BBA Ballot: National League MVP Award, Pt. 1

Congratulations to Zack Greinke of the Kansas City Royals and Tim Lincecum of the San Francisco Giants for winning the BBA’s award for the AL and NL Cy Young awards respectively. This is a great year to debunk the myth of the pitcher win-loss record, and I’m happy that our group’s voters saw that these two pitchers were simply amazing and deserved their awards.

As for the MVP award, this season the National League is going to be a ridiculously difficult season to judge the MVP. There are as many as five excellent candidates for the award this season between pitchers and position players. How do we determine who is deserving of the award between these excellent players?

The Definition

Well, to begin, I will define the award and then once again go into the process of evaluating it using the necessary tools. I’m not going to go into aspects of the term “value,” which the official BBWAA is very vague about. My vote for value is strictly in terms of production. Just like in my previous votes, I am not assigning any value to being on a winning team, because my opinion is that the evaluation process which I am using (we’ll get into that in a moment) determines your sole production (mostly) outside the context of your team, and it would be unfair to the player to penalize his production for being on a poor team (that’s why my vote for AL MVP, if I were to have one, is actually Zack Greinke, though Joe Mauer will be a deserving winner nonetheless). Any qualms with the concept of production outside of a team context will not be heeded, mostly because I define value as context-neutral production and nothing else.

To evaluate production, I am once again going to use WAR. This time, I’m going to about it in a slightly different fashion. I wasn’t totally happy with my linear weights as they came out compared to those shown in other places such as FanGraphs and StatCorner, so for position players, I’ll be using FanGraphs’ wRAA to determine offensive contribution. However, I did introduce something different into the data, in the form of baserunning. Currently, FanGraphs includes stolen bases and caught stealing into the equation for wOBA. I used my wOBA weights determined through the same method FG uses (originally shown by Tom Tango here) to determine the stolen base and caught stealing totals that were added or taken away. I subtracted those and recalculated wOBA accordingly, using my recalculated wOBA scale. From there, I park adjusted the wOBA using Patriot’s 2009 park factors published here and recalculated wRAA.

After that process, I added the remaining inputs. For baserunning, I used Baseball Prospectus’ Equivalent Baserunning Runs. All other inputs for position players come from FanGraphs, including defense by UZR and the appropriate rate adjustments. Once the season is over, I’ll revisit the defensive numbers giving my usual weighting to Fan’s Scouting Report and TotalZone, to get a more accurate picture of defense, but for now, this is all I have and this is all I’m going with. For the runs to wins conversion, I used 10 runs / 1 win.

For pitchers, I did what I did for the rookies: I took an average of FIP, scaled to runs allowed, from FanGraphs and the average between StatCorner’s and FanGraphs’ tRA and used that as the pitcher’s runs allowed. I then used Pythagenpat to determine the WAR, with a replacement level of .38 for starters. One thing different in this case was the run environment. With the rookies, I used a run environment strictly involving NL play, tossing out Interleague Play. In retrospect, that seems inaccurate, as NL pitchers faced AL lineups this season, and those games shouldn’t just be discounted. To accommodate, I added in the AL’s offensive numbers from Interleague into the runs/game, getting a run environment of 4.49 runs/game average offense. I then park adjusted that total accordingly using Patriot’s park factors.

Two final things went into consideration as “tiebreakers.” A lot of people have difficulty with regards to the defensive input shown here because of UZR’s measurement error. Of course, the best way to get rid of such error is to have more data, but in this case we simply can’t do that. To account for this, I gave the benefit of the doubt to ties to the players with less defense, provided the defense was rated sufficiently higher than average. Yes, this does penalize players with great defensive numbers, but since we can’t be too sure of those numbers, I figured a tiebreaker should go to the player with the more certain numbers. I did this tiebreaker when players had a run difference of one to two runs*. This was strictly at my discretion, so it is a bit subjective, and I apologize. This tiebreaker came into play twice in the ballot’s 10 voted players

*Note: Keep in mind that this would not have mattered in my Chris Coghlan vs. Andrew McCutchen debate, because the two were very far from each other.)

