Tuesday, November 28, 2006

Baseball's MVP debacle: if voters know more, Morneau no more?



I can't claim to be an unbiased observer either.
Artwork by James Blagden of No Mas.
The front pages of the New York Daily News and Post last week decried the injustice that Yankees shortstop Derek Jeter, long considered the frontrunner to be named the American League's 2006 Most Valuable Player, lost out to the lesser-known Justin Morneau of the Minnesota Twins.

It was a close final tally:
  1. Justin Morneau: 320 points (including 15 first place votes)
  2. Derek Jeter: 306 points (including 12 first place votes)
  3. David Ortiz: 193 points (no first place votes; one second place vote)
  4. Frank Thomas: 174 points (no first place votes; three second place votes)
  5. Jermaine Dye: 156 points (no first place votes; one second place vote)
This list betrays several biases among baseball journalists. One is that they consider designated hitters like Thomas and Ortiz, who do not play the defensive half of the game, essentially never valuable enough overall to merit MVP status. Since the Baseball Writers Association of America first started awarding the prize in 1931, exactly one DH has won: the Angels' Don Baylor, in 1979. Frank Thomas, Mo Vaughn, and Jason Giambi, all of whom have spent parts of their career at DH, have all won, but only when they spent most of the season at first base, though that is one of the least demanding defensive positions in baseball.

Another bias is against pitchers. There are plenty of good hitters in the league, but notoriously few ace pitchers. Measured against the average-quality pitcher who would likely replace him, a great pitcher like the Twins' Johan Santana is worth a good 15 or so wins, which is more than you can say for most hitters with 30 or 40 home runs. When the award began, pitchers were often chosen; the (then Philadelphia) As pitcher Lefty Grove was picked in the award's first year, and in 1968, ironically considering this year's World Series matchup and the poor showing by Tigers pitchers, starting pitchers for the St. Louis Cardinals and the Detroit Tigers won in both the National and American Leagues (they were Bob Gibson and Denny McLain, respectively).

Since the following year's awards, there have been 77 MVPs of the National and American Leagues named (there was a tie in 1979), including Joe Torre in 1971, when he hit .363 for the Cardinals, and Barry Bonds seven times, in 1990, 1992-93, and 2001-04, a testament to either herculean discipline or drugs or both. Only five of these 77 have been pitchers: the As' Vida Blue, the Brewers' Rollie Fingers, the Tigers' Willie Hernandez, Roger Clemens in his Red Sox years (many say he deserved the title last year), and Dennis Eckersley of the As.

It's hard to compare the value of starting pitchers to that of offensive players, who play five times as often and play all game, and whose defensive value is much harder to quantify. But the controversy over Morneau's MVP win this year comes because he is so easy to compare to Derek Jeter, another hitter and infield defender. Jeter's defense as a top shortstop was certainly more valuable than Morneau's as first baseman, and Jeter's batting average was higher, but Morneau led American League fielders in RBIs, behind Ortiz, who was (usually) kept at DH. And RBIs, which are thought to reflect a player's ability to focus his ability at clutch times when it is most needed, rule the roost of clucking sportswriters, eclipsing batting average and home runs as a determining factor in choosing the MVP.

RBIs are certainly important, and any Boston fan will agree that Ortiz seems more reliable with every man on base when he comes to bat. But according to Michael Lewis's Moneyball, statistical analysis shows that the most accurate of the common measures of a player's offensive value is OPS--on base percentage plus slugging percentage. Even better is on base percentage plus twice slugging, which I'll call OP2S.

Average OBPs are something like .300, which means the player gets on base, through a hit or a walk, 30% of the time. Average SLGs are maybe .350, and are calculated like batting average but with multiple-base hits counting double, triple, or quadruple, depending on how many bases the hit earns; a .350 SLG means the player earns on average 35% of one base per plate appearance, which might mean a single 35% of the time, of a home run 9% of the time. The average OP2S turns out to be something like 1.000, while the maximum possible OP2S is 9.000--a batter who hits a home run every time. Since the median batting average is around .225 or .250, I suggest recentering the OP2S s by dividing by 4.5; this produces numbers in the familiar range of batting averages, where .275 is decent, .300 is very good, .325 is great, .350 is incredible and anything higher is phenomenal.

