Sabermetrics have not gone too far

Pat Andriola and I accidentally wrote about the same exact thing today. Thankfully, we approached the topic in different ways, and both versions are up for your perusal

Over the last decade, baseball has undergone a major transformation. We started the decade in the heart of the steroid era, then came the Moneyball revolution, and now we’re finally seeing the effects of that revolution really shape how the game works. In that same span, sabermetrics have started to become relatively mainstream, as more and more eyes look at more and more data. Inevitably, there will always be people to take it one step further than the rest, though that’s not necessarily a bad thing.

Today at Minor League Ball, John Sickels writes about those people who have taken the study of baseball beyond even his level of interest. For those of you who don’t know, writing about baseball is John Sickels’ job. The noted prospect expert said this on facebook a few days ago:

I got sick of grad school when the things they wanted us to study (19th century Belgian weavers for example) became so granular as to become meaningless. I’m starting to get the same feeling about sabermetrics sometimes.

That’s his own personal feeling, and there isn’t anything in there that’s necessarily incorrect. The math involved in some more recent work is beyond my understanding, and really is not even intended for a large audience. Take, for example, a piece from Patriot a few years ago, in which he talks about EqA, calculus, tangent lines, and other things a random Joe Schmoe wouldn’t know a whole lot about. I could quote a portion of the text, but I wouldn’t even know where to begin.

Getting back to Sickels, I think almost everyone can agree with him to a certain extent. Not all sabermetric content is for everyone. One of the reasons the internet is so great is because of the sheer size of the thing, and the number of different resources you can go to to read about baseball. There are sites for individual teams, there are sites for entire sports leagues, and there are sites that take a whole bunch of smaller websites and put them under a single umbrella. If you’re looking for something different, there are sites that can make you laugh, sites that make you think, and sites that help you remember. That’s why, when Sickels says something like the following, I’m a little confused.

“But I’m finding that as I read the most advanced sabermetric stuff regarding major league players, my eyes glaze over and I start to get the grad school feeling again: why am I reading this? I’m not enjoying it. I want to watch a baseball game.”

Baseball writing on the internet is a meritocracy. Sabermetrics isn’t spreading because we say it is. It is spreading because there is an increasing number of fans that find it useful. There is no such thing as “required reading.” If you don’t find a particular aspect of sabermetrics useful anymore, there won’t be any negative repercussions should you choose not to read it.

This brings me to the final paragraph of the Sickels post:

So am I just entering my dotage prematurely? Or is advanced sabermetric analysis becoming so specialized that no one but physics and math majors can understand it, leaving us humanities majors behind, let alone the average fan? If that is true, what can be done about it? I don’t mean stopping research; obviously it needs to go forward. But I mean, how do we find ways to disseminate the new knowledge and make it comprehensible for the non-math folks among us? How do we integrate and explain the new knowledge?

This highlights a common misconception about sabermetrics: It’s not just one thing. There are plenty of “sabermetric” blogs out there that aren’t research- or math-heavy at all. Just look at THT Live alone. Last week, for example, we had postings on the persistence of platoon splits, data reliability issues for pitch velocity, whether or not teams should shift against Albert Pujols, and some insight into the Johnny Damon drama.

Sabermetric knowledge is not becoming too specialized. It is only the extremely specialized portion of sabermetrics that is becoming too specialized for Sickels (if that makes sense). Knowing whether to use outs or plate appearances when creating a run estimator is not essential to studying baseball, but it is an essential question for certain people to answer because of the specialized work that they do. One of the many goals of the sabermetric movement is to spread knowledge to the average fan. That’s why pieces like Dave Cameron’s Win Values series and the recent influx of sabermetric primers have been so important. Because they bring these seemingly granular advances in baseball knowledge out into the public eye, making them more transparent and understandable.

To wrap this thing up, I’d just like to reiterate that there is something out there for everyone to read and enjoy. Even if one particular thing is not your cup of tea, the advancement and proliferation of baseball knowledge is something we can all take part in and benefit from.


