Poisoning the well

According to Google, there are roughly 22,300 web pages containing the terms “mother’s basement” and “baseball.” A classic example of the genre is this slanderous screed:

Sabermetricians. Ugh. It’s like Major League meets Revenge of the Nerds, except the nerds aren’t fun and they think they’re better than you. And Bob Uecker isn’t there to keep things interesting with the occasional witty one-liner.

I can’t say that sabermetrics as a statistical form of analysis isn’t effective, because in many cases it is (on-base percentage, for example). I just really don’t like the guys behind the numbers. Poindexters from Ivy League schools who lack the social skill to relate to other people, but can explain the value of a ground ball by drawing a diagram and involving advanced mathematics (and they say baseball is boring).

This is not of course a lone example; Fire Joe Morgan of course made a career of lampooning that kind of thing.

So why does it matter?

The logical fallacy

It’s a clear and simple case of poisoning the well. It’s a form of personal (or “ad hominem,” if you prefer) argument that attempts to discredit what a person has to say before he actually gets the chance to say it.

In a larger sense, it’s an effort to discredit sabermetrics (and thus avoid arguing with its premises) by discrediting the sabermetrician, or sabermetricians in general.

An odd form of slander, isn’t it, that attempts to make “advanced mathematics” into an insult, is it not? Heaven forfend that intelligent people seek honest answers to questions!

It’s a way to avoid asking questions, or answering them. Whenever someone starts poisoning the well, you should start asking why they can’t be bothered to address the contention being argued.

An example

Let’s talk about Harold Reyonolds.

Reynolds was one of the key participants in the now-infamous MLB Network roundtable on Baseball Prospectus’ PECOTA projection system. His contribution to the discussion, aside from saying “stats are ridiculous” repeatedly and calling sabermetricians “stats mongers,” involved introducing the idea of someone with a quality .215 batting average. And I don’t think he meant some Adam Dunn-like Three-True-Outcomes deity.

I especially don’t think that’s what he meant after reading this most recent blog posting of Reynolds’:

But what I’ve been witnessing while I’ve been a broadcaster is everyone using these stats to try and explain the game of baseball. Not all statistics work. Some do, some don’t. And one of the stats that has become real popular is OPS. On-base plus slugging. All of a sudden, it’s this stat that defines whether a guy is a good ball player or not. And the fact of the matter is, if you’re a power hitter then the situation will dictate what a pitcher does with you—either walk you or pitch you real careful. So more than likely you’re going to end up on base and therefore your on-base percentage goes up. This in my mind has become the stat the everyone thinks is the be all and end all. It is not. If you have a ball club that’s a great offensive team then that changes everything. But if you have a guy like Adrian Gonzalez, for example, his OPS is going to be high—he’s got a lot of home runs and walks a lot … because you’re not going to pitch to him. Power guys like Giambi and Dunn have always had high OPS because no one wants to pitch to them. But it takes two hits to score them from first.

I’m not here to take down Reynolds’ argument (others have done so, if that’s what you want); nor am I here to concede the aspects of it that are technically correct (like the fact that OPS is not the perfect measure of offense).

I’m here to ask: What’s the point in engaging Reynolds in this discussion at all?

Does anyone think that if Reynolds is led by the nose to a handful of correlation coefficients or a linear regression he’s going to go, “Oh, well that makes sense?” Is there any reason to think that if someone shows Reynolds a context-dependent stat like WPA or linear weights based on base-out state, the scales will fall from his eyes and he will become a born-again sabermetrician?

A Hardball Times Update
Goodbye for now.

Well, it depends. Will any of those newfangled metrics conclude that, say, Ichiro is a better hitter than Adam Dunn?

Reynolds’ attitude towards advanced offensive metrics seems to be that they are wrong because they don’t agree with what Harold Reynolds thinks.
That’s it. The rest is incidental—if you prove those wrong, Reynolds will offer something else, or go back to talking about how “stats are ridiculous.” The conclusion comes first, the argument second.

In that case, does it make any sense to argue with Reynolds? Any reason to suspect that Reynolds is going to argue in good faith?

