The AL MVP: Keep on Debatin’ in the Free World

From the newspaper to the Internet, from ESPN to MSNBC to numerous pages hosted by Blogspot, the A-Rod/Ortiz AL MVP debate just won’t stop.
Some people have shown A-Rod to be more deserving of the MVP. He out-homered and out-fielded DH-extraordinaire David Ortiz. Others have, using convoluted metrics, done everything to show why Ortiz was more valuable to the Red Sox than Alex Rodriguez was to the Yankees.
One of the more popular, and pointless, arguments floating around the baseball world was the one espoused by Mike Celizic on MSNBC. Calling the vote an “injustice,” Celizic manages this gem:

But the numbers that count more than any other are production from the seventh inning on games when his team is either a run down, a run ahead or tied. That’s clutch time, when games are won or lost, when a team’s best hitters are standing in against the other guy’s best reliever, when the pressure is ratcheted up to nearly intolerable levels.
Ortiz had 78 at-bats in clutch time and he emerged with a .346 average, 11 home runs and 33 RBIs. A-Rod had 75 at-bats in clutch time, and he hit .293 with four dingers and 12 RBIs.
That’s nearly three times the homers and RBIs when Ortiz’s team most needed them. There’s not a player on the Red Sox who won’t tell you that without Big Papi, they don’t make the playoffs. No one in Boston will argue that anyone else on the team or in the league was more valuable…
But if you asked Yankees fans who the team’s MVP is, I don’t think A-Rod would win the vote. Rivera would probably win for the very good reason that no one was more critical in the Yankees’ winning the division than Mo. Jason Giambi might even finish higher than A-Rod, because, as good as A-Rod was, Giambi’s hitting from June through the end of the year was the real difference-maker in the Yankees’ run. Derek Jeter, the most feared clutch hitter in the Yankees lineup, would also take votes away from A-Rod.

Celizic’s argument is absurd on so many levels. Yet, this has been repeated on countless blogs on numerous discussion threads arguing the award. First, why do those close and late numbers count more than any other? It’s just as important to score runs in the first inning as it is in the seventh inning. If you’re winning, you’re winning. The runs count the same.
Next, Celizic gets into some good old fashioned Derek Jeter worshipping. Let’s look at what he’s saying about the close and late situations and throw that out the window. As much as I think Derek Jeter is the current heart and soul of a team growing increasingly soulless, Derek was not clutch this year. If Ortiz wins the MVP award for being great in close and late situations, it’s borderline ridiculous that Celizic would then argue that Jeter deserves the MVP on the Yankees over A-Rod. In those same close and late situations, the Yankee Captain hit a pedestrian .267/.353/.400 with 3 home runs and just 11 RBI in 90 at bats. With runners in scoring position this year, Jeter hit .261/.386/.355 with 2 home runs and 51 RBIs in 138. Now, I love the way Jeter plays, but let’s not get carried away here. He didn’t really come through that often in the clutch this year. It happens. But it’s absurd to say Jeter would take away votes from A-Rod because of his clutch hitting when it didn’t happen this year. If close and late were the only indicator of an MVP candidate, as Celizic seems to suggest, well, cross Derek off that list.
Meanwhile, with everyone so fascinated by this close and late scenario, I took a look at another scenario. This time, I compared A-Rod to Ortiz when playing against the American League’s other .500 ballclubs. I looked at how A-Rod did against the Red Sox, the White Sox, the Indians, the Twins, the Angels, and the A’s. I looked at how Ortiz did against the Yankees, the White Sox, the Indians, the Twins, the Angels, and the A’s. I then calculated runs created using the 2002 version of runs created. Here’s how the two compared.

A-Rod Ortiz
AB 213 214
Hits 71 59
HR 20 13
RBI 47 48
BB 26 38
K 42 48
AVG .333 .276
OBP .418 .385
SLG .648 .533
RC 68.31 47.73

So what we see emerging here is a different picture of clutch hitting. While Ortiz may succeed in “late and close” games (which, by the way, does not distinguish between games when the Red Sox were ahead by one run in late innings or behind, an important distinction), Alex Rodriguez seems to excel against the better teams. He had a higher on-base percentage by over 30 points, a higher slugging by over 110 points, and he helped create more runs.
So does that mean that Alex Rodriguez was definitely the MVP? People arguing for David Ortiz say that his close and late hitting is a good enough reason to award him the trophy. I, however, say no, this doesn’t prove anything.
All my thought experiment shows is that if you take the right numbers you get a situation where one of these players emerges clearly superior to the other. A-Rod here steps it up a notch against better teams. Ortiz seems to excel with the game on the line. Does one player’s winning the trophy over the other constitute an “injustice,” as Celizic said? No. Of course note.
This year, the MVP race was exceeding close. Both players deserved it, and only one could win it. While people can argue who is the better MVP candidate until they are blue in the face, the truth is that both of David Ortiz and Alex Rodriguez was very valuable to their teams. Beyond that, it’s simply up to the voters, however flawed a process that may be.