Here are some of the top AL hitters, with their 2006 OP2S scores (including postseason), recentered:
  • Travis "Pronk" Hafner (DH, Indians) - .390
  • David Ortiz (DH, Red Sox) - .374
  • Jermaine Dye (Right field, White Sox) - .362
  • Jim Thome (DH, White Sox) - .358
  • Manny Ramirez (Left field, Red Sox) - .358
  • Jason Giambi (DH, Yankees) - .340
  • Justin Morneau (First base, Twins) - .332
  • Vlad Guerrero (Right field, Angels) - .330
  • Paul Konerko (First base, White Sox) - .330
  • Frank Thomas (DH, Blue Jays) - .327
  • Joe Mauer (Catcher, Twins) - .321
  • Carlos Guillen (Shortstop, Tigers) - .320
  • Alex Rodriguez (Third base, Yankees) - .320
  • Mark Teixeira (First Base, Rangers) - .311
  • Raul Ibanez (Left field, Mariners) - .308
  • Derek Jeter (Shortstop, Yankees) - .307
  • Miguel Tejada (Shortstop, Orioles) - .306
  • Michael Young (Shortstop, Rangers) - .283
  • Ichiro Suzuki (Right field, Mariners) - .267
One thing to notice about this list--besides that half these guys are from AL Central division teams--that there are several DHs, basemen and outfielders, and even a few shortstops, but just one catcher: Joe Mauer. Catchers don't usually bat .347, and they definitely don't usually have OPS numbers to compare with, say, Frank Thomas and A-Rod (who was notoriously inconsistent, but actually had a great year).

Also notice that Pronk Hafner was number one with a bullet in OP2S, pacing Ortiz, before he injured his hand in September. He still managed to have an incredible six grand slams, as well as 117 rbis, almost as many as Morneau though he did it with 150 fewer at bats.

Finally, note that Frank Thomas has no place being nominated, least of all being placed second in three ballots. There are four DHs in the league who teams would take over Thomas in a heartbeat; well, three and Jason Giambi, who had a great season but still makes the entire city of New York hold its breath for eigth months of the year.

So who should have been MVP? Here are, as I see it, the top ten offensive players who should have been considered for the MVP, in reverse order. Actually eleven, since I cheated and manufactured a tie. Note that I excluded pitchers entirely.

10. Miguel Tejada (.306 OP2S)

Not much to say; he's a great shortstop, and the rare longball hitter who doesn't strikeout too much. He played every single game of the season. But he doesn't have Jeter's fire--the fire that makes Jeter steal bases and make insane defensive plays.

9. Travis Hafner (.390 OP2S)

Manny Ramirez and Pronk played only 130 and 129 games, respectively, but though injuries kept them from having full seasons, they still posted outstanding numbers. Hafner's 117 RBIs is comparable to Morneau and Dye's, though he had 150 fewer at-bats than Morneau. While I understand that a major injury reduces a player's value, you can't deny that his hitting was at the living legend level of Albert Pujols (whose OP2S was .394), and he was incredibly valuable to a team that would have been in the pits without him.

8 (tie). Paul Konerko & Vlad Guerrero (both .330 OP2S)

Like Jeter, Vlad makes hits (both had 200+) and gets on base; and like Jeter, once there he steals more. You never see him strike out, he had 116 RBIs this season, and he's a great outfielder.

Konerko is solid and dependable, and buried Jeter in home runs and RBIs, with similar numbers in other categories. But niether Konerko nor Guerrero brings the defensive importance of Jeter.

7. Derek Jeter (.307 OP2S)

What can you say about Jeter? He's been excellent for his whole career, and steals bases like he's still 25. He's not top tier with home runs or RBIs, and he strikes out more than he walks, but he makes pitchers work and he hits sweet, reliable short balls, especially doubles. If it's possible to make a defensive play, Jeter will make it. Even without the notoriously subjective "intangibles" like Jeter's supposed leadership, he's incredibly valuable.