Let’s move on to a favorite whipping boy of the online sabermetric movement, Murray Chass. He wrote, a while back:

Things I don’t want to read or hear about anymore:

Statistics mongers promoting VORP and other new-age baseball statistics.

I receive a daily e-mail message from Baseball Prospectus, an electronic publication filled with articles and information about statistics, mostly statistics that only stats mongers can love.

To me, VORP epitomized the new-age nonsense. For the longest time, I had no idea what VORP meant and didn’t care enough to go to any great lengths to find out. I asked some colleagues whose work I respect, and they didn’t know what it meant either.

Finally, not long ago, I came across VORP spelled out. It stands for value over replacement player. How thrilling. How absurd. Value over replacement player. Don’t ask what it means. I don’t know.

That’s a perfect summation of the school of thought I am discussing here: VORP is nonsense, even though (or especially because!) Murray Chass doesn’t know what it means.

Now, there are legitimate arguments against VORP as an estimation of player value (I have made some of them myself). Does anyone think that if presented with them, it will create an opening to discuss other, more correct metrics with Chass? Or are you simply providing Chass with enough ammunition to use against VORP without providing him the means of coming up with something better?

The answer

Imagine you’re on a baseball team (it could be your local rec league, it could be the New York Yankees; for this example it makes little difference). The other team strikes out three times in a row, and yet they send another batter up to the plate.Nothing convinces them to take the field and allow your team a chance to bat.

Should you pitch to the guy?

Of course not! Only one option is presented to you: to leave the field of play.

It’s the same in rhetoric as it is in baseball. Both sides have to play by the same rules, or there’s no point in playing along. It’s an old aphorism: never wrestle with a pig, because you get dirty and the pig likes it.

Now, aren’t there some on the sabermetric side of the aisle who similarly move to the ad hominem to close off debate, and who will argue their position despite the evidence? Of course there are! Is this right? Of course it isn’t! But in either case, it is simply appropriate to walk away from the argument until that person is ready to have a fair fight.

The right approach

What should you be doing, then?

The correct approach is to point out poor arguments and bad logic, but to be non-confrontational about it. Remember, you don’t want to argue with people who won’t argue in good faith. What you do want to do is lay the groundwork for an argument in good faith, regardless of whether or not the other party wants to participate.

Because there is typically a third party, of course, one who is neither persuaded to the cause of the sabermetrician, nor set against him as one who doesn’t bother to watch baseball games. It’s those people who are pliable, who can be persuaded with evidence.

So persuade them with evidence. Don’t let the carnival barkers and the hucksters and the swindlers of the world distract you from this. Don’t take the bait.

And remember that the inflexible, the ones dead set on a conventional wisdom that’s long since fallen out of date, are a dying people. Let them pass. There are plenty of people out there who aren’t so calcified in their views, who cut their teeth on the works of people like James and Palmer and who are receptive to an evidence-based approach to baseball.

And they are the future. If some wish to remain in the past, then by all means let them be buried there.

References & Resources
Anyone who wants to argue on the Internet should absolutely peruse this list of logical fallacies. Another great resource is the Forrest of Rhetoric.


13 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Marc Schneider
14 years ago

I agree with sabermetrics in general but I have to say that my eyes glaze over at times with respect to some of the stats.  But that’s ok; I don’t have to read everything sabermetric.

What I find disturbing is that so-called journalists (including broadcasters) that supposedly report on baseball for a living refuse to do even enough research to develop enough of a basic understanding of the stats to do a reasonable critique.  Can you imagine if Murray Chass was a foreign correspondent (say in Iran) and he said something like “I checked out the feeling of the populace about the mullahs but I couldn’t understand what they were saying and found it boring so I didn’t bother anymore.”

rockundrolla
14 years ago

Marc Schneider’s point I think is key… really it’s THEIR JOB to understand the issues. Not to like them, but to understand and not dismiss without the research. To extend the analogy.. people in their professions are required to keep on top of new technology.. designers, programmers and new software… doctors with new equipment and techniques. If they aren’t continuously learning then they cannot do the job to the best of their ability. Stats don’t have to be used because they’re there, but they can lead to a greater understanding.
It’s just laziness and sadly, most people are guilty of it, a lot of the time.

dan
14 years ago

“The other team strikes out three times in a row, and yet they send another batter up to the plate.”

