THE BOOK--Playing The Percentages In Baseball

If there is jealousy, it’s from the PHD crowd who wish they could come up with ideas that actually make sense.  Not to disparage those who post here who have PHDs.

Some guys have great ideas and also have credentials.  In that case, it is the idea that counts.  For those without ideas, they have to fall back on their credentials and that’s all they talk about.  They are the ones who try and create an us vs them barrier.

Nothing to say, just posting this to be notified of future comments.

I think Phil does a good job of discussing the views in the article.  My personal opinion of the open source commenting is (despite what others here might think) closer to Phil’s than an old guard academic’s.

I disagree with regard to the ‘recognition’ comment in the paper (and was confused about it when I first read the paper).  In fact, I see the very opposite happen in academia: theories and statistics being named after their creators (by their own doing or not).  We have a Gaussian distribution; the Welch test; Two-Stage Heckman; Wilcoxon Rank Sum Test; I could probably go on forever.  The acronymous statistics I see in (as you say, Saberia) are relatively far from this practice we see so often in academic statistics. 

As for the ‘jealousy’ comment, I think it was made for lack of a better term, rather than with direct criticism.  Sabermetricians of course want their work to be recognized as a legitimate contribution (and, often, it should be).  The fact that those who consider themselves the authority (whether or not this is true) may not accept them is frustrating at the very least, I’m sure.  But I think calling this fact ‘jealousy’ is, in general, fairly imprecise.  There are certainly tensions I’ve observed, but the use of jealousy in this respect may be a result of some reliance on the views of Bradbury and Berri’s interactions.

One thing I do want to highlight (which Phil does nicely) is that despite the view that academics are so resistant to change--often true--plenty of intelligent academics feel that open source information are a great way to convey information.  I think Dr. Fort portrays this view clearly in his introduction paper.  I have said many times that I find plenty of value in the very interesting things people do online.

The big question is how to go about compiling the information so widely spread across the internet.  Not to necessarily institutionalize it in a peer review process, but to give proper access and direction toward the original writings and quality work that goes on.  Sifting through so much information has its drawbacks and finding those posts/essays/articles that fully lay out the proper methodologies is not always an easy task.  I think this has been discussed here before, but having a central hub (less for control, and more for organization and identification of quality/reviews) for all of this information is important in pushing forth and making the great information from such a diverse crowd available and accessible/fully reviewable for all interested.

Sorry, one more comment.  I believe that Fort strongly supports his own views on open access with his website: http://www.rodneyfort.com.

Academic articles are not cheap, and are often kept in university vaults unless you want to spend on them (a complaint I’ve heard, and sympatize with).  However, Dr. Fort’s site not only includes a large amount of data, but also every single paper he’s written (as well as a reference list of just about ever single paper he can think of in Sports Economics).  While he can only provide his own papers, the list has proven to be a valuable tool for me, and the open access on his site is a good start for anyone interested in seeing where Sports Economics is going vs. Sabermetrics.

Personally, I think the whole “credit” issue is a red herring.  I suppose it’s true that academics have at times failed to properly cite or acknowledge sabermetric work. But to me that’s a minor concern—and frankly, I haven’t always agreed with Phil when he has raised it (for example his criticisms of Bradbury’s DIPS article).  Our objection to most of the sports analysis done by economists is not about citations, its about the poor quality of the work.  The problem isn’t the lack of footnotes, it’s the lack of accuracy and insight.  Most of this work gets important things wrong, and usually the best it can achieve is “re-discovering” something saberists already knew.

Economists have told us:
1) baseball players peak at 29-30 and age slowly thereafter;
2) Football and baseball teams have no ability to identify young players’ talent in drafts;
3) OBP was radically undervalued before “Moneyball” then suddenly became much more highly compensated;
4) Pitchers throw too many fastballs;
5) The NL had fewer HBP because pitchers feared retaliation when they batted;
6) Minutes played in the NBA has essentially no correlation with players’ actual contributions to winning;
7) HR rates have been consistent across baseball history, and Sosa, Bonds and McGwire were just rare talents who happened by chance to break HR records in the same decade;
and on and on.  But ALL of these claims are either flat out wrong or completely unsubstantiated.  Tango and Phil could add many more examples.  I’m not cherry-picking a few obscure, poorly-done studies—I’m talking about major papers in refereed journals.  And it rarely adds to our knowledge, and in many cases actually misinforms people. 

