What Can Be Learned From the AL MVP Voting?

by Jeff Zimmerman on Nov 21, 2009 10:40 AM EST in Awards & Honors

On Monday afternoon, I fully expect Joe Mauer to win his first MVP award after finishing sixth in 2006 and fourth last year. Joe led the AL in batting average while hitting for respectable power (28 HRs) in only 138 games, most of which were as a catcher. He led the Twins into the postseason after they trailed by three games with only four games left, thereby forcing and winning a playoff game with Detroit.

Other than Mauer winning, I expect Mark Teixeria to finish high in the voting after leading the league in HRs and RBIs (categories that have traditionally been important to voters). I also expect Derek Jeter to finish high even though he doesn't have great traditional stats this season (.334 AVG, 18 HR and 107 runs scored). He did have one of his best years with a 7.4 WAR value, his highest since Fangraphs began keeping track and his third-highest total when compared to Rally's WAR ranking.

Truthfully, I have no problem giving the award to Mauer. He had a great season and helped the Twins to the playoffs. The real question with today's vote, to me anyway, is where do Zack Greinke and Ben Zobrist end up in the rankings?

Using WAR values as a guide, both Zack (9.4 WAR) and Ben (8.6 WAR) had a better season than Joe (8.4 WAR). The problem is that the mainstream media is barely touting if at all Greinke or Zobrist. Going to MLB.com, Zack gets an honorable mention while Ben gets no mention (though four of Ben's teammates got mentioned: Evan Longoria, Jason Bartlett, Carl Crawford and Carlos Pena). At cbssportsline.com and USAToday.com they have no mention of either Greinke or Zobrist.

Two groups of internet voters have released their results. SBN, our parent website, had a vote and Zack came in sixth overall, while Ben came in second. When the Internet Baseball Awards released their results, Zack came in fourth and Ben came in sixth. I really don't expect either of these players to finish as that high in the final votes with the BBWAA, even though no two players in the AL did more to help their team win games this last season.

Let's look at how Zack and Ben final voting placement can say something about the makeup and thought process of the BBWAA.

Zack Greinke - Do pitchers even matter in the MVP voting?

Here is a little background on how pitchers have done in the MVP voting:

Since 1956, when the Cy Young Award was created, a pitcher has won the MVP Award 8 times or 8% percent of the time. Before the Cy Young was created, pitchers won the award about 1 in 5 times (19%).
The last pitcher to win the MVP was Dennis Eckersley in 1992.
Since 1992, the average final position for the top pitcher vote getter is 8.4, with Pedro Martinez about winning the award in 1999 by finishing with only 3% less vote than Ivan Rodriguez.

There is not much precedent for MVP voters voting for pitchers, especially ones with 16-8 records. The average finish of the top pitcher in the MVP voting has been between the 8th and 9th spot, so this makes a good dividing line to see if the writers noticed Zack's great season (despite the horrible defense and run support he had all season). Zack had the 31st highest ranked season ever in ERA+ with 205, which is the highest total since 2003 when Pedro Martinez finished with an ERA+ of 210.

If Zack finishes eighth or higher, hats off to the voters. If ninth or lower, shame on voters.

Ben Zobrist - Do name familiarity and defense matter in the MVP voting?

Before the season, many people were familiar with the final MVP candidates, but few were familiar with Ben, and the few that knew of him didn't expect great production. Offensively, his 2009 season was nothing to laugh about (9th in the AL in Runs Created), but his defense is where he separated himself from the others for the top WAR values for positional players. He had a UZR rating of 26.4 which was good for 2nd best among all position players in the AL.

The question is, "Will any of this production really matter." Even though the Internet voters gave Ben quite a bit of love, I can see the BBWAA giving him a cold shoulder. I would be surprised to see him break the top 10. I really want to see the percentage of voters that put him in their top 5. This will give a nice understanding of the percentage of voters that look at and use WAR (or similar metrics) to compare players. If the percentage of voters that put Ben in the top 5 is above 10%, I will be amazingly surprised.

---

The award season for MLB is a nice time to look back at the the players that had a great season. The problem is that sometimes players that should receive recognition don't.

Tweet Comment 30 comments | 0 recs |

Do you like this story?

Comments

Yabbut don’t other defensive ratings show Zobrist as not being such an amazing defender? I’m inclined to think he isn’t quite /this/ good.

by cubsforever on Nov 21, 2009 11:57 AM EST reply actions

+/- has him at +32 (in runs saved)

I take that with a large grain of salt. I agree that Zobrist had a stellar season and will probably be overlooked in the MVP voting but I would put him behind Mauer, Greinke and Jeter on my hypothetical ballot.

by James Kannengieser on Nov 21, 2009 1:23 PM EST up reply actions

“Using WAR values as a guide, both Zack (9.4 WAR) and Ben (8.6 WAR) had a better season than Joe (8.4 WAR). "

Does Joe’s WAR value even include defense?

by Karcotte on Nov 21, 2009 12:12 PM EST reply actions

Nope. Fangraphs WAR excludes catcher defense.

