Wednesday, October 29, 2008

Foul Ball!

If Connie Mack never heard of it, I don't want to hear about it.

Lee Huang pointed me to an op-ed in the New York Times coauthored by unlikely colleagues Billy Beane, Newt Gingrich, and John Kerry. They advocate the creation of a federal institute for "evidence-based" medicine, as opposed to what they call the "experience-based" approach that most of our care is based on now. Basically, that means introducing a lot of hard statistical analysis into practice.

That may well be a fantastic idea,* but I reject the analogy with which they drive their argument: sabermetrics.**

As their story goes, sabermetrics is transforming baseball, propelling to success teams like the Oakland A's of earlier this decade and the Tampa Bay Rays of today. Furthermore, dinosaurs clinging to the old ways, like the Yankees, Mets, and Red Sox, are sitting at home.

I dispute every step of this narrative. Yes, the A's had a good run from 2000 to 2003, but they never even made it to the World Series. And where are they today? They've finished ten games under .500 the past two seasons.

And the Rays? It's true they're in the World Series with baseball's second lowest payroll. But they haven't done so predominantly by getting hidden gems for a steal off of the trade market. Rather, finishing in last place every year continually presented them with favorable positions in the draft. This standing definitely required deft choices in said drafts, which many other teams failed to do, but they certainly had a leg up by being awful year after year.

In short, if the A's' success earlier in the decade and the Rays' success this year is evidence in favor of sabermetrics, why aren't these teams' failures since and before evidence against it?

Now, how about the "dinosaurs" (my word, not theirs). The article implies that the Yankees, Red Sox, Mets, et. al have gotten where they are by following the "old ways." Quite the opposite. My understanding is that the strategy of building a team out of the most expensive free agents year after year is a relatively new one, pioneered by the Yankees in the 90's.

Again, it's true the hazards of such a strategy have been made manifest, in no team more clearly than the Mets. The result is a team full of players past their prime who all want to be stars - in short, a whole equal to less than the sum of its parts. But these hazards are by no means the result of following tradition and intuition.

If you want to look at a truly traditional approach to baseball, look no farther than Charlie Manuel and Pat Gillick, the Phillies' manager and general manager, respectfully. They have 135 years between them. Charlie's following of his gut frustrated Philly fans for years until they realized our team was winning. And Gillick has done the same miracle work here as he did in the early 90's with the (cursed) Toronto Blue Jays and later on with the Seattle Mariners.

And we're about to win the World Series.

So, I cry foul on Beane, Gingrich, and Kerry. They spun a story about baseball to make an analogy with healthcare seem pleasing, when in fact it is a false one. Not surprising, coming from politicians, but not true.

Now, the fact that the analogy is a false one may be irrelevant. But to the extent that it rings true, I would be very cautious about undervaluing a doctor's intuition.

*It probably says something about me that I was far more incensed about the baseball aspect of this article than the healthcare aspect. Sorry, Health and Societies friends.
**Sabermetrics refers to the statistical approach taken by the Oakland A's (under Beane) late in the 90's. They evaluate players using a lot of new-fangled statistics in addition to traditional indicators like batting average, ERA, and intuition.
*** Though unrelated to baseball, I find hard to swallow their statement that the United States has worse health quality than most industrialized nations. There's no question we need to do a lot better, but my (flawed) understanding is that if you're insured here (a big if, I know), you're a whole lot better off than in most other places.


LH said...

Questioning Billy Beane is blasphemy. There, I said it. (What can I say, I own a T-shirt that says "In Billy we trust" on the back.)

OK, all kidding aside, you are somewhat correct when it comes to your assessment of sabermetrics, although Beane et al would probably agree that it has its limits:

* Playoffs are too short a sample size, so throw sabermetrics out the window in a 5- or 7-game series.

* Money still matters, whether it means hiring more scouts or buying the best talent.

* Individual stats don't factor in that ever elusive intangible called "chemistry," whether that means teammates getting along, managers getting the most out of their players, or Jamie Moyer offering invaluable mentoring services to Cole Hamels.

However, the core of the Billy Beane approach is still sound - the use of data to determine which skills are undervalued relative to the market, and then to coolly buy low and sell high while others buy high and sell low:

* The ability to draw a walk used to be undervalued, so Beane loaded up on hitters with good eyes. (He almost got a minor leaguer named Kevin Youkilis for free because, when no one wanted him, Beane coveted his ability to draw walks.) Once the market caught up to that fact, stats told him defensive efficiency was where it was at, and he's drafted accordingly.

* Beane will never draft high-potential high school pitchers, who almost never pan out but are universally lusted after (goodbye, Todd Van Poppel!); he will always draft high-performance college pitchers, who often pan out but are often panned because scouts fear they have accumulated too much mileage (hello, Joe Blanton!).

* Closers are always overvalued, so Beane installs a no-name pitcher, knowing that anybody can earn 30+ saves if given the chance, and then trades him after he's had success. (See Koch, Billy; and Foulke, Keith.)

I grant you that missing the playoffs four out of the last five years is not a good track record. But an average of 90 wins a season over a decade is nothing to sneeze at. And, says this lifetime A's fan, there's always next year!

l e i g h c i a said...

In relation to your health care footnote:
- I agree that you can receive good care in this country IF you have health insurance. And probably alot faster than you would in some other countries, that are known for longer waits for medical treatment.
- However, overall we fare poorly on health care because there are so many people who are uninsured. Any system's quality should probably considered in light of how they treat the "lowest" and "poorest" members of society, rather than the richest.
- The fact that insurance is so complicated also drastically increases the cost of care-- America spends twice on admin costs relating to healthcare than many other countries. Much of that is related to insurance and billing.
- Anyways, I'm Canadian so I'm rather biased on the topic of health care :)

Zachary said...
This comment has been removed by the author.
Zachary said...

Okay, so I actually read the article now. :)

-Nurse Zach here... I completely agree with what these three guys are saying (despite the fact that my thesis gives a lot of attention to embodied knowledge and the centrality of intuition in expert clinical practice) because they are completely right about how our healthcare stacks up to other industrialized countries, i.e. shockingly poorly. The sad truth is that we spend more money on healthcare than anyone else and get worse outcomes than most. The socialists currently have us beat.