Casting the conflict between academics and amateurs as largely an issue of hurt feelings over lack of credit allows economists to skirt this much more serious complaint of saberists. The fact is that “peer review” today in sports economics serves mainly not to elevate expertise or provide quality control, but quite the reverse: it works to protect academics from having to confront criticisms from much more knowledgable SMEs.  It’s anti-meritocratic.  But I don’t believe it’s sustainable.  Eventually, some young, ambitious economists will realize that amateurs on the Internet have essentially written their journal articles for them, and they will use those insights to write devastating critiques of the current generation.

Well…

The idea matters, no doubt. But clearly there are nuances and levels of analysis that a trained ~whatever~ is more facile with than a typical saberist. This doesn’t make their ideas better, and subject matter expertise trumps most of this advantage, but I know I can’t properly rebut what I don’t fully understand, nor can I incorporate feedback that goes over my head.

What I like best about the saber community is that at heart we are all student/teachers, since we are all constantly learning. Pointing out a possible issue with an analysis and then declaring that your feedback wasn’t properly accounted for is very different from helping someone fix the issue or better understand how to fix it.

In short, perhaps some/most/a plurality/enough saberists are too slow to acknowledge their limitations (of their current technical ability).

I just wonder if we genuinely asked for help if we would get it.

Guy,

On #7, I assume you’re talking about the “HR and Law of Genius” paper by DeVany?  I believe work on rebutting the claim about constant HR rates has also been done by Dr. Winfree, an Economist working just down the hall from me, and John DiNardo, a well-known econometrician.  (working version of the paper here: http://www-personal.umich.edu/~jdinardo/lawsofgenius.pdf)

I think the last sentence in your post is pretty spot on (with the exception of “writing it for them"), and while I’m no economist, my discussions with people online (including yourself) have led me to some interesting research projects that do dispute some strange conclusions I’ve seen made in academic literature.

I do agree that Phil’s portrayal of the DIPS theory that Bradbury mentions is overarching.  Bradbury does confirm the originator of the theory, and listing every study he could google isn’t an efficient way to go about explaining it.  I find it true that some citations have been lacking (but could easily be a product of not being aware of other work, which is similarly the case with some of the sabermetric work that I read citing academic work).

There absolutely are plenty of examples poor quality of work in academia, as there are on the internet.  The idea that the ‘credit’ issue is to skirt criticism is something I disagree with, but I think arises out of the lack of understanding how a saberist has reached his or her conclusion--ultimately, a result of the difficulty following from the inception of a theory or idea (say, DIPS), to the saberist writing something new and assuming it’s general knowledge.  An economist is faced with significant criticism every day, which skirting around will do no good.  Choosing which criticism to pay attention to does happen, given the incentives of an academic, but I don’t think the ‘credit’ issue is purposefully used to put up a cover here.

Does that mean Tango should explain the basis of everything he’s ever done in every article and make it 40 pages?  Of course not, but supplying the paper trail to the new conclusions is also important.  Tango does this fairly often, which I like, but older and older links are difficult to search.  That is improving.

My perception is that what’s happening is more than just an issue of “credit” ... the problem is not a lack of a footnote or two.  It’s the implication among (certain) academics that non-academic sabermetric research doesn’t exist, or that, as a body of work, it should be (almost) completely dismissed because it wasn’t academically peer reviewed.

Guy’s point that a lot of academic research is wrong or unsubstantiated ... well, that’s related.  A major reason that erroneous conclusions can get published is that the authors can get away without deferring to prior sabermetric work.  If academic authors begin to accept that non-academics are responsible for most of the progress in sabermetrics, what would happen?  Perhaps members of the community would be seen as legitimate referees of the sabermetric aspects of sports economics papers, and this stuff would get caught before going to print.

Or perhaps not.  But, in any case, we wouldn’t be urinating into the wind when we criticize some of these papers.  If academic authors take Tango’s work seriously, then, by extension, academic *readers* will start to take Tango’s work, and opinion, seriously, and progress will be made in getting some of this substandard research caught in time.

