BP Comment Quick Links
November 12, 2012 Pebble HuntingHow Many MVP Votes Will Ryan Braun Lose?
A few years back, Jay Jaffe introduced an MVP Predictor formula called JUMP on Baseball Prospectus. It was, if his descriptions of his spreadsheets are any indication, a spectacularly messy equation, befitting the complex and irregular methods voters use to choose their MVPs. As Jay wrote at the time,
Which is to say: stories, not just statistical inputs, tend to guide MVP voters, and no two stories are ever the same. Even when stories resemble each other, they’re not quite the same. Which is a way of saying that what I’m about to talk about is limited by all the known unknowns and unknown unknowns. Ryan Braun (bold denotes a league-leading mark):
You could hardly draw up two more similar seasons, but of course there are two plot twists in the second story: Braun's team didn't make the playoffs in 2012, and Braun had been tainted by a positive test for testosterone by 2012. Baseball writers love to reward players who make the playoffs. My general sense, and probably your general sense, is they also like to punish players who have been connected to performance-enhancing drugs. This wouldn't make writers bad people; it would make them like most other groups of people. Or maybe you and your significant other were the two people who saw The Beaver on opening weekend? From an ESPN chat at the end of the season:
Exactly how much will Braun's vote suffer from his positive test? On Thursday, we'll find out where Braun finishes, and with how many votes. Today, we can talk about where he should finish, based on the established voting standards of the Baseball Writers Association of America. In 2001, Jonathan Bernstein wrote on this site about his MVP prediction system. It's much simpler than Jaffe's:
Bernstein later concluded that his system, which predicted the National League's winner in all but three seasons between 1969 and 1993, wasn't quite as effective in the Wild Card era. Perhaps, as far as predicting the winner itself, he's right. But in predicting the order of finish, his formula correlates just about as strongly now—over the past three years—as it did from 1987 to 1989 (a stretch I used for comparison because the average run-scoring environment was reasonably close to the current one and because it covers a variety of run-scoring environments). Among offensive players who received at least one MVP vote in those three seasons in the late 1980s, Bernstein's points had a .64 correlation to MVP vote totals. From 2009 to 2011, the correlation was .62; if we treat wild card berths as equal to division titles, the correlation for the latter stretch goes to 0.63. Anyway, long way of saying it's not a perfect system, obviously, and no system is; but that it's got some merit. This year, Braun gets three points: he led the league in home runs, he batted better than .300, and he drove in more than 100 runs. Buster Posey gets five points. Three .300 hitters on playoff teams—Jon Jay, Yadier Molina, and Marco Scutaro—also get three points (reflecting, perhaps, the post-WC flaw that Bernstein indicated). A bunch of people get two: Headley, Laroche, some Cardinals. But, given the basic outline of voters' tendencies that Bernstein identified, Buster Posey should win the award and Ryan Braun should finish second, or close to it. Caveats with Bernstein's method aside, let's look at some more specific comparisons. Here’s how I would describe Ryan Braun’s candidacy, as a non-PED narrative:
Who are the players who, in recent years, have most closely matched that description? 2011: Matt Kemp led the league in home runs and OPS+ (and RBI) while playing for a team that was competitive but missed the playoffs. He batted over .300 and also stole bases. He played an up-the-middle position and wasn’t bad at it. He ranked first in an average of the most famous win-above-replacement metrics, including WARP. MVP finish: Second 2011: Joey Votto finished third in home runs and fifth in OPS while playing for a team that was competitive but missed the playoffs. He batted over .300. He played a corner position but wasn’t bad at it. He ranked third in an average of the most famous win-above-replacement metrics, including WARP. MVP finish: Sixth 2010: Jose Bautista led the league in home runs and OPS (but not RBI) while playing for a team that was competitive but missed the playoffs. He batted under .300. He played a corner position and was bad at it. He ranked sixth in an average of the most famous win-above-replacement metrics, including WARP. MVP finish: Fourth 2010: Miguel Cabrera led the league in RBIs and OPS+ while playing for a team that was competitive but missed the playoffs. He batted over .300. He played a corner position and wasn’t good at it. He ranked sixth in an average of the most famous win-above-replacement metrics, including WARP. MVP finish: Second 2009: Prince Fielder led the league in RBIs (but not home runs) and was second in OPS while playing for a team that was competitive but missed the playoffs. He batted .299. He played a corner position and wasn’t good at it. He ranked fifth, seventh, and 13th in the most famous win-above-replacement metrics, including WARP. MVP finish: Fourth 2008: David Wright finished second in RBIs and seventh in OPS while playing for a team that was competitive but missed the playoffs. He batted over .300 and also stole some bases. He played a corner position, but it was third base and he wasn’t bad at it. He ranked fifth in an average of the most famous win-above-replacement metrics, including WARP. MVP finish: Seventh 2007: Magglio Ordonez led the league in batting average and was second in RBIs while playing for a team that was competitive but missed the playoffs. He batted over .300. He played a corner position, and the metrics disagreed about whether he was good or bad at it. He ranked fourth in an average of most famous win-above-replacement metrics, including WARP. MVP finish: Second 2007: Chipper Jones led the league in OPS and was second in batting average while playing for a team that was competitive but missed the playoffs. He batted over .300. He played a corner position, but it was third base and he seemed to be okay at it. He ranked fifth in an average of the most famous win-above-replacement metrics, including WARP. MVP finish: Sixth Again, we find that no story is the same as Braun's story. Kemp's case, for instance, was stronger than Braun's is, as he played the more valuable position, was better by stathead measures, and led the league in RBIs. Chipper Jones' case was weaker but isn't a perfect analogy because third base is neither up the middle nor as valueless as left field. Etc, and so on. But Braun's 2012 season would blend in nicely if thrown into a basket with these years, and the voting results of these years were pretty consistent: 2, 2, 2, 4, 4, 6, 6, 7. No winners, but usually in the top five. So, without the PED scandal, Braun could have expected to finish somewhere between second and sixth, and probably closer to second. He could not have really expected to finish first. You could make an argument that he should win; but the BBWAA can't make the argument that, based on their past voting tendencies, he should win. It's also hard to argue that he should finish lower than fifth. And we already know that he won't, as the BBWAA announced its finalists and Braun is among them. The gap from second to fifth, in votes, is usually pretty large. So a few voters could leave a player off the ballot and it wouldn't usually be enough to drop him from second or third place all the way out of the top five. That might have happened to Braun, and we'll know in a few days. But, to sum this thing up: assuming he doesn't finish first, Braun will land within the exact range of where he likely would have without the testosterone results. When it comes to HOF voting, the BBWAA seems intent on asserting its values. But when it comes to MVP voting, this dog didn't bark.
Sam Miller is an author of Baseball Prospectus. Follow @SamMillerBB
7 comments have been left for this article.
|
Braun's seasons really does pose the question, what impact do PEDs really have.
I ran across a column saying everybody should vote for Braun because this season PROVES he wasn't on steroids last year because otherwise he would have gotten worse. I tend toward your read on it, though.