MVP Voting and WAR

I agree, however what I said is that WAR is the best all-around value stat. I didn’t say it was perfect. A better method is to average the WAR from fangraphs with the WAR from baseball-reference, as one uses UZR and the other uses DRS. It just would have taken me too long to do that. Until we get better info from the Field FX system, UZR and DRS are the best thing available to us.

Batting average, like UZR, takes three years to fully normalize. How many times have we heard that Curtis Granderson doesn’t have a high enough average to be MVP?

People like to beat the drum about UZR, and the sample size is a relevant observation, but if defensive ability is going to be a factor in these types of things, I’d much rather the voters be conscious of UZR than of fielding percentage, gold gloves, or vague anecdotal evidence of that one diving catch.

Total Zone’s numbers are usually easier to swallow. There’s no better statistic for evaluating a single season’s performance than WAR. If you don’t like the defensive metrics, just use rWAR and judge defense with your own opinion.

There is no doubt that Howard’s 2006 season was the best of his career and he put up some gaudy counting stats as well. His rate stats were as follows: .313/.425/.659, good for a 162 wRC+. I think people are forgetting, though, how incredible Pujols has been over the past decade. In 2006, Pujols hit .331/.431/.671 – better than Howard across the board – for a wRC+ of 175. He also added more than a win on defense by both UZR and Total Zone, and was slightly above average on the basepaths. Howard was essentially average in the field and below average on the basepaths, and Pujols’ better line helped him put up a higher oWAR than Howard (7.0 to 5.6) despite playing in only 143 games vs. Howard’s 159. Pujols also hit the most home runs of his career (49, a personal best which still stands) in a season in which he missed nearly 20 games. Howard won the MVP essentially because he hit 9 more home runs and had 12 more RBI. It was fairly close – Pujols received 12 first place votes to Howard’s 20 – but it’s still ridiculous.

And what exactly is value other than the player who increases your win total the most? I can understand if you want to mix it with something that is not context neutral, like WPA or RE24. I’m not saying it should even necessarily go to the leader in WAR, but WAR can tell you how valid of a candidate someone really was. Everyone agrees that Morneau was a terrible choice, and WAR agrees. Because of the instability of UZR, if players are close to each other in WAR, you need to use other methods, but in the case of a vote like 2002, where Arod had 5.3 more WAR than the MVP, Tejada, it should be a no-brainer.

And why do you think comparing a player to the major league average is a bad idea? Is it of “dubious value” when economist use index years to compare inflation rates? They compare players to league average in a year because that average can change a lot, and it would be impossible to compare players across years without doing this.

Here’s an example. Player A hit 10 HR, had 53 steals and had a triple slash line of .354/.415/.542. Player B hit 49 HR, 7 steals and a triple slash of .336/.432/.685. Which player was better? I think it’s pretty clear from this information that it is player B, by a wide margin. Now, what if I told you that Player A was Honus Wagner in 1909, which was the year where there were the fewest runs scored per game in MLB history, and Player B was Todd Helton in 2001, in the middle of the steroid era? Are you still sure of who was better in his day? Stats like wRC+ and WAR answer this question. Wagner had a wRC+ of 205, meaning he was more than twice as productive as the average player in that year, and a WAR of 12.9. Helton had a wRC+ of 160 and WAR of 7.6.

So, although Helton’s counting stats were much better, comparing the player to league average in that year shows that Wagner was much much more dominant in his day. That is the value of comparing a player to “a player who does not actually exist.”

Yeah, that’s all I really meant it as. I’m not saying that voters should check fangraphs and vote for whoever is #1 on the list, because WAR still isn’t perfect until we have a more stable defensive stat. However, there are some instances where WAR shows you that the MVP choice wasn’t even close to being the best player, sometimes not even on his own team (here’s looking at you, Morneau).

The voters didn’t know what they were doing because as good as Howard’s numbers were, Pujols’ were better and he was on a playoff team. The voters seem so hesitant to vote for a player who didn’t lead his team to the playoffs, and in this instance not only did he not lead his team to the playoffs, he was not as good as another player who did.

I think the voting guidelines need to be cleaned up. Think of it this way, how could an AGon or Granderson be more valuable to his team than a Bautista? The Blue Jays have all of like 2 or 3 position players who are above average, while every player on the Red Sox and Yankees are above average. Taking Bautista off the Blue Jays impacts them much more than taking AGon or Granderson off their team.

The Blue Jays aren’t going to make the playoffs, but that isn’t for Bautista’s lack of effort, it’s because most of his other teammates suck. Should a player on a playoff team be rewarded because his teammates were good, while a player who was better and more valuable to his team is punished because no one around him played well? This is why I don’t like the “have to be on a playoff team” argument.

I’m not saying that voters should go on fangraphs, look at the WAR leaderboard, and vote for that guy, but it wouldn’t be a bad place to start. From there, you could take a look at things like WPA to get an idea of how much of the value he provided was in important situations. All I know is that the current “method” of voting has produced some clearly inferior MVPs. I would prefer if there was some equation that defined the MVP, but I know that will never happen.

Let me guess, you read that article at It’s About the Money Stupid? Carlos Lee plays left field, AKA the second easiest position to play on the field. Corner outfield is where most teams put their lumbering power hitters, and the ones who actually have good arms go in right. So Lee is being compared to some of the worst starting outfielders. That said, UZR says he’s saved 7 more runs than the average left fielder this year. That could mean that he’s had a larger portion of balls hit right at him that he doesn’t have to run much to get to. I don’t disagree that single season UZR isn’t the best option for the defense part of WAR, but that doesn’t completely trash WAR as a whole.

Lol. Go UZR! Go defensive WAR!

Is the first quote about pitcher WAR? The third quote I think is very key. We always have to remember that there is some error bar on WAR. You can’t definitely say that a player with 7.8 WAR is better than a player with 7.6 WAR, but the odds that a player with an 8 WAR is better than a player with a 6 WAR is pretty good.