Maybe you dropped the third strike? wink

Brad
14 years ago

This is the classic paradigm shift. Had Moneyball never been published, it would’ve arrived somewhat more slowly, but was nevertheless inevitable.  (Also, the A’s would probably suck less now due to inferior competition, dammit!). 

The fact that virtually every MLB team has embraced sabermetrics to at least some degree means the cat is totally out of the bag.  The crotchety folks can piss and moan about it, but we (well, *you*, I’m no sabermetrician) can just sit back and revel in being a part of the now-dominant paradigm.  Let the revelry begin…

MatthewA
14 years ago

never wrestle with a pig, because you get dirty and the pig likes it.

Who said anything about Sarah Palin?

Rimshot, please.

But seriously, folks: Marc S. nails the issue on the head with regards to baseball writers not bothering to get a grasp on stats research. They’re so frightened by issues of accessibility (which has some merit – some sabermetrics discussions can get pretty dicey if your math skills aren’t so sharp) and dismiss it outright. It’s really to the writers’ detriment, and will catch up with them eventually.

That’s why I continue to find it refreshing to read Rob Neyer’s stuff at ESPN. He’s proven his chops with handling sabermetric research, and he writes now speaking from that experience. He’s very accessible, but is happy to show where the stats are to back up his info. If writers like Chass or journalists like Reynolds could even do that, the baseball world would be a much happier place.

CrotchetyOldGuy
14 years ago

Maybe it would be easier to engage us old guys in productive conversation if not for smug, self-important sites like “Fire Joe Morgan.”

There’s two pigs wrestling around in this mud puddle.

Winston S.
14 years ago

If there is hope, it lies in the proles.

Gerry
14 years ago

I think the line is, “never wrestle with a pig, because you *both*  get dirty, and the pig likes it.”

Jake
14 years ago

I appreciate sabermetrics, but sometimes the more esoteric stats do get a bit intractable, especially the “total value of a player” stats.  I can grok simple stuff like OPS+ and even EQA, but something like Win Shares is just beyond my comprehension…

and then there is WARP and VORP – I know one of them accounts for defense and the other does not, but otherwise I don’t understand how they are calculated, and that to me makes them indistinguishable from magic and devalues them in my eyes somewhat.

One of my favorite Baseball Prospectus features is the ability to mouse over a statistical term and immediately pull up its definition … but some remain undefined, and that troubles me.

Nick
14 years ago

Jake- I know exactly how you feel.  When I first got into sabermetrics, I knew about the metrics and what they attempted to measure, but given that I didn’t know how they were created, they didn’t hold much meaning with me.  But I stuck with it and read a lot and am now able to somewhat understand the intricacies of the stats and can see why they are a lot more predictive and meaningful than the less esoteric ones. 

BTW if you want to know more about how WARP type stats are calculated, there are some helpful primers out there:

http://www.beyondtheboxscore.com/2009/6/20/919602/war-lords-of-the-diamond-pitchers
http://www.beyondtheboxscore.com/2009/6/12/906943/war-lords-of-the-diamond-position

And of course if you don’t understand things you can ask questions, that’s what blogs are for.

Jake
14 years ago

Nick – those two links … good stuff!

now who can explain Win Shares so elegantly?  I love Bill James’s Historical Baseball Abstract (2001 edition is the one sitting on my shelf) but it does not do a good job of explaining just *how* Win Shares are formulated.

Nick
14 years ago

Colin- the article you quoted at the start was actually a parody.  Just FYI.

Dave Studeman
14 years ago

Win Shares isn’t as inherently elegant as WAR, which is a real basic problem.  But here’s an overall description of how the system works.

http://www.baseballgraphs.com/main/index.php/site/details/#sharecalc