Vice Sports on "How Wall Street Strangled the Life out of Sabremetrics"

[Terminator]

I thought it was an interesting read so I thought I'd share.

 

https://sports.vice.com/article/how-wall-street-strangled-the-life-out-of-sabermetrics?utm_vicesportstwitter

[KingKat]

I thought it was an interesting read so I thought I'd share.

 

https://sports.vice.com/article/how-wall-street-strangled-the-life-out-of-sabermetrics?utm_vicesportstwitter

 

Something about that article doesn't quite jive but I can't put my finger on it. Maybe I need to sleep on it.

[BTS]

Something about that article doesn't quite jive but I can't put my finger on it. Maybe I need to sleep on it.

 

Yeah, I thought the same thing. Really, it was a lot of words that didn't say much of anything.

[KingKat]

Yeah, I thought the same thing. Really, it was a lot of words that didn't say much of anything.

 

I feel like he might be on to something but he just kind of hinted at an argument that he never really fleshed out. I guess what he's saying is that whatever market inefficiencies are being found right now are being closely guarded and not shared and that means that the market isn't rewarding players properly but that's ******** really. Look at framing data. It's all right there in the open and the market still hasn't adjusted so we have evidence that even when analytics are freely shared market inefficiencies are slow to be corrected. There are obviously lots of competing mentalities still at play in baseball. We also don't really have evidence that front offices are hoarding secrets the way the author suggests.

[TDotttt2005]

I believe this is a terrible article, which really is trying to be way too much. How does Wall Street have anything to do with proprietary baseball stats and openness? I guess he's attempting to make the connection between JP Morgan organizing his business in the late 1800's/Rickey and the institution of the amateur draft in the 1950's/and the fact that Sternberg and Friedman came from Goldman Sachs, but really, other than a catchy 'everyone hate on wall street' era title, its a laughable connection at best.

 

And on the lack of openness of Sabrmetrics, has he heard of the concepts of trade secrets and Competitive advantage in business? Should every team just put their whole database on the internet, transcripts of every meeting, phone call, text message (No offense Astros), in the name of openness. Should Coke give away its formula for everyone to make? This 'strangling' he mentions is what leads to progress. Entrepreneurs/Smaller companies with 'ideas' are hired/bought every day, making the founders rich, and bringing them into the fold of larger companies, who can make actual products out of simple ideas, which in turn, are used by millions of people every day to make life simpler and more productive. Once these products are eventually deemed obsolete, it incentivizes new ideas to take hold, and new products to be developed. Its the same thing with SABR. For example, take a stat like WAR. Someone invented and refined the statistic and put it out there (not sure if it was James). One team hired that person, paid him a lot of money to develop it further, and put it to use to acquire talent and win baseball games. Eventually, that competitive balance is eroded, as every team used that stat, and it spawned the need for more ideas, more stats, more potential competitive advantages, etc. Thats how we've get progress in the SABR discipline, and any discipline really.

 

I think the author is trying to be a populist, socialist, capitalist, anti-wall street, pro-labor, pro-minority, sports geek, and a journalist all in one article. What we got was this terrible mishmash of ideas.

[Laika]

The author's point seems to be that sabermetrics has been consumed by MLB teams operating like Wall Street firms because any public sphere sabermetrician who shows any sliver of actual competence or ingenuity gets gobbled up by a team almost immediately. What historically has been a collaborative outsider's hobby is now a war of secretive, proprietary discovery. Due to the great motivating power of market competition, a wonderful open source journal of baseball knowledge is now utterly and thoroughly an arms race of encrypted and guarded secrets. As such, sabermetrics as we know it is dead.

 

I would reject his thesis based on the very nature of how ingenious discovery seems to work.

 

If you've ever read one of Taleb's books you're probably at least somewhat convinced that game-changing discoveries cannot and do not come from targeted research. It is extremely difficult to cure cancer by hiring the five smartest research oncologists you can find and telling them to go "find a cure!". No, game-changing discoveries overwhelmingly tend to come from think-tank style environments - large groups of competent people collaborating for the sake of discovery with no particular goals aside from discovery and research on their own. Any actual leaps forward happen randomly from the collective goop of the think-tank environment. Think-tank style research environments for sabermetrics are stronger than ever, and they can't be harmed simply by MLB teams plucking out a few big dogs like Mike Fast and keeping their outputs under lock and key. Nobody can really know what aspect of baseball will be the subject of tomorrow's next big sabermetric thing, but what we do know is that it is far, far more likely to pop out of a Fangraphs "community research" page, or out of a blog like "Breaking Blue", than it is to pop out of the fingers of someone like Colin Wyers who sits down at his Thinkpad everyday and tries to find the next OBP level market inefficiency. Front office analytical departments aren't reinventing the wheel every day, what they're doing, in large part, likely amounts to basic application and implementation of publicly available saber knowledge, and very small tweaks on commonly available aspects of sabermetrics.