The excuse that we’re not peer reviewed is just that, an excuse, and one that allows authors to virtually ignore any previous studies that pertain to their work.  Guy says, correctly, that many academic studies just re-discover what saberists already knew.  But, won’t it be obvious that they’re reinventing the wheel if they have to *cite* the wheel?

In any case, it’s not a matter of which researcher gets the credit, or which community gets the credit.  It’s a matter of which RESEARCH gets the credit.  The idea is that you cite the best and most relevant studies to your research, because that’s how the reader can best evaluate your claims.  My problem with the lack of citation isn’t so much that it’s an offense against the rebuffed author—it’s that it’s an offense against the field of sabermetrics, and an offense against the reader.

In the case of Bradbury/Berri and DIPS ... my problem is that they didn’t cite EVEN ONE study evaluating DIPS—just the original Voros post.  I mean, come on.  Is that really appropriate, to ignore absolutely everything that’s been done on the topic in the past 10 years?  That’s more than just a matter of hurt feelings.  It’s more than just a nitpicky failure to put in a footnote.  What it is, really, is a dismissal of an entire body of knowledge.

I agree with just about everything there, Phil.  No qualms, I just didn’t think Bradbury portrayed the DIPS theory purposefully to ignore the others (but I could be wrong).

I do hope that the participation of those in academia (like Rod Fort or any others) on your site, this site, or any other site show that much of the work is taken seriously.  I think the discussion by these people particularly signals that it’s catching the attention of academic groups interested in sports. 

I know that plenty of academics feel that the saber community is fantastic at measurement of performance, and I would encourage anyone I know to read and understand these in order to use them for academic economic or management research and theory.  Even in disagreement, the discussion can shed light on shortcomings from both sides.

Considering the open source, I’m not sure there’s any defense to use BA or OBP for any research of contribution of a player to a team, for example, in labor economics.  Most of these metrics are just as easy to calculate, or have someone who has already done that for you.  I agree with many here that this should be the wave of the future with respect to sports analytics in academia, beginning with comprehensive academic reviews of sabermetric type measurement techniques.  Without this, I think much of the academic community will end up ignoring it.

One change I do see coming is upcoming students in academic fields paying more attention to this stuff due to growing up in the information age...assuming they’re truly interested in sport research.  I imagine that the academic research in sport will have its own niche of students of the game (and sabermetrics) as well as economics, management, statistics, etc. in an academic sense.  I try my best to do so, even if I get grumpy at times.

or **EVEN** OBP **ITSELF**

As for Guy’s point that “it’s about the poor quality of the work, the lack of accuracy and insight” ... well, honestly, I don’t have that big a problem with that.  Poor quality work is something that’s pretty common throughout academia, and I’m sure a lot of it is filtered out by the peer review process. 

Poor work has to be expected, inside academia or out.  Part of our “job,” as colleagues, is to be critical when necessary.  Us bloggers complaining about poor research is like a proofreader complaining about typos. 

What we SHOULD complain about is when the author refuses to correct his typos, when the proofreading process repeatedly doesn’t find the typos, when authors refuse to acknowledge that the word in question is spelled differently outside of academia, and when the authors insist that we can’t possibly know anything about spelling because we haven’t studied spelling techniques for many years. 

When wrong conclusions get published and popularized, it’s not because academics are dumb.  It’s because the process that’s supposed to weed out the bad research is faulty.  And why is it faulty?  Because peer review isn’t working.  And why doesn’t peer review work?  Because the academic community doesn’t recognize the sabermetric community as subject matter experts.

The problem is that the machinery of academia simply does not understand or care that bad research is getting published.  It doesn’t understand or care that good research is getting ignored.  That makes us look bad—and, more importantly, it makes the *good* researchers look bad.  Who would trust any sports economist, when just this year, one (DeVany) got an article published that’s not only completely and obviously wrong, but was convincingly refuted three years ago (see Millsy’s link)?  What does it do for the reputation of sports economists when other authors argue points that can be refuted in one sentence, but get published and promoted anyway?

>I just didn’t think Bradbury portrayed the DIPS theory purposefully to ignore the others (but I could be wrong).

Do you mean that you don’t think Bradbury was aware of any of the previous studies on DIPS (like, say, Tippett’s)?