So really it’s reasonable to assume that Mauer produced at least 9 WAR of value, given his reputation as a good defensive catcher.

by cubsforever on Nov 21, 2009 12:15 PM EST up reply actions

Rally

Using a 5-4-3 weighting of the catcher defense used by Rally, Mauer’s defense would probably come out around 10 runs this year. Perhaps the number is less if you discount the games he played at DH, but it wouldn’t be less than an extra half win.

I think it’s entirely possible that Joe Mauer just had the best season a catcher has ever had. It is certainly not worse than fifth best all time, up there with Bench in ’72 and Piazza in ’97.

by Tommy Bennett on Nov 21, 2009 12:49 PM EST up reply actions

Had him at +6 runs I believe based on projection and d_f's Driveline article numbers for 2009

Marlin Maniac, a Florida Marlins blog
Come attend Intro to Sabermetrics 101!
Check me out at Beyond the Box Score as well.

by SFiercex4 on Nov 21, 2009 3:06 PM EST up reply actions

"In which Fangraphs WAR=concrete value"

even though no two players in the AL did more to help their team win games this last season.

I’m getting sick of it. There has to be a better way for people as smart as the writers here(lots of writers are guilty of it, actually) to compare players; this only pushes people away.

by Daniel Berlyn on Nov 21, 2009 1:36 PM EST reply actions

What other players would you nominate besides Zobrist and Greinke in this category?

I think that it just so happens that WAR is an accurate description in the regard.

The only player who really comes close in the AL is Jeter. Baserunning only adds about one run to his total, and if we’re taking Zobrist’s defense with a grain of salt, put me down for a salt-shaker’s worth for Jeter’s UZR numbers.

You’re right people may abuse Fangraphs WAR but I just don’t see how this is an example of it.

by Tommy Bennett on Nov 21, 2009 1:48 PM EST up reply actions

and yet you hate it

and still return. RBI don’t do it. W/L don’t do it. WAR, in all its imperfections, and the fact that Mauer’s defense isn’t included is one of them, is still a better measure of how much a player did to help his team win than anything else out there. If that pushes people away, so be it.

What measurement would you suggest? Team wins?

by chuckb on Nov 21, 2009 2:24 PM EST up reply actions

Here's my problem with people complaining about the MVP

It’s long been considered a position player’s award and even longer has been considered a position player’s award and also an award where offense is weighed most heavily. The voters still manage to mess it up, and they do this quite a bit, but people are complaining for the wrong reasons.

What you’re saying is that we should have a measurement for the best player. This is where you’re wrong; it should be up to the voters to use as many metrics and statistics as they can to determine who’s the most valuable. Using WAR gives a general idea but when comparing two players with similar value the margin for error is too big to simply say the player with the higher WAR is better.

by Daniel Berlyn on Nov 21, 2009 2:38 PM EST up reply actions

It's rather difficult to tell if this is a troll.

Are you denying any ability to quantify the number of runs a player contributes to his team? Or rather that there are error bars on the measurement? The latter is justifiable, but the former probably isn’t.

For the AL MVP this year, there are three players who were far and away ahead of all others. So yes, I agree, it may be difficult to sort out exactly who performed best. But what isn’t difficult is separating those three from their peers. They simply stood above.

by Tommy Bennett on Nov 21, 2009 2:52 PM EST up reply actions

Yes, and we don't really need WAR to see that

every evaluating method should be explored, not just the simplest. When some people say Ben Zobrist was a more valuable player than Albert Pujols and that’s that, I’m not buying it.

by Daniel Berlyn on Nov 21, 2009 2:59 PM EST up reply actions

No one here is saying that.

by Tommy Bennett on Nov 21, 2009 3:09 PM EST up reply actions

Or Mauer, for that matter.

by Daniel Berlyn on Nov 21, 2009 3:24 PM EST up reply actions

WAR is a starting point for the discussion and some people believe Ben had a decent year.

He was 9th in Runs Created. Should that only be used?

Should people with extreme outlying seasons not be considered for the award or people that are steady from year to year?

What does Win Shares have for a ranking (I am not a subscriber to BJOL)? They limit how much weight that there is given to position and defense?

I also believe some record being broken should get extra consideration (some hit .400 or HR record broken).

Jeff Zimmerman - Protecting the world from RBI's and Wins from my mom's guest house.

by Jeff Zimmerman on Nov 21, 2009 10:00 PM EST up reply actions

I'm interested in seeing how Zobrist does as well, with full expectation that he won't do so hot

His defense has to be taken with a grain of salt, so a tiebreaker should go to Mauer.

As for pitchers, there is doubt about where the replacement level is and how to calculate it, so we definitely can’t be too sure about the production of Greinke vs. position players (vs. pitchers is a different story, no contest).

That being said, all you have to do is define how you are voting, and it shouldn’t be an issue. If you take Zobrist’s (and everyone else’s) UZR at face value in your voting, and you determine the replacement level for pitchers, then there’s nothing wrong with casting your ballot with those definitions in mind. I did that for my thing, and honestly I probably would have voted Greinke at the top, followed by Mauer and Zobrist. Nothing you can say, other than that my definitions are wrong, and we can go for hours on that argument.

