One of the first ‘advanced’ formulas I encountered as a young baseball fan was a stat called ‘Runs Produced.’ The old Beckett Baseball Card Monthly ran an article detailing ‘Runs Produced’ late in 1990…I remember that Bobby Bonilla led the majors in ‘RP’ that year, with an even tally of 200 (he had 112 runs scored, 120 RBI, and 32 HR).
Older and wiser, I can see the obvious flaws in a metric like ‘Runs Produced’…maybe I can’t see any flaw in the metric of ‘Puns Reduced,’ but I know enough to realize that R+RBI-HR isn’t a sophisticated measure of the quality of a hitter.
Way back in 1990, however, the eleven-year-old version of me really thought that Runs Produced was something. If the eleven-year-old me had been given a vote for the NL MVP, I almost certainly would’ve gone with Bonilla over Bonds. What, after all, are the forty-eight extra stolen bases, twice as many walks, and drastically superior defense of Bonds, when set against Bonilla’s fifteen Runs Produced? Give that trophy to Bobby Bo!
(It’s worth noting that one voter did vote for Bonilla over Bonds in 1990. I’m not sure who it was, but I’m sure it wasn’t me. Really. I absolutely did not submit a fraudulent ballot for the Houston sportswriter. I didn’t know what the ballots looked like! I’ve never even beento Houston! Stop looking at me!)
Anyway, I think part of the reason I liked Runs Produced is because it gave me an argument about the game that could potentially separate my perspective from the common one. That’s part of the appeal of non-traditional statistics: they allow us a way of understand the game that is separate from the conventional understanding.
I was always looking for that: my favorite player was Dwight Evans; you had to look pretty far afield to find any metric that made Evans look better than Jim Rice. Or Dale Murphy. Or Dave Parker. One reason I’m so susceptible to the charms of advanced metrics is that advanced metrics typically consider walks. Dwight Evans drew tons of walks….it’s one of the few categories printed on the back of a Topps card that show up in bold for him (three times!) It’s one of the categories where Dewey trounces Rice.
So, Runs Produced was a pretty cool metric that I spent a bit of time tallying in notebooks. Then I discovered girls and forgot all about it.
* * *
I bring this up because we have a couple of interesting MVP votes coming up, and because I’ve noticed that a whole lot of people who write or talk about things like MVP votes have been talking a lot about the hot statistic du jour : Wins Above Replacement, or WAR.
WAR…(huh)…(yeah)…what is it good for?
Absolutely everything: WAR is one of those statistics that attempt to measure everything: to evaluate the totality of a player’s contributions to a team. How many wins did Mike Trout give the Angels, over a replacement-level centerfielder? Approximately ten wins, a staggering, historic total. The best total for a rookie, and the best total for a twenty-year player in baseball history.
I love WAR. I reference it in a lot of columns, including a few length columns about the worst MVP’s and Cy Young selections measured by WAR.
That said, I’ve noticed a somewhat alarming trend regarding this year’s MVP debate: a lot of people are saying things like:
“Mike Trout has a WAR of 10.7. Miguel Cabrera has a WAR of 6.9. So Mike Trout should be the MVP, obviously. Anyone who thinks otherwise is a twit.”
That’s a direct quote. Granted, it’s a quote from me, spoken to the blank wall in my office. Still, a google search of ‘Trout’, ‘WAR’, and ‘MVP’ also turns up:
“Mike Trout has a WAR of 10.3 while Miguel Cabrera is at 7.1. Trout has led in AL WAR ranking every full month this year with the exception of September, when he was second behind Adrian Beltre. As I’ve stated, I’m not a huge fan of saber metrics, but that stat is very impressive.”
That’s from a guy named Brandon Weeland, from Bleacher Report. He isn’t even a fan of sabermetrics (or sabers and metrics), but he’s impressed by Trout’s showing on the WAR boards.
ESPN’s Jason Stark is a fan of sabermetrics. What says he?
[T]hose of us who believe that (Mike Trout should be the MVP) don't believe it because we worship WAR, or because we see that Trout has accumulated more wins above replacement than Cabrera or anyone else.
Good….no WAR worship there.
We just understand that Trout's insane 10.5 WAR are one more clear indication that he's a better baseball player than even one of the greatest hitters of our lifetimes.
Ah.
Just to be clear: whatever version of WAR (fangraphs or bb-reference) you prefer, Trout does kick ass:
AL
|
fWAR
|
rWAR
|
Mike Trout
|
10.0
|
10.7
|
Robinson Cano
|
7.8
|
8.2
|
Miguel Cabrera
|
7.1
|
6.9
|
Looking at the three serious candidates for the AL MVP, Trout is clearly ahead of Cano and Miggy. It’s a bit of a pity that no one has really given attention to Robinson Cano’s incredible season, but the Trout/Miggy debate leaves no room for third-party candidates.