22 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Kinanik
14 years ago

The division of labor is limited by the extent of the market. As the market for Sabermetric analysis grows, the more diverse the content out there to be read. It’s the same with everything else – as the market for fiction increases, for every Great Novel that came out in 1900, we get a things from Twilight, Harry Potter, Twilight fan fiction, etc to Bel Canto, Kite Runner, The Road. If any one person tried to digest everything written they would find a good portion of it unpalletable, but that’s not the point – the point is that people get to choose what they want to read about and readers are overall better off. Sickels is doing the equivalent of going to the book store and grabbing ten random novels and noticing he doesn’t enjoy all of them. Well, of course.

Kinanik
14 years ago

Although now that I read his comment in the other thread, I think I misunderstood him. If he wants Sabermetrics to be better integrated into the narrative in baseball, that’s a good thing, sorry if I misunderstood!

John Sickels
14 years ago

Kinanik, that’s what I was trying to stay in the essay…how do we integrate all this new knowledge in a way that is meaningful and accessable?

Matt
14 years ago

Well of course it’s his “John Sickels’) decision whether to read the material.  But willfully disregarding it will open him up to the typical dumb-Neanderthal criticism from the sabermetric flock.

A couple years ago when I was first starting to get into sabermetrics, I emailed a couple leading bloggers to ask how OPS+ is calculated.  I know what it is, I understand the concept.  But how exactly are park effects factored in?  The responses I got back were all along the lines of “well, some ballparks are actually more conducive to scoring runs than others…”  “Thanks a lot, aholes,” I thought.  I’m a civil engineer; I’d like to believe I can understand these things, but I can’t even find the formula!

The reason that people are still clinging to BA, RBI, etc is not that they believe them to be the most accurate measure of a player’s value.  It’s because they’re so transparent.  I can find them for my local minor league or even amateur team without splitting the field into 78 zones or assigning a leverage index.  (Don’t get me wrong, I love the idea of UZR and WPA.  But the numbers could have been pulled out of a hat by a medicine man for all I know.)

It’s great if you believe that nothing is “required reading” and that some of these things may be too much for some fans. But then don’t expect them to ever become “mainstream.”

Sabermetrics in 2010 seems strikingly similar to to investing in 2008.  So many people are so confident in their approach, designing ever more complex prediction models, their bullishness feeding off each other’s, blinding them to their errors.  (What’s up with OPS, we know OBP should be weighted about 1.7X more than SLG)  I’m afraid you all are headed right off a cliff.

Joe R
14 years ago

2008 market crisis -> 2010 SABRmetric crisis.

Yes, these are totally similar things.

Paul Moehringer
14 years ago

I tend to agree with what Sickels is saying here.

You ask alot of these sabermetric people what makes someone like Justin Upton valuable, and you’ll get all the stats in the world thrown at you.

Ask the same person precisely what fundamentals he’s lacking at the plate, they’re usually clueless.

I have an exremely hard time taking someone trying to explain why person a is better then person b, when they themselves can’t even figure out basic hitting concepts.

I do appreciate sabermetrics in what it has brought to the game, and it has definately changed how the game has been viewed in the past 30 years, and I appreciate how it’s increased interst in the game, but it can only take you so far.

If all you can do is stats, there’s going to be alot of basic concepts that go right over your head.  But your not even going to realize it, because you’ve so thoroughly convinced yourself that sabermetrics are the be all end all of analyzing baseball, and that just simply isn’t the case, no matter how much people want it to be.

Jonathan Sher
14 years ago

Joe R – Matt didn’t say the 2008 market crisis and 2010 sabermetrics were “totally similar thing.” He said that shared some specific things in common. If you disagree you should refute his specific claims. Otherwise, your misrepresentation of his argument lends the impression lends the impression you don’t know how to refute his argument.

Please understand, I’m not suggesting Matt is correct in his contention. But at least he was specific and provided an example that can either be supported or refuted. His approach is scientific. Your response is not.

The Real Neal
14 years ago

The article is fine, but this statement is both ironic and preposterous:

“If you don’t find a particular aspect of sabermetrics useful anymore, there won’t be any negative repercussions should you choose not to read it.”

Ever heard of a website called “firejoemorgan.com”, Dan?

Matt
14 years ago

Thanks, Jonathan.

