.2 |
[th=""]Player[/th][th=""]fWAR[/th][th=""]rWAR[/th]
The various types of Wins Above Replacement calculations share a common goal: To measure each player's individual value as accurately as possible. For a position player, that means totaling up the value of everything he produced as a hitter (and the value he destroyed every time he made an out); plus the value he produced or destroyed on the bases; plus the value he produced or destroyed on defense. For a pitcher, it means adding up the value of the outs he generated and subtracting the value of the hits and walks he allowed; it may also mean adjusting the value of the balls he allowed into play to try to back out any help received from his defense. In all contexts, the statistics should be adjusted for ballpark, although they're not adjusted to reflect the unbalanced schedules big leaguers face.
The defensive numbers bundled into WAR are somewhat controversial, of course, in part because they're new, in part because they're opaque, and more than anything else because there are multiple stats purporting to measure the same thing but giving us different results. But I respect the statistics even with their limitations, because when we're trying to get a reasonable measurement of defensive value, these statistics are better than a scout's eyes in a small sample of games, which is in turn better than a divining rod, which is in turn better than fielding percentage, which is the worst thing to happen to baseball since Bowie Kuhn finished trying to run the game into the ground.
With all of that in mind, here's a rundown of the leading candidates, including some more qualitative arguments about their cases, starting with the obvious -- or should-be-obvious -- MVP to date.
[h3]
Jose Bautista[/h3]
He's leading the AL in on-base percentage and in slugging percentage, with giant margins in both categories, while playing adequate defense across two positions. His OBP is somewhat boosted by intentional walks, but without them he'd still lead the league in OBP by 12 points. (Even without those intentional walks, his fWAR is still 7.0, tops in the AL). He's been far and away the most valuable player in the league, and the only reason I can see that he's not getting his due as such is that he plays for a non-contender.
Because, let's face it, that's the real problem with Bautista's candidacy: For the voters still clinging to an outdated notion of value, there's nothing he could do to earn the MVP award while he plays on a non-playoff team. But where can the argument be? You can't even make a win-probability argument against him -- that is, to say that his performance hasn't directly resulted in wins -- because he's leading the AL in win probability added (WPA), which factors in the timing (inning, score) of offensive performance and credits the hitter with the change in the team's probability of winning the game. And he's not leading by a little bit -- his net WPA of 6.50 is nearly 50 percent higher than the second-highest figure, 4.39 by
Miguel Cabrera. I don't advocate the use of WPA to determine the MVP because it's still context-dependent, but if you care about context, it's a sound measure, and points to Bautista in a landslide.
[h3]
Dustin Pedroia[/h3]
Pedroia's case here rests to a surprising degree on where he plays: he gets two boosts from playing a position up the middle and from playing it exceptionally well this year, a fairly extreme outlier compared to the rest of his career. Pedroia has made himself into an above-average defender, producing fairly consistent defensive ratings in each of his first four full seasons. This year, his UZR figure, representing runs saved above an average second baseman, is higher than the combined total for the previous two years and has him as the best defensive second baseman in the game (although the current crop of defenders at that position is pretty weak overall).
As a hitter, Pedroia ranks ninth in the American League in FanGraphs' Batting Runs, dead even with
Ben Zobrist, and only moves up the rankings on the high fielding rating and the boost he gets for playing second base. (In fact, the difference between Pedroia and Zobrist this year is almost entirely fielding, with small adjustments in Pedroia's favor for playing time and because Zobrist has played 33 games in the outfield.) An argument for Pedroia over Bautista involves team record and a fervent belief that the large uptick in Pedroia's defensive numbers is accurate. I'm skeptical on the second part, but even if I take it at face value the first part holds no water with me.
[h3]
Jacoby Ellsbury[/h3]
Ellsbury also plays a position up the middle and has had the good fortune to come up in enough critical situations that he's racked up the third-best WPA figure in the league in addition to earning some mainstream "clutch" points or positive qi or whatever sort of woo some use to decide whose team is good enough for its players to be considered for an MVP ballot. Ellsbury was one of the better defensive centerfield prospects I've seen, at least in terms of range, but his defensive scores as a big leaguer have been all over the place (more evidence in favor of the argument that a single season isn't a sufficient sample when attempting to judge a player's defensive ability).