In the NL, the waters are a tad murkier:
NL
|
fWAR
|
rWAR
|
Buster Posey
|
8.0
|
7.2
|
Ryan Braun
|
7.9
|
6.8
|
Andrew McCutchen
|
7.4
|
7.0
|
If you go by fangraphs’ version of WAR, Posey is neck-and-neck with last year’s MVP, Ryan Braun. Baseball-Reference has a three-way race between Posey, Braun, and Andrew McCutchen, the guy who led the MVP race for much of the season, but faded down the stretch.
The sudden emergence of WAR as a well-known metric has taken the thunder away from other equally interesting metrics that measure the totality of players. There is, for instance,
Total Runs. This metric measures a players offensive, defensive, and base running contributions, and makes an adjustment for position. In the case of Buster Posey, you get:
Name
|
Runs Created
|
Runs Saved
|
Baserunning Runs
|
Position
|
Total Runs
|
Buster Posey
|
110
|
-1
|
1
|
30
|
140
|
Jason Heyward
|
94
|
20
|
6
|
19
|
139
|
Posey was the better hitter than Heyward, but the Braves outfielder gets some runs back on defense and base running. Posey, as a catcher, gets a higher position adjustment.
Looking at the MVP candidates by Total Runs gives us this:
AL
|
Total Runs
|
NL
|
Total Runs
|
Mike Trout
|
173
|
Andrew McCutchen
|
153
|
Robinson Cano
|
162
|
Ryan Braun
|
153
|
Miguel Cabrera
|
150
|
Buster Posey
|
140
|
Mike Trout still gets the AL MVP, but we suddenly have a dog fight between Andrew McCutchen and Ryan Braun in the National League. David Wright and Michael Bourn actually finished ahead of Posey in the Total Runs leaderboard, but we’re not getting them involved.
Total Runs attempt the same thing as WAR, but it comes to a different conclusion, at least in the NL.
Then there’s Win Shares.
Whereas WAR measures players against replacement level players and Total Runs measures players as pure contributors, Win Shares measures a player’s contributions to the number of games a team wins. If a team wins 100 games, that team is credited with 300 Win Shares.
People who talk about MVP’s frequently talk about the importance of ‘team’…of how much a player’s efforts relate to the success of a team. I think this is a legitimate point. What I wonder is this: if you’re going to use any advanced metric to decide the MVP, why use a catch-all metric that sets the player against an imagined replacement player, when you can use a metric that measures the player against their team’s actual wins?
If we use Win Shares, the three MVP candidates look like:
AL
|
Win Shares
|
NL
|
Win Shares
|
Mike Trout
|
38.4
|
Andrew McCutchen
|
39.6
|
Robinson Cano
|
33.9
|
Buster Posey
|
37.9
|
Miguel Cabrera
|
31.7
|
Ryan Braun
|
28.4
|
Mike Trout is still ahead in the American League race…he is clearly the AL MVP, by any advanced metric.
In the National League, Pirates centerfielder Andrew McCutchen pulls to the front of the pack. Astonishingly, Andrew McCutchen rates as a slightly better player than the much-lauded Mike Trout…if we round his tally up, McCutchen becomes one of the very few players in history to post a 40-Win Share season.
* * *
There’s nothing wrong with falling in love with a stat. I loved Runs Produced for a while, and I still hold some affection for it.
And it’s understandable why we fall in love with cool statistics: they let us see the game in a new way, in a way that challenges us. It’s fun to argue about Cabrera/Trout. And: it’s possible we’re getting closer to a truer way of seeing the game.
But...it seems hardly better to hold up a statistic like WAR to be the only line of discussion than it is to praise RBI’s as a sacred cow of hitterly greatness. Arguing that Miguel Cabrera’s Triple Crown automatically makes him the MVP is silly, but arguing that Mike Trout is the MVP because of his 10+ WAR is almost as bad.
(No, it’s not! Mike Trout all the way! TRRRRRROUT!)
Just to wrap this up: the MVP candidates, by Runs Produced:
AL
|
Runs
|
RBI
|
HR
|
Runs Produced
|
Miguel Cabrera
|
109
|
139
|
44
|
204
|
Mike Trout
|
129
|
83
|
30
|
182
|
Robinson Cano
|
105
|
94
|
33
|
166
|
NL
|
Runs
|
RBI
|
HR
|
Runs Produced
|
Ryan Braun
|
108
|
112
|
41
|
179
|
Andrew McCutchen
|
107
|
96
|
33
|
170
|
Buster Posey
|
76
|
103
|
24
|
155
|
My vote would be for Mike Trout and Andrew McCutchen as this year’s MVP’s. But my eleven-year-old self says Miggy and Braun all the way!
Dave Fleming is a writer living in Wellington, New Zealand. He welcomes comments, questions, and suggestions here and at dfleming1986@yahoo.com.