A (satirical?) example of what I’m talking about:
http://joeposnanski.com/JoeBlog/2010/02/16/baseball-predictions-in-february/#more-3092

Nick Steiner
14 years ago

John, I am curious as to the point of your article, as I don’t think you articulated it will and I’ll give you a chance to explain before I decide to get really offended by it.

Jonathan Sher
14 years ago

Thanks Matt for the fun link. Posnanski is a hoot.

Nick Steiner
14 years ago

On second read, it appears that this is what you are trying to get at. 

In the article it appeared that you were saying that the more esoteric research in baseball is not interesting to you because it doesn’t add much to your knowledge and is a pain to read through.  That’s fair enough, however, I’m not sure why you expect everyone to share those feelings. 

You use Pitch f/x as an example.  As a prospects expert, understandably, it offers limited uses for your work.  However, as an avid Pitch f/x researcher myself, I know that the insights gained from Pitch f/x research have informed many, many people and are relatively easy to inject (you definitely don’t need a physics degree to understand how much better a curveball is when followed by a high fastball http://www.hardballtimes.com/main/article/pitch-sequence-high-fastball-then-curveball/). 

So if you are using Pitch f/x work as an example of something that is “too far”, then I would say that is clearly a direct result of your niche and not of it’s esotericism.  If you did not intend that part of your article to be read that way then I am sorry, and would appreciate further clarification.

Dan Novick
14 years ago

Fire Joe Morgan made fun of stupid arguments and pointless ramblings. John Sickels makes fantastic points without the use of sabermetrics.

Jonathan Sher
14 years ago

Dan, I don’t know John Sickels personally though I do enjoy his work, but I suspect you are reacting to a point you think he made but did not. You seem to take from his article that Sickels has been turned off by sabermetrics generally. I don’t take that as his point. Consider this passage:

“But when it comes to the MOST ADVANCED sabermetric stuff” (emphasis added).

I don’t think he’s talking about the use of batted ball data to decide whether to shift for Pujols—I like that piece, by the way. He’s talking about a subset of sabermetrics, that which uses advanced math. So when you use an excerpt of his comments as an example of a common misperception, let me suggest that isn’t the case and that he views sabermetrics in its many nuances and flavors just as you do, perhaps not to the same level of detail as you since that is an area of your expertise, but a similar perspective nonetheless.

Of course I’m speculating base on what I know of John’s work—but perhaps you can engage him directly.

Bob Wall
14 years ago

I too am a former history major and a liberal arts guy.  I have been fortunate to have followed the growth of Sabermetrics from those early Bill James articles that I read to get the jump on my Strat competition.  And there are times recently when the math level does rise a bit, and the swimming gets tough.  And yet…it is just me, or does anyone else see this year as a giant field test of the new defensive metrics, the most hotly debated Sabermetrics area?  Can defense be quantified as offense and pitching have been to date?  How can this not be fascinating to the baseball statistics crowd?

CH
14 years ago

It doesn’t seem that Sickels is fed up with the entire SABR world, just the ultra-specific part of it that doesn’t appeal to his sensibilities at all.  It’s a perfectly reasonable response to advanced statistical analysis.

What I think he’s trying to say is, there are important studies being done in SABR circles, but if they’re not communicated effectively to the average or even above-average fan, they turn into cryptic conversations between a very small group of people.

Take Hawking’s “A Brief History of Time,” for example.  His editor warned him that, for every equation included in the book, his readership would be cut in half.  So, Hawking included only 1 or 2 equations in the entire work.  People still read and enjoyed the book, even though it may have still been over their heads, because most people feel comfortable attempting to understand the greater concept behind something, even if they don’t necessarily understand the math.

Would anyone suggest Hawking should stop using math altogether?  Or that the math isn’t important or doesn’t tell us anything new?  Of course not.  But if he’s trying to present his work to a larger audience, it needs to be communicated in a way that a large number of people can understand it, or at least understand the concept behind it and why it may be important or useful to them.

TCQ
14 years ago

CH, I think most people understand that if Sabermetrics is going to become more mainstream, it needs to be articulated in a way that’s clear to a layman. But the point being made(well, my point, at least) is that there are plenty of people *already doing this*. Dave Cameron, Carson Cistulli, Matt Klassen, even a lot of Tango’s work is pretty clear(Joe Posnanski was linked above, and he might actually be the best example of all. He’s definitely not on the cutting edge of stats – still uses OPS rather than wOBA, etc – but he’s about as accessible as you can get and furthers the basic principles of the sabermetrics every time he writes).