Phil/12, presumably Bradbury was aware of all the research listed here:
http://www.futilityinfielder.com/dips04.html
since he included a reference to it in his article on DIPS published at THT:
http://www.hardballtimes.com/main/article/another-look-at-dips1/

That list of DIPS research includes the famous studies by Tippett and MGL.

I did a quick Google search, and the first thing that popped up was this:

http://sabermetricresearch.blogspot.com/2006/10/chopped-liver.html

Bradbury is well aware of sabermetric DIPS research.

I was both impressed and disappointed with my involvement with Sakes/Hauer.  There were quick and forthcoming about the data, but very disinterested in discussing the obvious shortcomings. 

- To them, they were true to their data source, and, yeah, they were, but, that was wrong (because of backloaded deals). 

- And, they called a “free agent” year anyone who signed their last contract as a free agent, even if it was many years ago.  So, when they find that “free agents” had their OBP corrected properly in one offseason, well, their definition of free agent bears no resemblance to everyone else’s definition of free agents.  How can a market correct itself after one year on contracts that were ALREADY signed two+ years ago?

- And it costs more to pay for an infielder’s (2b,ss,3b) hitting run than an outfielder’s hitting run, all other things equal.  Well, they didn’t find that to be the case in some years, which can only mean: all other things were NOT equal.  It is impossible for an OF to be paid more for the same batting run.

***

As for Fort, I had one exchange with him on JC’s site, and neither of us grew on the other.

http://www.sabernomics.com/sabernomics/index.php/2008/12/im-not-a-sabermetrician/

#
tangotiger says:
December 19, 2008 at 6:07 pm

“And, learned first-hand in my interactions with them, SABRmetricians simply are not interested in the MR part.  So, no surprise, whenever they do bring it up, they do it in ways that bother economists.”

Rodney: I don’t think that your experience with sabermetricians is broad enough to make this statement about sabermetricians.  There are some who apply MR, by service class, in establishing valuation, including applying discount rates.

The safer thing to say is that both sides can learn more from each other. But, there is already tremendous overlap.  You simply haven’t seen it.

#
Rodney Fort says:
December 21, 2008 at 3:50 pm

I don’t know who you are, tangotiger, but I’ll be the judge of the breadth of my experience thank you.

JC moderated out my response to Rodney, in which I made a one-sentence response like: “Rodney, thanks for making my point.”

I think that may have been the last time I posted on JC’s site as well.

To me, the bold part was a reasonable compromise position.  When I extend a reasonable compromise position to someone I’m arguing with, and that person doesn’t take it, I know I’m dealing with someone unreasonable.

Phil/12,

No, I’m sure he’s aware of it.  JC, from what I know, is fairly well-read within the saber research.  I was just implying that it was more of an efficiency thing, rather than a purposeful “F-U” type thing.  But like I said, I could be wrong.

DeVany actually isn’t a Sports Economist (not even close), and I think it’s important to separate those who are (Fort, Winfree, Zimbalist, Sauer, Humphreys, Longley, Bradbury, Berri, etc.) with those who are not (Dubner, Levitt, DeVany) when referring to expertise in the matter.  There are some who make it a point to understand the game, and others who simply want to find some easy data and say ‘look what I can do’.  Though, it’s a good example of the latter, I think.

Millsy/16: Well, maybe you’re right, maybe I’m overreacting to JC not mentioning other research in the DIPS context.  I still think he should have described it as an established result, rather than as a “suggest[ion]” in an “essay” from Voros, but I will consider that I might be reading too much into it.

Still, even if he had just mentioned his own Hardball Times post (the one in Mike/13), and maybe added the words “and others,” I’d have been happy. 

In fairness, he did include two non-academic “peak age is 27” references in his aging paper, after four academic ones.  (In unfairness, the non-academic ones came last, even though Bill James’ 1982 study was the first chronologically.  But, still, I won’t complain too much.)

Tango/15,

I could understand your frustration with the comment, but you do leave out the point that Fort is very favorable to the possible learning from one another:

“Indeed, there is much to learn everywhere from both SABRmetricians and economists on this part of the interesting problem.  What distinguishes economists, in the main, is the rest of the story on MR.”

