From the Archives
September 28, 2000

Much Valueless Pablum

Bias Inevitable in MVP "Voting"

September 28, 2000

It's September, time to turn to the perennial hot-stove-league issue of who the Most Valuable Player is in each league. And since the Presidential election is around the corner, we might ponder the nature of voting for a spell.

The major controversy about the MVP award, as ever, usually has to do with what it actually means. The criteria for it are vague. Some voters consider it to mean "most valuable to a team's success", and interpolate this to mean "most valuable to a winning team's success". Others simply vote for the player with the most distinguished playing record. Some voters take defense into account, others don't. Some voters don't consider pitchers as being worthy, since they have "their" award, the Cy Young, while others note the fact that a pitcher getting three starts in two weeks faces about as many batters as an everyday player has at-bats during that same period and logically include pitchers in the balloting. Some writers exclude DHs because they "don't play a position"; others would vote for starters but not relievers. Some voters publically discuss their ballot and their criteria, others treat it as a private matter.

These are all inconsistencies that both make the award an imprecise one and add to the annual fun in arguing about it. I'll of course join the fray at a later date and suggest my own candidates and winners. My topic today is why the voting itself is so messed up.

Yes, it's just an award. But I think it counts for something, especially in the long run. We wouldn't be discussing the MVP outcome in 1941 if as baseball fans it wasn't representative of something near and dear to the game. We love talking about excellence, and baseball is a game divided by seasons more than careers or individual games. I'd like the MVP selection each year to mean something, since a hundred years from now, it will still be written down somewhere.

The sad fact is, the MVP is selected by a group that, as a whole, is perhaps the least qualified in the current baseball community to do so: the baseball writers. Specifically, long-time members of the Baseball Writers Association of America, which excludes internet writers from membership, incidentally.

To refresh your memory, the MVP is voted on by two newspaper beat writers from each major league city, with cities with two franchises getting two extra writers. This is purely an artifact of the time when newspaper beat writers were the only outsiders covering the game, and from an era in which there were as many as ten daily papers in some cities. Times, obviously, have changed.

I love good baseball writing, and I don't mean to malign these folks as a group. I listen especially carefully when I hear a writer interviewed, and I love reading out of town newspapers. They're certainly, as a group, the most expert individuals you can find out of uniform. But they're underqualified to vote and be the sole determinants of the MVP and other awards, whether or not the MVP criteria are clear. Here's why.

* Writers are interested in good stories, and thus tend to vote for players who make good stories.

The joke goes that "MVP" stands for "Much-Valued Piece", aka a good story.

I heard one voting writer being interviewed on ESPN about the MVP recently describe one player as "having great numbers, but he's as exciting as a sofa". He also described another player, a previous winner, as being an interesting candidate purely from the perspective of whether it would make a good story or not if he were to repeat.

I'm just as guilty of this as the next writer -- we love our own stuff and love to re-read it, and players who are more interesting get our attention more consistently.

* Writers emphasise the contributions of the last two weeks of the season.

The voting takes place between the end of the season and the beginning of the post-season -- a one-day window. It seems obvious that recent impressions would sway a decision that should be based on a sober reflection of a full season.

Again, I draw my sources from those writers who have discussed their voting criteria -- I've heard at least one voting writer recently describe Mike Piazza's chances as being based on whether he got hot in the last two weeks of the season, which seems to be willfully ignoring 95% of the season -- but you can also look at the results of some years. Last year, Chipper Jones was hot down the stretch, having a celebrated streak at the end, and won the award despite having almost identical end-of-season numbers to a few other players whose teams also made the playoffs. The last-man-hot always wins the tiebreaker.

I'd add that this applies equally to Cy Young balloting. The ludicrous idea that there's a better pitcher anywhwere in 2000 than Pedro Martinez has surprisingly currency among the followers of David Wells to suggest that because he has a better winning percentage, he was a better pitcher. That is, it did until Wells fell off the last few weeks. Their relative standings in ERA, quality starts, strikeouts, opponents' batting average, etcetera, haven't really changed, but Pedro has pitched better on average than Wells the last two weeks.

* Writers are baseball insiders.