CH
14 years ago

TCQ, I don’t think anyone is trying to have a go at Fangraphs or the writers there, and I certainly didn’t mean to imply that there aren’t ANY great SABR-oriented writers.  There are many.  I apologize if it sounded like some kind of indictment of the SABR community.  I just think it’s important that advanced metrics are used in a way that they can win over the average fan while still holding the interest of the more statistically inclined.  Granted, that can sometime feel like an almost impossible balancing act, so I like to give every writer the benefit of the doubt. 

However, I think Sickels has every right to voice his concern that maybe it’s started to tip too far in one direction.  Of course people have the right to disagree with him, but the important part is that sabermetrics is supposed to be about reason defeating mindless tradition.  If sabermetrics becomes too stale or too granular it will develop its own sort of mindless tradition, where people are afraid to question the orthodox SABR view, for fear of being somehow blackballed from the community.  Every new study or new statistic has to prove itself to the baseball community as a whole, neophytes and experts alike, otherwise it’s nothing more than a purely academic exercise.

TCQ
14 years ago

Oh, crap, that list of writers was pretty FanGraphs-centric, wasn’t it?

But the main thing I was trying to say is that while yes, absolutely advanced stats have to be explained in such a way that the average fan gets them, that isn’t any reason for the more esoteric angle to be explored also. wRC+ can be explained in a series of primers while SIERA can be broken down in a way that’s way over my head – that’s a good example, actually, as I read the first part of Tango’s series and bowed out on trying to understand it, but I *love* that it’s out there, because I respect him greatly and believe that if there was an big flaw in SIERA, he’d point it out(and maybe already has, I haven’t been keeping up with it).

But I totally agree with you that there’s a danger of the sabermetric community becoming so insular that it isn’t interesting to anyone outside of said community, but there are just so many great layman-friendly primers out this offseason that it’s hard for me to fathom that that’s an imminent danger. Always have to keep an eye out, though, and Sickels certainly shouldn’t be black-balled for stating his opinion – maybe the next time someone says something similar I’ll agree.

Trey
14 years ago

“Among 929 plants, 705 bore violet-red flowers and gray-brown seed-coats; 224 had white flowers and white seed-coats, giving the proportion 3.15:1”
– Gregor Mendel

“Everything is so over-analyzed. Why don’t we just look at the rose and realize it’s a pretty flower?”
– Ernie Harwell

Matt
14 years ago

Hawking may have had a best-seller, but still no one knows what string theory is.  The one equation that he included, e=mc2, is universally familiar but not at all understood (myself included), so I’m not sure that’s the best example.

As TCQ said, there are plenty of writers who make the implications of the newest data accessible to “a layman.”  But as TCQ also said, we trust them because we respect them, we admire their work, we know they’re smarter than us in these arenas and we’re sure that they would point out their own flaws if there were any. I’m not so trusting.

One of Hawking’s main theories in his book was disproved 16 years later.  Scientists are always eager to disprove each other’s theories, knowing that the original proponent will not be offended.  I don’t sense that same willingness in the sabermetric community.  I’ll admit I’m an outsider, but it seems to me like everything that Bill James or Tango comes up with is treated like gospel based almost entirely on their reputations and past accomplishments, without much examination of the formulae involved even by those writers who so effectively translate it for us laymen.

That’s why I compared it to the economic recession.  It seems like there are thousands of us who kind-of get what’s going on, but only a tiny handful who truly understand it on the deepest level, and of whom all the rest of us are just tacit followers.

TCQ
14 years ago

“and we’re sure that they would point out their own flaws if there were any. I’m not so trusting.”

Not really what I was trying to say; I don’t think that Tango would always point out his own flaws(I don’t think he’d publish something if he saw overt flaws in it, but that’s not really the point), but I think the SABR community at large – whether that’s Bill James looking at Tango or Patriot looking at James or whatever – is *very* good at examining what’s put out there in a scientific way. But as I’ve said before, we obviously always have to keep an eye out for it becoming too insular or too dependent on the few at the top.