I’m not sure the comment to you was meant for anything other than knowing the breadth of one’s own experience better than another...not a terrible assumption.  In my first interaction with Dr. Fort, we talked about measuring performance, to which he said, “That’s what sabermetricians are for, because they’re so damn good at it.  I want to focus on the economics of the issue.”

I would agree that, in general, the majority of saber slanted things I read are toward measurement, rather than MR.  But I don’t think anyone would argue that there is no one involved in that side in the saber community (and I think this has slowly shifted toward the business/$$ side as well, given the hardly marginal improvement in (at least hitting) measurement that is out there at this point).

Millsy: his statement here:

“And, learned first-hand in my interactions with them, SABRmetricians simply are not interested in the MR part. “

Obviously shows that he doesn’t deal with enough sabermetricians to make that statement as true.  So, when I tell him his breadth of experience isn’t wide enough, he starts off by saying he doesn’t know me (which makes my point), and then he said he’ll decide if it’s enough. 

Basically, he’s dealt with 1% of saberists out there, extrapolates to all saberists because he thinks it’s a representative sample, and… well, whatever, what he said was b.s.

Phil:
You are certainly right that the economists would make many fewer errors (indeed, wouldn’t publish half of their papers at all) if they respected and consulted the relevant saber research.  Still, if they somehow managed to make important discoveries while studiously ignoring the whole body of saber work, I would still want to read it and learn from it (however irritating their posture might be).  And if their superior technical expertise paid off with new or sharper insights, we’d even subscribe to the journals.  In some sense that’s a fair description of Bill James’ status:  mostly ignoring what everyone else does, but still worth listening to because he knows so damn much and you don’t know what he may figure out next. 

But the sports economists have not taught us much about baseball.  Bradbury did a nice study (with Drinen?) showing that lineup protection is a myth.  I’m not 100% sold on the conclusions, but they make a very strong case.  After that, there’s....well, I draw a blank.  Is there anything else of any significance?  The vast majority of the work just hasn’t added anything to our body of knowledge.  (Phil is arguing that they might contribute more if they recognized the existence of this body of knowledge, and maybe that’s true.  Hard to know.)

*

Millsy:  yes, I was refering to DeVany’s article. Good to know it is being contested; sad that it got published.  I pointed out some of the problems with the paper to JC Bradbury when he was promoting it on his blog four years ago.  The exchange, which I think nicely illustrates the way many economists have responded to criticisms from “amateurs” is here:  http://www.sabernomics.com/sabernomics/index.php/2006/01/letter-to-bryan-burwell/#comments. 

Final point:  as Tango’s example shows (and my exchange with Bradbury), the problem here is not only that the academics lack subject matter expertise.  Economists should understand multi-year baseball contracts better than we do. They should be able to understand why looking only at players with long career biases an aging study.  Etc.  Much of this work features fundamental errors of research methodology and/or logic (the understanding of seasonal performances as samples of underlying skill is especially lacking).  So it’s not like we need to marry the subject matter knowledge of the saberists with the quantitative skills of the academics, and head off on our honeymoon.  The sad truth is that the amateurs are ahead on both fronts.

I wonder if Bradbury would tell John DiNardo and Jason Winfree that they didn’t understand DeVany’s article.

One thing that has bothered me about the way that (some) academics seem to treat baseball research is that many of them treat their field as the one and only field equipped to answer these problems.  Sabermetrics is an inherently multi-disciplinary endeavor that can involve contributions from (at least) mathematics, statistics, economics, physics, and computer science.  As such it bothers me when (some) academics of a particular field act like they are experts because of their degrees when a) their field might not even be the most relevant and b) other fields are also relevant.  Also, not that the degrees matter, but some of the academics seem to forget that many of the online sabermetricians might have degrees in relevant fields too.

Tom -

Actually pretty funny interaction. And the whole “tangotiger” part just reeks of disgust.

Win!

20/Guy: I’m not arguing that sports economists would “contribute more if they recognized the existence of this body of knowledge.” I’m arguing that if, as an institution, they start recognizing this body of knowledge as a first step, it might get us to the point where they don’t produce as much flawed work, or where the flawed work is caught before publication.

Sometimes, at “By the Numbers,” at the extreme, we get papers from newbies who really have no idea what sabermetrics is all about, and have come up with a weird new stat that adds and multiplies stuff randomly.