The other tiebreaker was a difference between pitchers and position players. For a difference of a similar amount as the previous one between a pitcher and a position player, I gave the tiebreaker to the position player because of the uncertainty of the replacement level for pitcher in comparison to the relative agreement between multiple sources in the replacement level for position players. This came into effect once in the top 10.

Enough chit-chat, right? Do you understand how I did my ballot? Because if you do, what follows is the ballot itself, with some brief explanations. Keep in mind that this year, the differences between slots, especially at the top tier, were very small, as you’ll see in the ballot, so nothing is assured, believe me.

WAR below is listed to the second decimal place. I claim no accuracy to two decimal places, it just lets me differentiate between a few players. Let’s get to it.

10. Chris Carpenter, RHP, St. Louis Cardinals (6.08 WAR)

Three seasons ago, Chris Carpenter was the best pitcher in the National League by an enormous margin and was primed to dominate the league for years going forward. Instead, he suffered through two seasons of debilitating injuries and was a huge question mark going into this year. Well, Carpenter answered all the questions with his fantastic season. How does leading the NL in ERA, posting a 2.78 FIP that was second best in the NL, walking hitters at the fourth lowest rate in baseball, and leading a St. Louis staff into the playoffs along with teammates Adam Wainwright and Joel Pineiro. Carpenter was a house to face, and had he pitched more innings (he was the only pitcher among the Cy Young candidates to not reach 200 innings), he would have undoubtedly overtaken most of the other pitchers in the race along with being Comeback Player of the Year.

9. Prince Fielder, 1B, Milwaukee Brewers (5.93 WAR)

Fielder actually ended up 0.1 WAR behind the player Carpenter and was the beneficiary of the tiebreaker. Fielder hit .299/.418/.602, good for a park-adjusted wOBA of .424 in my calculations. This came out to a staggering 53 runs above average, which was only slightly mitigated by his poor baserunning (-5.4 runs on the bases) and the fact that he was a first baseman. This was also apparently his best season on defense, as UZR rates him as just average after three seasons of DH-level defense. Whatever the Brewers are planning for Fielder, he certainly helped their case by massively bumping up his trade value this season.

8. Dan Haren, RHP, Arizona Diamondbacks (6.25 WAR)

I spoke already about the excellent season Haren put up this year, but it bears repeating in light of this fact: Arizona’s Chase Field had the second highest run environment in the National League, behind only Coors Field in Colorado. That means despite Haren’s season not “impressing” as much as some of the others in the Cy Young race, his performance in a heavy hitter’s environment notched him up a good deal and, in my book, netted him a nice appearance in an MVP ballot.

7. Javier Vazquez, RHP, Atlanta Braves (6.34 WAR)

Again, I already discussed Javy Vazquez’s season, as he had a nice breakout year moving into the National League. As mentioned before, if it weren’t for a certain lanky kid in San Francisco, Vazquez would easily have my vote for Cy Young thanks to that deadly combination of high strikeouts and low walks. He’ll have to settle for eighth place in an MVP voting.

6. Adrian Gonzalez, 1B, San Diego Padres (6.34 WAR)

First base was a very plentiful position in the National League this year, and Gonzalez definitely made his case as one of the best in the league this season. Consider this: Gonzalez had a .402 wOBA playing in the most run-depressing environment that baseball has seen easily since the ’80’s and likely since the ’60’s (not sure about this, but it wouldn’t surprise me). After adjustment, Gonzalez put up a wOBA of .422, essentially identical production as Fielder given their run environments. Where Gonzalez had a nice advantage was in his defense (worth 4.3 runs this season by UZR), which combined with some other factors to yield an ever so slightly better season. It’s amazing that both of these first baseman paled in comparison with the guy further ahead in this list.

Here we then have a giant gap in WAR between the 6th and 5th best player in the NL (according to these measurements) this year. It seems like as good a time as any to stop and split up the ballot. Come back in a little bit to see #5 – #1!