10 Responses to The AL MVP: Keep on Debatin’ in the Free World

  1. Johnson says:

    Good stuff Ben… It amazes me that nobody seems to have considered that Ortiz had Manny Rameriz batting behind him… look at the RISP stats… they are about as good, perhaps better, than those of Ortiz.
    A pitcher facing Ortiz in a late/close situation has little option but to give him good pitches since Manny comes up next and hits over 100 points better with runners on.
    Situational hitting is dependent on a lot of things and should be taken with a grain of salt… overall hitting gives a better complete picture.

  2. BosoxBob says:

    As the author/researcher who produced the “convoluted metrics”, I’m sorry if my story did not explain my methods clearly enough. The three metrics I combined to account for situational performance (Win Probability Added), defense (Win Shares), and Replacement Level (Equivalent Runs/Runs Above Replacement Player) are all well-established sabermetric measures and cover everything discussed in the story by Larry Mahnken (of the Replacement Level Yankees Weblog). There was no hand-waving or mysterious methods involved.
    The individual metrics showed that: a) despite having very similar overall stats, Ortiz’s had a much greater impact on producing wins for the Sox, b) ARod’s defense was not worth as much as some think this year (another example of this would be his Zone Rating, which dipped significantly from last year (.786 to .735)), and c) the difference in replacement level between a 3B and a DH was not nearly enough to outweight Ortiz’ “clutch” numbers.
    As far as your +.500 opponents example goes, ARod’s numbers are largely driven by his outstanding 3 HR/10 RBI game against the Halos on April 26th. And if you think the level of competition is significant, then what does it say about ARod that he had a .710 OPS against the Royals – the worst team in the AL?

  3. Benjamin Kabak says:

    BoSoxBob: Seriously, there’s no need to get so up-in-arms about your argument over here. I’m fully aware of your methods and I fully understand the stats you cherrypicked to tell you story. Insulting other writers with condescending comments won’t win you any converts. Furthermore, my post didn’t dispute your findings; rather, it looked at the popular “close and late” argument making the rounds.
    But while we’re on topic, here’s what I have to ask you:
    1. Since when has determining the MVP given out by the BBWAA been about WPA, Win Shares and any equivalent stats? If you want to talk about those stats in the context of the awards, do what David Pinto has been urging us to do for a few weeks and start some sort of Internet baseball award thing that’ll rely on stats and gain traction with the statistics community. Otherwise, you’re making your judgment on a method YOU see as best, just as the writers are deciding on a method THEY (and sometimes only they) see as best.
    2. My point isn’t that you’re wrong and I’m right. If you read closely enough, my point is that you can make a compelling case for either of these two players as MVPs. It just depends on how you pick stats. I’ve seen on Musings other readers debate your methodology. There’s no need to rehash arguments here. Sometimes, MVP just doesn’t boil down to pure statistics. There is more to game than that.
    3. I never came out on a side in this post even though you automatically assume that I’m supporting A-Rod. I never said that he deserved the MVP over Ortiz.
    4.To look at A-Rod’s stats against the Royals and criticize them is absurd. You know as well as anyone else that 20 at-bats does not constitute a big enough sample size to draw any conclusions whatsoever.

  4. Marsha says:

    Forget all the arguing and stats interpretations and give the MVP to Mariano Rivera. Mo was robbed in the Cy Young voting so the voters had a chance to wake up and repair the damage.