I tell them, effectively, “go learn some sabermetrics.” I’m not really thinking that they’ll then suddenly start producing brilliant stuff, but at least they might know better than to produce bad stuff.  And, you never know.  Every good sabermetrician started not knowing anything about sabermetrics.  All current young researchers learned from Bill James and Pete Palmer and Voros McCracken and Tango/MGL/Andy Dolphin to get to where they are today.

Guy/20: I never saw that thread or discussion.  That is classic JC, or likely more accurately, classic JC the blogger.  I presume that since he’s friends with Forman and Drinen that he’s a fine fellow in real life.  I’d hate to think he actually talks like that in real life.

There’s this fellow “Ask Tom” Kyte from Oracle.com .  Tom is an oracle genius, knowing everything about how to tune Oracle.  He’s rough and abrasive, but he knows his sh!t, and everyone takes his attitude because of how well he knows everything.  And, he will go in-depth with you, and would not think of cutting corners in his explanation.  If you have to be abrasive, at least be welcoming.

click my name

He simply goes the extra mile to explain things.  “Followup” is Tom.

I’m going to say something confrontational and controversial here, but I think a pursuit of the “academicization” of Sabermetrics is a bad idea.

The reason why academia works is because academics are subsidized by the universities that hire them. We have no such luxury since most teams are generally free riding at this point.

Any topic involving the future of Sabermetrics that focuses more on credit and less on money is missing the point IMO. I’ve gotten plenty of one.

Voros for the win on this one. 

And I completely agree with him, although I have made enough money from sports analysis to remodel my basement.

Voros/27: I don’t get it.  You’re saying that money will start to drive the field?  You’re saying that sabermetricians care more about money than credit? 

I’m confused.

Millsy,

I find it interesting that you refer to academics in economics as ‘Dr. so-and-so’.  But you don’t call Pizza Cutter Dr. Carleton.  What’s the difference?

@28

The one in your mom’s house, right? Where you live and work.

*sad trombone*

Hawerchuck/30,

What does that have to do with anything substantive? 

I have the utmost respect for anything Dr. Carleton/Russell/Pizza Cutter does, which I have mentioned before.  Much of it has to do with those I know personally as Dr. so and so and those who I fomerly knew only as a pseudonym.  I interchangably use ‘Dr. Winfree’ and ‘John DiNardo’ as well (likely because ‘Dr. Dinardo’ sounds like something out of a comic horror movie).  But, honestly, I’m not sure why that interests you.

Voros/27, I don’t think you’ll find as much disagreement here as you seem to expect.

None from me, anyway.

Millsy,

It’s an appeal to authority.  And it shows your bias. (I am biased against academics who work outside their area of expertise but dismiss the work of non-academic experts in that area.  So be it.)

Academic economists must be a really uptight group if you don’t work on a first-name basis.  In grad school, I never met an academic born after 1950 who would have been called “Professor so-and-so” or “Dr so-and-so” by grad students.

Josh @31 - exactly.  It’s pretty sweet now - I’ve got a bathroom, a microwave and a sink full of dishes.  Just how you’d expect.

It’s an act of habit for the most part, Hawerchuck.  I refer to JC Bradbury as JC or Bradbury interchangeably in many contexts, as well as Dr. Bradbury.  Same with Dave Berri, or anyone else with a prefix.

Announce the bias to all who care, as I surely don’t.  Certainly I sway toward my own interests.  I’m not blind to that fact.

However, if you had read any substantial portion of more my comments, you would realize I thought Phil had very interesting points about the article, point directly to a paper by an economist that is being refuted, as that researcher was working on the fringes of his area of expertise, mentioned that there isn’t much excuse to not using better metrics (designed outside academia) in academic work, that academic work is best left to open access in terms of it’s ‘philosphical’ goal of being not for profit...shall I go on?

If the point of your comments is to make me go away and not bother wasting my time, then you’re doing a fantastic job.  It’s unfortunate, too, as I think there are plenty of interesting comments here.

Phil,

I guess my point was that you’re correct in that academicization of Sabermetrics won’t work. But the reason why (no funding) is also the reason why the current “freewheeling” model is suboptimal as well.