[KingKat]

The author's point seems to be that sabermetrics has been consumed by MLB teams operating like Wall Street firms because any public sphere sabermetrician who shows any sliver of actual competence or ingenuity gets gobbled up by a team almost immediately. What historically has been a collaborative outsider's hobby is now a war of secretive, proprietary discovery. Due to the great motivating power of market competition, a wonderful open source journal of baseball knowledge is now utterly and thoroughly an arms race of encrypted and guarded secrets. As such, sabermetrics as we know it is dead.

 

I would reject his thesis based on the very nature of how ingenious discovery seems to work.

 

If you've ever read one of Taleb's books you're probably at least somewhat convinced that game-changing discoveries cannot and do not come from targeted research. It is extremely difficult to cure cancer by hiring the five smartest research oncologists you can find and telling them to go "find a cure!". No, game-changing discoveries overwhelmingly tend to come from think-tank style environments - large groups of competent people collaborating for the sake of discovery with no particular goals aside from discovery and research on their own. Any actual leaps forward happen randomly from the collective goop of the think-tank environment. Think-tank style research environments for sabermetrics are stronger than ever, and they can't be harmed simply by MLB teams plucking out a few big dogs like Mike Fast and keeping their outputs under lock and key. Nobody can really know what aspect of baseball will be the subject of tomorrow's next big sabermetric thing, but what we do know is that it is far, far more likely to pop out of a Fangraphs "community research" page, or out of a blog like "Breaking Blue", than it is to pop out of the fingers of someone like Colin Wyers who sits down at his Thinkpad everyday and tries to find the next OBP level market inefficiency. Front office analytical departments aren't reinventing the wheel every day, what they're doing, in large part, likely amounts to basic application and implementation of publicly available saber knowledge, and very small tweaks on commonly available aspects of sabermetrics.

 

Reminds me of anecdotes Voros McCracken told about his time in baseball. He basically spent his time in baseball staring at a blank screen waiting for the next idea to come. You can't just give a guy money and expect him to come up with the next FIP. An ingisht like that is just something that sort of happens. You can't just pump those out on command and you're absolutely right that it's just as likely to come from an unknow than from someone who has already had their moment and has been scooped up.

[Terminator]

The author's point seems to be that sabermetrics has been consumed by MLB teams operating like Wall Street firms because any public sphere sabermetrician who shows any sliver of actual competence or ingenuity gets gobbled up by a team almost immediately. What historically has been a collaborative outsider's hobby is now a war of secretive, proprietary discovery. Due to the great motivating power of market competition, a wonderful open source journal of baseball knowledge is now utterly and thoroughly an arms race of encrypted and guarded secrets. As such, sabermetrics as we know it is dead.

 

So far you're the only one to get the point of the article.

 

He does kind of have a good point IMO but I see yours as well.

 

As the article says, there doesn't seem to be a ton of new breakout developments lately are there? A decade ago you could easily point them out. Now I don't find it to be so clear. But that's all conjecture based upon my own opinions so I could be wrong.

[Abomination]

As the article says, there doesn't seem to be a ton of new breakout developments lately are there? A decade ago you could easily point them out. Now I don't find it to be so clear. But that's all conjecture based upon my own opinions so I could be wrong.

 

It shouldn't be surprising. At some point, you (mostly) exhaust what you can do with the available stats. Then you wait until data collection improves and begin again. With MLBAM rolled out next year to all 30 teams, we should see an explosion in the next year or two.

 

Also, as each new stat improves on previous ones, the amount of improvement is generally smaller. At some point, people (at least the public) don't care about the difference in the small amount of improvement, especially if it also comes with an extremely complicated formula.

[Laika]

As the article says, there doesn't seem to be a ton of new breakout developments lately are there? A decade ago you could easily point them out. Now I don't find it to be so clear. But that's all conjecture based upon my own opinions so I could be wrong.

 

I wouldn't say that it's even true. It's just like any area of research though, future breakthroughs happen within the foundations set by past developments. It's not really possible to reinvent DIPS - but something like TIPS can be just as ingenious within the context of today as DIPS was at its inception.

 

A lot of the steps forward today are more technical, less foundational, and within sabermetric niches. But that doesn't mean that they aren't important or aren't as creative and cool as the foundational discoveries which they operate within.

 

So I would say that it's not that teams have diluted sabermetric research at all by hoarding intellectual capital, it's just the nature of research in general within any fairly new field. I mean, in some ways, you could say that there haven't been very many equally important leaps forward in genetic research since things like the discovery of the structure of DNA, the development of PCR, and the sequencing of the genome.