Again, I don't mean to suggest anything sinister or dishonest, but the fact is baseball writers rely on close relationships with players and management to earn their living. They're on the inside of the game, and tend to adopt the insider's concepts of what constitutes a valuable player. So even if they truly believe they're being objective, they're going to get skewed towards the viewpoint of the people they like and trust. There are lots of other sources of information, some of which can be more objective than the inevitable subjectivity of somebody so close to the game and its play. Writers are experts, but only one form of an expert.

* Writers don't see a lot of baseball games.

That may sound insane, since beat baseball writers see 162 games a year, in person, plus spring training. But think about it: they see one team play 162 games a year, another 13-15 teams play six to twelve times each, and that's it. They may catch a few games on TV, but they've got to cover their own teams before and after any given game, so they're seeing their own team in great detail -- and missing the bird's eye view of the majors as a whole. They thus have a poor basis for comparison, or at least a skewed one.

[continued]

I don't mean to hold myself personally up as being better than a beat writer, but I'd like to use myself as a counter-example. I see games at every level, from little league to the best major league games, and in a dozen or more cities. My obsession with the sport and my satellite dish conspire to let me see parts of as many as fifteen major league games a night (really) and as many as four complete games in a day. I don't know a given team in as great a depth as its beat writer, but I've seen every major league team play this year, more than once. I simply cover the sport differently than a beat writer.

National broadcasters also see virtually every team play, and they review lots of games for their highlight shows. They have large staffs doing detailed research before and after broadcasts. They've got the characteristic of serial depth on the games they see.

Even the much-derided (by insiders, at least) fantasy-rotisserie industry has a unique perspective. The people who write about and run leagues, the killer experts there, follow the inside numbers of a vast number of players at every level. That's an angle most writers can't get.* Writers vote on MVPs for both leagues, which is loopy.

This means that most writers get a vote on the best player of a league where they see, at most, six or eight games and at most three teams (via interleague play) every year. My suspicion is they tend to vote for candidates based on the recommendations of their colleagues, or possibly the raw stats, but not on a lot of first hand experience with players from the "other" league.

* Writers seem to believe in the RBI.

Writers aren't alone in this category, but I wouldn't be the first to observe inside baseball people are perpetually overimpressed with the significance of the RBI. Bonds is slugging up a ton, but because Jeff Kent has more RBI than Bonds -- largely due to the fact that Kent bats behind a great OBP guy, name of Barry Bonds, thus drives him home a lot -- I've heard several voting writers say they're not going to vote for Bonds. There's an extremely high correlation between the RBI champ in a given year and the winner of the MVP -- too high for it to be a coincidence. I can make the case against the RBI in another column, but as an inductive proof, I'd argue that this aspect of the results defaults the current voters as a group automatically.

Similarly, writers overvalue Wins in Cy Young balloting; you can tell just by the outcome of the voting.

* Finally: writers can be petty, like anybody else.

Gosh, what a shock, that a human being can be petty. Not to say this happens a lot: far from it. But there are a few celebrated examples, as when Ted Williams was left off the ballot by disgruntled Boston writers. I've no doubt this is repeated in miniature and more quietly every year, if only by one or two writers. The problem here is not writers being writers, but that there are so few voters, so that just a few people acting on emotion or personal like or dislike of a player can skew the voting.

So, as ever, I've got a favored solution. I'm not sure who came up with the idea originally, but basically you fix the skew towards the writers' views by adding a lot of other skews, to get a more well-rounded selection process.

Writers are like any guild: they're going to continue to maintain their privileges, and do it on the basis of their undeniable expertise. They can't help being who they are - nobody can. We've all got our biases and experiences to draw from.

I'm not in favor of taking the MVP (or Cy Young, or Hall of Fame) ballot away from newspaper writers - I want to enfranchise the rest of the baseball world. To paraphrase de Tocqueville wildly out of context, there's nothing wrong with Meritocracy that more Meritocracy can't fix. In context, this means extending the ballot to other segments. Let broadcasters vote. Let the players vote. Let the coaches and managers vote. Let SABR have a vote as a body. And yes, let the poor internet writers and other associated hangers-on who follow the game at various levels vote. Weight each sector's vote accordingly, and come up with a points formula to determine a consensus MVP.

Forget the controversy over the vague criteria. Until the voting base for these post-season awards is broadened, the MVP won't really mean anything.
The Baseball Crank may be contacted at crank@thediamondangle.com.
(c) 2001 Matthew Wall/The Baseball Crank.