Don’t get me wrong, the detailed discussions on the real guts of doing this kind of analysis that occurred on the old fanhome boards was amazing and taught me a bunch about this stuff. But at some point your desire to do a 100 hours of work with the best case scenario being that someone else harvests the value of that work while you get “credit"…

...well that desire starts to wane. Selfish, maybe, but it’s where I’m at right now. And I can’t be the only one who feels this way, and I think that sort of disincentive can’t help but drive down the quality and quantity of research.

So to me I guess the question is how someone can monetize at least a portion of the added value his research brings. I don’t have any answers on that, but I wish I did.

I really doubt that you’re going to go away just because I question your motives.

I would hope no one goes away.

***

Voros: try posting under a different name.  For some reason, your posts keep getting flagged.

I don’t think Millsy is going to go away.  He’s the only person around here who seems to enjoy taking Berri and Bradbury’s side when we trying to figure out what’s wrong with them.  If going up against 20 guys on his own didn’t sour him on this site, then me wondering whether he reflexively respects people based on their credentials isn’t going to do anything.

Tommy,

How’s this?

Tango/39,

I certainly do not want to go anywhere, as I enjoy reading many of the discussions.  However, I don’t find comments purposefully attacking credibility very savory, especially when I have not been here to do anything of the sort to anyone else’s opinion.

Hawerchuck/40

What side do you speak of?  I really don’t understand your distaste for me.  I’m not taking any side here in this thread, nor to I have motives besides discussing views of academic and non-academic work.  I sympathize with that of an academic crowd, which I openly admit for obvious reasons.  I find the discussion here interesting (outside of your recent comments).  I challenge you to find somewhere in this thread where I specifically take some sort of ‘side’ in common with Bradbury or Berri.  In the past, I’ve specifically stated that I am not familiar with a significant portion of Dave Berri’s work in basketball, and do not have the breadth of knowledge there to make any sort of educated opinion on the matter.

But if you want to personally attack me and my objectivity, credibility, or whatever else you seem to have a problem with, so be it.  It’s a shame, because I’m sure you are an intelligent person who has much to add to sport research.

I find it ironic that, of all people, a trained engineer claims bias against those who do work outside their respective area of expertise (based on academic degree).  That’s not to disparage any work in sports that you do--and in fact, I know you do interesting work--but I’m not sure you’re in a position to make the claim you do above...and neither are a large number of those immersed in sabermetrics. 

Many baseball opinions I hold highly include those with degrees in Public Policy, Economics, Psychology, Ecology, Engineering, Statistics and Operations Research, and those with none at all.  I really don’t care about these things, as long as I see value in something being done.  You can question that fact all you’d like, but I’m not sure what the value in that is.  There’s always the potential to use the degree as a signal, but it can certainly be a noisy one.  I reflexively respect people that deserve that respect, not for a credential on the wall.  Perhaps you and I are off on the wrong foot, but I find your comments and implications rude and unwarranted.

Voros/37,

I believe you make an excellent point.  There will always be those who want to do this type of work as a hobby, but the hours are quite daunting (especially with the enormous amounts of data mining ability needed these days), and I see much of the sabermetric advances becoming proprietary when they require more and more time.  I could be absolutely wrong on that, though.  It’s not that sabermetric interests are specifically for the $$ (I believe there was a discussion to dispell that belief here in the past), but that the time spent on these things is valuable, especially when considering the skills that many involved in the subject have.

Voros: not good. Try avoiding putting in your email address.

Yeah Millsy, please stay.

Voros, I feel you. But I’m not sure most saberists want to really embrace the alternative, which is: instead of trying to help Theo Epstien win his division, you instead should help Joe Sixpack win his fantasy league.

If everyone charged for their analysis, there would be room to build a brand with free analysis. But as it stands *everyone* is offering free analysis. I’ve published exactly zero of my stuff (which you should rightly take with a grain of salt) but the reason is deliberate. In my view data should be free, analysis you should have to pay for.

Anyway, I’m not suggesting I have all the answers. However, it seems a slam dunk to me that Tango and MGL should charge for their content.

But! they should also continue to publish their formulae and methods so everyone can keep moving the ball down the field (even if they have to pay to keep up).