Marlin Maniac, a Florida Marlins blog
Come attend Intro to Sabermetrics 101!
Check me out at Beyond the Box Score as well.

by SFiercex4 on Nov 21, 2009 3:13 PM EST reply actions

Yeah, although I will add

I expect each one of these players to be significantly worse next season.

by Tommy Bennett on Nov 21, 2009 3:27 PM EST up reply actions

I agree with Daniel

WAR – at least FanGraphs’ implementation – has a lot of flaws and shouldn’t be the be all and end all of value. It doesn’t include catcher defense and base running. It also relies too heavily on UZR – does anyone actually believe that Zobrist played like a +27 run defender this year? That would be one of the best defensive seasons of all time, and coming from a guy who was below average at short before this year, I can’t believe that UZR represents his actual contributions on defense.

Furthermore, WAR includes no aspects of leverage or clutch hitting (or fielding for that matter), and to some people, that is considered an important part of value. To me, the main things WAR gives us are a baseline for fielding and playing time. The positional and replacement level adjustments are very close to perfect. All that’s needed then is estimates for offense and defense, and wOBA and UZR may not just cut it for you.

by vivaelpujols on Nov 21, 2009 3:54 PM EST reply actions

I get that +27 is very high

But I’m having a hard time finding a defensive stat that didn’t love Zobrist’s defense this year. He has a higher RZR than Utley, and as James noted above, +/- loved him. Certainly he should be expected to regress, but is there evidence that he wasn’t actually that good other than that it was unexpected?

by Tommy Bennett on Nov 21, 2009 4:11 PM EST up reply actions

I think so

First off, you should always regress defense, even for end of season awards. There is simply too much measurement error to take it at face value. Plus, the fan’s saw Zobrist as a middle of the pack defender this year.

http://www.tangotiger.net/scout/index6.php?prim_fld_cd=4

by vivaelpujols on Nov 21, 2009 4:17 PM EST up reply actions

Regressing

Usually you only regress in aid of prediction. I’m not sure I see the argument for regressing to evaluate actual performance. Measurement error swings both ways and you might be destroying useful information if you just regress everyone to the mean.

by Tommy Bennett on Nov 21, 2009 4:19 PM EST up reply actions

You regress

Because of the amount of uncertainty involved in UZR or other defensive metrics. When you see a guy put up a +27, you should assume that is part real skill, part luck (or deviation from true talent level), and part measurement error. Given the fans, and past UZR scores point to Zobrist only being a slightly above average defender, a lot of that +27 is built up by luck and measurement error. Luck shouldn’t be counted against him, but measurement error should – which is why you should regress.

by vivaelpujols on Nov 21, 2009 4:24 PM EST up reply actions

Viva -- Ignore defense and position

Do you think Ben will make the top 9? We was 9th in Run Created.

I don’t think he deserved it, but it seems that some people believe was good enough to be the 2nd or 6th best. Do you think this will extend to BBWAA?

Jeff Zimmerman - Protecting the world from RBI's and Wins from my mom's guest house.

by Jeff Zimmerman on Nov 21, 2009 9:54 PM EST up reply actions

No

Half the BBWAA don’t give shit about the Rays. Plus most of Zobrist’s value came from OBP, and we all know how the BBWAA think about that.

by vivaelpujols on Nov 21, 2009 10:14 PM EST up reply actions

I agree, to a point

I think WPA should be used as a factor, b/c it tells us something about “clutchiness.” I think a lot of metrics should be used. And I don’t think Zobrist was as good defensively as UZR has him but apparently other metrics have him as having a terrific defensive season as well and while he’s been a below average SS in the past, he only played 62 innings at SS this season. What does that have to do w/ how well he played 2B or RF?

Additionally, WAR’s not perfect b/c it omits baserunning but the difference between an average base runner and the best in baseball is about 1 win per season. It doesn’t matter that much. Does it make WAR slightly less accurate to not include it? Yes but the difference isn’t so great that we can’t say that Ben Zobrist had a fantastic season.

by chuckb on Nov 22, 2009 12:50 AM EST up reply actions

What does that have to do w/ how well he played 2B or RF?

It means he is less likely to be a really good defender, meaning that it’s more likely that a lot of his +27 score was due to measurement error.

by vivaelpujols on Nov 22, 2009 1:40 AM EST up reply actions

Another thing to keep in mind

Zobrist’s defense includes both OF and 2B, so his UZR is inflated as he racked up runs in the corners. However, he pays a price in positional adjustment, which is -1.1 runs versus the typical +2.5 runs accorded a second baseman.

by Tommy Bennett on Nov 21, 2009 4:17 PM EST reply actions

The part about 2B

he played 715 innings there, certainly enough to get a gauge but nothing concrete. Same goes for his RF defense. I’m sure if he played either for a longer period of time it would have regressed. There is, however, certainly something to be said about playing multiple positions well.

by Daniel Berlyn on Nov 22, 2009 3:46 PM EST up reply actions