But his defensive value isn't the big reason for the jump in his WAR totals -- he's started hitting for power like never before, generating more offensive value in four months this year than he had in over 300 games coming into this year and more than doubling his career home run total. He has shown the ability to drop the bat head and drive the ball out to right-center this year, so I'm not convinced it's a fluke, but the value created by the home runs is real either way. I am a lot more comfortable with his defensive ratings than I am with Pedroia's, though.
[h3]
Justin Verlander[/h3]
I'm not sure what sort of voting calamity would have to occur for a pitcher to win an MVP award, but indulge me for the moment, since the rules for MVP voters do not in any way exclude pitchers, and any voter who omits them entirely is violating the spirit of those rules, if not their letter as well. And such a voter is also ignoring the value that a top-end starting pitcher can create, whether it's
Roy Halladay this year in the National League or Verlander in the American.
The FanGraphs version of WAR relies on FIP, a very simple ERA estimator that is probably too simple -- a back-of-the-envelope calculation where a more rigorous one would do the job better -- and might slightly overrate Verlander, who fares a little worse in any of the other three major ERA estimators (SIERA, tERA, and xFIP) I know of. (An ERA estimator looks at the individual results allowed by a pitcher over which he has significant control, including strikeouts, walks, home runs, and groundball or line drive rates, and uses them to construct a more "neutral" ERA-like statistic that should give us a more accurate picture of how much the pitcher contributed to his team. They assume many things, including that all pitchers suffer roughly the same loss of effectiveness when pitching from the stretch.) Baseball-Reference's version of WAR is based strictly on a pitcher's ERA, so the boost Verlander gets from his ballpark is still present there, and the fact that he's been "lucky" on balls in play -- one might say he's been helped by his defense, assuming one hasn't actually seen Detroit's defense in action -- isn't factored out either.
That "luck," for lack of a clearer term, shows up in Verlander's career-low BABIP, or the batting average he's allowed on balls in play, of .234. Pitchers have little control over whether a ball in play becomes a hit beyond some ability to generate more groundballs or fewer line drives. So Verlander's dominance this year may be boosted by some good fortune that isn't likely to continue; many voters would disagree, but I prefer to use heavily normalized statistics where possible to try to isolate individual value, which would mean taking credit for that low BABIP away from Verlander, dragging his WAR even further below Bautista's. Even if he throws 260 innings or is credited with 26 pitcher wins, he's probably not going to be the most valuable player in the league.
[h3]
Adrian Gonzalez[/h3]
Yes, the RBIs are pretty, but they're a function of all the guys Gonzalez has had on base for him, and he doesn't play a premium position. Great player. Nowhere near MVP status.
[h3]
Curtis Granderson[/h3]
The most valuable position player on what may be the team with the best record in the American League by the end of the season, and if you throw out the defensive runs figures on FanGraphs -- not that I'm advocating you do such a thing -- Granderson is second in the league behind Bautista, as UZR shows him at nine runs below the mean for all AL centerfielders this year and Baseball-Reference concurs. He's also third in the AL in wOBA, a total-offense rate stat, behind Bautista and Cabrera. I don't know if Granderson is suffering from playing with a former centerfielder,
Brett Gardner, to his right, but I've never seen Granderson as a below-average defender, and he wasn't one statistically until this year. I wouldn't put him first on my ballot under any circumstances right now, but I could understand any voter who put him second or third due to skepticism at the implied collapse of his glove.
[h3]The National League[/h3]
Colleague Jayson Stark is breaking down the NL MVP race today, but since I know you'll ask, right now I'd have Halladay at the top of my ballot, followed by
Justin Upton and
Troy Tulowitzki, any of whom could end up the leader at the end of the season.