Folks lamenting the dearth of monetization options should look to Silicon Valley. There is no shortage of talented, passionate individuals who develop a web app whose entire business model is to wait for Google to swoop down and buy them out. Chances are it’s not gonna happen. Same with MLB ballclubs, which are far less well capitalized.

But if you go where the market is (fantasy) there’s a shot you can make a buck or two.

Millsy,

You posted a dozen or so comments on my blog defending Berri and Bradbury’s work against my criticisms.  But you admitted that you hadn’t read the work I was referring to.  I don’t think it’s unreasonable for me to wonder what your motivation is.  I think you’ve wondered quite openly, both here and on my blog, what my motivations are.  I don’t take it personally, and I don’t think you should either.

Personally, I haven’t been impressed by Berri and Bradbury’s work, and I will admit that my first thought upon seeing any new work is that their conclusions are likely incorrect.  I am cognizant of my bias.

You wrote: “I find it ironic that, of all people, a trained engineer claims bias against those who do work outside their respective area of expertise based on academic degree.”

I didn’t write that.  Here is what I wrote:

“I am biased against academics who work outside their area of expertise but dismiss the work of non-academic experts in that area.”

As in, I am less likely to listen to someone who won’t listen to me.  That’s not a grave human failing.

Josh,

Much thanks on the fantastic injury data, btw.

Hawerchuck,

My intention there wasn’t to defend the actual work involved, but to attempt to give perspective as to why an economist may be interested.  That’s all.  I say that with complete honesty.

Fair enough on the final point.  I would understand any frustration you have with regard to mutual respect.

In grad school, I never met an academic born after 1950 who would have been called “Professor so-and-so” or “Dr so-and-so” by grad students.

How about 1946

I came across an article (linked to my name) which discusses someone else’s questions about the whole journal system and how little sense it makes. I figured I’d share it to give you a different perspective.

Alan - my advisor was born in 1941 and he insisted on being called “professor so-and-so”.  If you were a postdoc, you could call him by his first name; if you actually knew him well (my old boss did), you could call him by his middle name, which was what he went by with those he was close to…

I had the pleasure of meeting Alan last week, and when our host went to call him “Dr. Nathan” he interrupted and said “Alan, please”.

I would think it depends on the relationship - when you desire to be on an equal basis, you are fine with first names. When you see yourself in a superior position, you desire the title.

Personally, “Mr. Cartwright” just makes me feel old.

Voros/37: Agreed that it’s frustrating to do all that work for free.

But, I think that’s the way it’s going to stay.  There are too many people willing to do a lot of work for just the love of it.  I don’t know if The Hardball Times pays, but I bet if they do, it’s not very much.  But look at the quality of the stuff that comes out there ... even just the links from this blog in the last couple of weeks.

A few weeks ago, there was talk about a “baseball discount” for salaries.  It’s the same for sabermetrics, and the discount has pushed the compensation down to zero.

It’s like writing of all sorts—people do it for free, and are so thrilled to be published that they don’t care about money.  Same for music, and acting, and software: it’s so much fun to create, that people who want to make money doing it are going to have a bit of trouble breaking in.

(I have to admit, I’m a bit envious that academics get paid for doing what I’ve done for free (and, in my opinion, often better).  But that’s just the way it is, I guess, that universities will only pay you to do research if you get a Ph.D. first and get hired and publish in approved journals and call what you’re doing economics.)

There’s probably money to be made doing what you’re doing, consulting for a team on specific questions.  That’s the equivalent of developing software internally in a Fortune 500 company.  You have to do what they want, not what you want, but if you can find that rewarding, you can get paid. 

But, you know more about that than I do, at least the sabermetrics side. 

I got started in 1978 doing stats for a college summer league for $200. But because I hired the scorers, I could schedule myself. With doubleheaders and the national tourney, I think I made $2000 in 1982.

Flash forward, a few years after trying to establish myself on a national stage, get a chance to write articles for websites, provide fantasy content, raw data dumps, answer emails when teams are seeking advice, and it looks like I can pay for vacation this fall and maybe get a few things fixed around the house.

But it is still fun. Two days until I’ll be at Strasburg’s pro debut! (got a 10 ticket pack in Altoona, be there tomorrow and Sunday)