6. Carlos Guillen (.320 OP2S)

Guillen lags behind Jeter in several ways. Guillen can't really steal bases, though he tries, and he's not the wunderkind playmaker on defense. But he is even smarter than Jeter when at bat. He runs up pitch counts and walks a lot. He's great at smallball, mediumball, and longball; this year he had 174 hits, 41 doubles, 5 triples, and 19 home runs, surely approaching some kind of record for balance. He's slightly worse on the field than Jeter, but he's more than slightly better as a batter.

5. Jim Thome (.358 OP2S)

His fellow teammate Jermaine Dye, when endorsing Morneau and Jeter both for MVP at the end of the season, announced that he recognized that "to the winner go the spoils". But can't we fight the bias within us? With the departure of Frank Thomas, Thome became the resident big slugger on the team. And though he was outhit by Dye, didn't play defense, and missed 19 games, he still hit 42 homers and brought 109 guys home. If the other guys on his team had gotten on base more often, he or Dye could have had the RBI crown.

Manny Ramirez, also of .358 OP2S, would be in this spot, but injuries and immaturity kept him down to 130 appearances. Still, note that Manny had more RBIs than Jeter, despite his 174 fewer at-bats.

4. Justin Morneau (.332 OP2S)

His second half of the season was incredible. His talent is undeniable, and he's clearly someone who rises to the challenge of tense situations. But RBIs are a fickle measure. Last year he had 79; you have to go back four years to find a year that Ortiz had that few (his last season on the Twins, ironically). Maybe he's on an upward trajectory. But I'll take anyone's even money bet that next year Morneau has fewer RBIs than he had this year, not because he'll be worse, but because the stars won't align for him to hit the RBI jackpot.

This raises a question of sports philosophy: when we decide on an MVP, what are we saying? That this player turned out to have been the most valuable? Then there is a strong case for Morneau--he was, at least, in the right places at the right times. Are we saying that, with reflection, that this player is the one we would pick first for a new team? Then Morneau isn't it. Neither is he the player that we would pick first for a new team if we had a crystal ball back in winter 2005.

3. Jermaine Dye (.362 OP2S)

Dye hits the long ball and gets runners in. Morneau had 10 more RBIs, but differences like that are small in a measure as much about luck as clutch value. Dye's slugging percentage is .069 greater than Morneau's, meaning that Dye hit for about 7% more bases than Morneau per plate appearance. No sane GM would give that up just to cross their fingers and hope that Morneau's entire career before last June, and not his three subsequent months, were the fluke.

Losing Dye would have been worse for the White Sox than losing Morneau would have been for the Twins.

2. Joe Mauer (.321 OP2S)

Mauer's OP2S isn't the highest in the league, but when you consider that he's an exceptionally talented catcher, it's clear that he was wrongly ignored in the balotting this year. He played 140 games, fewer than most MVP candidates but a lot for a catcher. He had only 54 strikeouts, more hits than Ortiz or Dye (including four triples), and a respectable 84 RBIs and 13 home runs.

You have to go down to the middling Jorge Posada or the hobbled Jason Varitek to find another catcher in the AL with anywhere near the hitting ability of this guy. As a catcher, he's widely acclaimed, and far more valuable defensively than all but the shortstops on this list; as a hitter, he rounds out the top ten. If Ortiz wasn't so unmistakably explosive, he'd be the most valuable player.

1. David Ortiz (.374 OP2S)

He's a designated hitter. He does not often contribute on defense. But when he plays first base, he's fine. Would the Twins trade Morneau for him if Ortiz had to play first, and the pay was the same? You bet they would. That goes for just about any player in the American League. If Joe Torre became GM of the Red Sox, knowing that the Sox need a shortstop and have depth at first base and DH, would he trade Ortiz for Derek Jeter? I don't think he would. Only Johan Santana and one or two other pitchers can compare with what Ortiz gives his team.

As for the MVP voters, if they are making RBIs king they should note that Ortiz had seven more of them than Morneau, despite appearing at the plate 38 fewer times.

And as for Santana, which player would a team rather have? I leave that to wiser baseball philosophers. For now, I'll just say that we'll always love you, Big Papi, and that you are, at a minimum, the AL's most valuable batter. Hands down.