Sabermetrics is a term that some deep geeking, baseball stat freaks use. It involves the study of some types of statistics that are different than the traditional batting average, earned run average etc. The proponents of sabermetrics have come to some conclusions that have changed the game.
Billy Beane has had a good degree of success as the G.M. of the Oakland A’s using sabermetric evaluations. Now, he has teamed with Newt Gingrich and John Kerry on this interesting N.Y. Times op-ed about why we should use a sabermetric type of approach to evaluate health care.
America’s health care system behaves like a hidebound, tradition-based ball club that chases after aging sluggers and plays by the old rules: we pay too much and get too little in return. To deliver better health care, we should learn from the successful teams that have adopted baseball’s new evidence-based methods. The best way to start improving quality and lowering costs is to study the stats.
Speaking as someone who has been there and run the MANOVA, once someone learns to datamine, you can pretty much put any spin on the numbers you choose. While with stats folks can check your numbers to see if you lying, the design of the research and choice of test will decide whether you find what you’re looking for or not.
My first thought was that doing sabermetrics on health care is exactly the kind of idea that Newt Gingrich would go for.
I’m not familiar with MANOVA, J. I started reading some of Bill James’ sabermetric based stuff about 20 years ago, I think. When, I started looking at this, as a baseball fan, a lot of what these guys were saying bugged me because they were debunking myths. For instance, the “clutch” performer. There are guys that had a reputation for being able to elevate their performance in big time situations. Sabermetric guys delight in proving that “X factor”, that intangible to be a load of crap. A lot of old school baseball guys ignored it. Billy, used this data when the money was on the line. He built teams that won based on data. No hunches. If we are going to pay this guy X amount off $ over a given period, there needs to be data to support that decision. So, I don’t see this as a case of manipulating numbers for spin. It’s more about, how can we take this data and find real long term tendencies?
I still say that any statistical results depend on who you have spinning the numbers. I had a lot of stats classes and learned enough to become jaded about the whole process once my study was finished. My dissertation was built on a factor analysis, where you have variance spinning around in factor-space, then you look for relationships and try to explain how the variables are related, so I know a bit about the spin. I hear what you’re saying about the “x-factor” and baseball, though; seems like they were betting everything on a guy’s tendency to perform outside his norm. Stats become interesting when people use them to judge song quality against the “hits”. (example = http://www.faqs.org/patents/app/20080228744 )
Can I get a witness? Does it not seem like our world-class USA leaders are apt to have more and more time to hang out leisurely and come up with spurious, yet catchy “variance” rubrics to sell to the G Q Publick?
Kerry, Gore, Clinton, Bill and Hill’ry, all standing around playing stink finger, then looking at a baseball application to “fix empirically” our Healthcare system, is a strange Leisuretronic Industry all in its own right. Come on help me out here.
I don’t really think the Healthcare situation is all that broken to begin with; and I am certain it is not a federal case, rather a state or local task… I still respect Jimmy Carter for getting out and swinging his hammer, unlike others swinging their twenty year hence aged organizational development wet dreams.
Dr J has a point: as in, “any song can be started on any note…” no?
I’ll second that.
I’m coming around to J’s point. I still think there’s some validity to the concept, but I can see some of the cracks.
I used to work in health care as a sysadmin. It is unlikely that this kind of thing would ever be developed in-house. This would either be developed by your EMR supplier, or at least with their help as that would be the interface. And they’re as much as part of the medical economy as the drug companies.
One of the things I did was maintain a web application. A scheduling application. Each building — each building where doctors worked, which didn’t include my building, which sucked — would have a drug company come in an provide lunch. Here’s the napkin and plates, the plastic forks and spoons, the pasta salad, the entree, the stack of pens and folders extolling the virtues of their .
Going from my limited understanding of sabermetrics, it’d be good for proving that an experienced player isn’t as good as his rep by looking at his major-league stats, but you only get data on the rookie by checking the minor league and high school stats. A new drug won’t have that data, so the system might encourage established drugs. Assuming that the programmers don’t fudge numbers for the company who bought them lunch.
It’s all about the Benjamins… ($100 dollar bills). Everything in my doctor’s office has the name of some drug company on it, including the facial tissues. I’d bet the drug companies would be happy to develop any software required to cut health care costs, especially if there was a prescription recommendation module.
Link = http://www.youtube.com/watch?v=WgoB2h_Wkco
“The Drugs I Need” by Austin Lounge Lizards
Did someone say “Lunch?”
[0061]Depending on a nature of the scale used for evaluation, the present adaptivity method may be applied to evaluate a range of type of files, i.e. compression format, nature of files etc. . . . with an increased accuracy in highly non-linear fields, by providing a dynamic learning phase.
[0062]Although the present invention has been described hereinabove by way of embodiments thereof, it may be modified, without departing from the nature and teachings of the subject invention as defined in the appended claims.
“Are you happy to see me, or is that a sabermetric embodiment in your trousers?”
… thanks for the witness, Dr J, I’m all in.
clik HERE: The Damnedest Lies
The success of fivethirtyeight.com is a credit not only to statistical prowess but also to keen intuition about social habits.
Of course, there is no foolproof mathematical method for determining how these weights should be assigned. And doing so successfully is as much of an art as it is a science. Modeling real-world systems, particularly those influenced by human behavior, also presents statisticians with another problem. There are an overwhelming number of factors that influence the behavior of the system. And attempting to keep track of and account for all these influencing factors would be something akin to trying to herd a large group of cats across Manhattan at rush hour.
So inevitably statisticians must determine which factors to consider and which to disregard. This ultimately amounts to making guesses. And while these guesses are educated, they are still essentially guesses.
But in this way statistical models can be deceptive. We often conflate the objectivity of statistical theory with the uncertainty of statistical results. The British politician Benjamin Disraeli once commented that as a consequence of this confusion, “there are three kinds of lies: lies, damned lies, and statistics.” (see: retrieved HERE )