  5. BosoxBob says:

    Insulting other writers with condescending comments won’t win you any converts.
    I didn’t think I was being insulting or condescending with my comments. If so, it certainly wasn’t intended. And while I realize that you presented a very balanced argument, I did take exception to your use of the term “convoluted”. It implies that I was engaging in some sort of subterfuge – hiding a fatal flaw in my methods by throwing around a lot of meaningless terms. I tried to write that story in as straightforward a manner as possible. I skipped any description of Win Shares, given that it took Bill James a whole book to explain his methods. And I didn’t get into detail about replacement level (although I admit that I could have done a better job explaining things there).
    I also don’t understand why you’re now saying that I cherrypicked my stats. The WPA calculation was done using every single plate appearance for both players, and also included stolen base attempts. And both the fielding Win Shares and the Equivalent Runs stats cover the whole season as well. I chose not to use a WPA approach for fielding, partly due to the problem in determining how to divide WPA between the fielders and the pitcher, but also partly because I didn’t want to be accused of cherrypicking when I brought up the negative WPA for ARod would get from his fatal 9th inning error against the Sox on 4/6 (leading to 4 unearned runs) and his error in the 4/11 Sox game (causing another 3 unearned runs).
    Now, to respond to your specific points:
    1. I’m not going to kid myself and think that the BBWAA voters are ever going to look much beyond the usual stats. And having Internet-based MVP from the sabermetric/statistical community is a great idea (we could call it the Al Gore Award 🙂 ). BTW, I found it ironic reading your “Since when has determining the MVP given out by the BBWAA been about…” line after hearing so many Yankee fans object to a similar line of reasoning from Ortiz supporters, who said that the BBWAA hasn’t considered defense in their voting in the past.
    2. “Sometimes, MVP just doesn’t boil down to pure statistics.”
    The voters pretty much decided based on 3 things: 1) ARod’s better overall stats, 2) Ortiz’ many clutch performances, and 3) ARod’s defense. All I attempted to do was to take those somewhat subjective perceptions and quantify them using some available metrics. I’m not advocating that the voters shouldn’t judge the players based on what they’ve seen, only that they look at a way of evaluating their perceptions using a common scale (wins added).
    3. Agreed. BTW, I agree with most of your critcism of the Celizic piece. I will say though that his mention of Jeter is not that far-fetched, given that the ballot includes this as one of the criteria: “General character, disposition, loyalty and effort.”
    4. Yes, ARod’s OPS against the Royals is pretty meaningless. But I was just trying to point out that cherrypicking stats has two sides to it.
    I think the bottom line here is that we had a case of two players having exceptional (even historic) years, but trying to compare them was like comparing apples and oranges. In almost any other year, either player would have been the runaway winner. But only one of them could win it this year.

  6. Benjamin Kabak says:

    I apologize, then, Bob, for my use of convoluted. I think I picked the wrong word. And for the condescenion mark. I understand now where you were coming from.
    I guess when I say cherrypick, I don’t mean in that you intentionally picked stats that showed Ortiz in a better light than A-Rod. I was thinking more in terms of picking numbers that will be obscure to a good number of fans. I think you present a compelling argument for Ortiz.
    In the end, having seen both players play for the entire season, I couldn’t award either one of them with the MVP. If the game was on the line in the late innings, I would have wanted David Ortiz up. But in other cases — particularly when playing the Red Sox this year — I would have wanted A-Rod up. I wish they had just given the two of them co-MVPs. It’s nearly impossible to say that one outperformed the other, even in light of your analysis.
    Now that we’ve cleared the air a bit, let’s get the ball rolling on that stat-based MVP award for 2006! If we get enough people on board, we could have a really impressive vote.

  7. Dan says:

    My one gripe with WPA is what makes it tick. It’s an interesting idea, but it’s not really the panacea to this argument. I’ve seen studies (sorry it has been a while and I’m not sure on the exact reference) saying that being “clutch” doesn’t really exist, and it’s more of a matter of luck than the skill of hitting better when the odds are against you. The problem with WPA is that it does reward players for hitting HRs, etc, late in the game when the score in close, but that doesn’t measure a player’s’ skill. Is it fair to give Ortiz the MVP because he hit his HRs in the clutch, which doesn’t show skill but rather luck? Personally I don’t think so.
    Just my two cents anyways.

  8. Kenny says:

    Re #7 – Dan, I generally agree with your position, and don’t like giving players credit for performing exceedingly better in the clutch than in other situations when it’s been shown that to the extent clutch hitting exists, it’s extremely small and smaller than random variations. However, I think this is actually the point where the distinction between MVP (“valuable”) and best player might be made — even though Ortiz was not the best player, and even though you cannot attribute his clutch performance to a skill, it still happened that in 2005 Ortiz happened to hit in the clutch, and that happened to provide the Red Sox with value, probably more value than if his hits were timed otherwise. Going forward his great clutch performance doesn’t have predictive value, so when evaluating how intrinsically good as a player he is I would not consider his clutch performance. But from the standpoint of MVP, the argument can be made. However, my preference is still to go with the best overall player despite everything I’ve written to this point – I just wanted to point out that an argument can be made the other way.

  9. BosoxBob says:

    I’d say Kenny just summed up my opinion perfectly. Although ARod is probably the best player in the AL, I think Ortiz was more valuable in helping the Sox win games this year. And as far as clutch hitting is concerned, I’m not convinced that there is no such thing. The studies I’ve seen have looked at season-to-season variations, and have observed that players who have a “clutch” year like Ortiz did usually revert back to the norm the next year. I don’t think that precludes a player having periods during a season when he might actually rise to the occasion in pressure situations. After all, players have hot streaks and cold streaks during a season.
    The thing is, we’re talking about people, not computers. There is a psychological aspect to this that stats can’t capture. We all have our moments at work where we are at our peak. But that level is impossible to sustain over any length of time. So I think it could be with “clutch” hitting – a short-term affect that levels out in the long-term.

  10. John says:

    Way Whay A-Rod Why????

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: