The Sabermetrician's burden

As I've alluded to in a few posts, some recent job duties have unexpectedly reunited me with an old friend--sports talk radio. Miind you, when I say friend, I mean the kind of friend that draws you in at first but whose lack of substance grows gradually more apparent until that day you realize he's nothing but a self-absorbed bore who constantly repeats himself. Much to your disappointment--but not at all to your surprise--the reunion after the long separation only affirms how little the friend has evolved in the time elapsed, and you remember why you let him drift in the first place.

(Both current and former friends of mine should find this all very familiar by the way.)

But my reacquaintance with sports talk radio reminded me of a very important fact: that media outlets such as these still has a very powerful and widespread influence over the way people watch and interpret sports--and for the purposes of this post, baseball in particular. I know that's hard to hear, but it it's the truth. While we spend our hours buried in the Baseball Prospectus, it's easy to forget that we're but a small minority compared to the millions who get their opinions from Colin Cowherd, Jay Mariotti, and the entire staff of the New York Post.

That's not meant to be some snooty condemnation of those media and their audience. It's simply to say that many radio hosts and columnists are quick to form opinions given limited facts and are reluctant to seek additional facts once an opinion has been formed.

As we all know, these individuals are often the most popular with the "casual fan," whose minds often work the same way. After all, fact-finding is difficult, and saying "I don't know" is a shameful admission of weakness. And if knowlege is power, then pretend knowledge will at least make one feel powerful. (Ok, so that was a little bit snooty.)

I guess the big question before I go any further is, do we care? Should we worry that Bob from Buffalo Grove thinks a lineup of .220 hitters could win a pennant or that Steve from Palos Hills thinks that Ryan Theriot is the Cubs' MVP? (I kind of raised this same question in a previous, more long-winded post.) Sometimes--as worked up as we get over such outrageous assertions--we seem to prefer it that way. Keep the knowledge to ourselves; keep the dummies in the dark; use them as fodder for our fun.

I'll admit--I'm as guilty as anyone. Last week I was listening to Mike North (the now internationally famous Mike North ever since he told Ozzie Guillen to "stay the hell out of the kitchen, junior!"), and he was embroiled in a half-conversation, half-rant deriding the emphasis on stats and trumpeting the ever-popular qualities heart and hustle, as they apply to last year's Jim Thome-Aaron Rowand trade. Of course, I was laughing it up on the inside, both repulsed and amused by the implications, caring not for the misinformation of North's listeners.

But at least ostensibly, the intent of the Sabermatrician--or of any other "higher-level" sports fan--is to change the ways of the willfully narrow-minded. If that's truly the case, the champions of Sabermetrics from all strata--not just the Baseball Prospectus--need to be inclusive. Given the drastically different perspectives of these two types of fan, that may sound impossible. But as is always the case with new modes of thinking, the burden lies with those spreading them.

As someone simply loves baseball and defends both perspectives when I can, the following are my humble suggestions to make Sabermetrics more accessible to that so-called "casual fan":

Embrace the arbitrary.

A big reason why the "casual fan" is so chained to the traditional stats of batting average, RBI, etc is the familiarity with round numbers and milestones. Those that swear by Sabermetrics tend to disparage the arbitrary nature of these numbers.

At the same time, there's no denying their appeal. 3,000, .400, 100, 755--these things stick with us. They make the game fun to follow. And while I never expect people to hear, "Alex Rodriguez has just closed within 10 win shares of Babe Ruth!" I do think casual fans could eventually latch onto the .300 EqA club or the 150 ERA+ club.

Fire Joe Morgan who used the QB rating analogy to make the same point. If football fans will marvel at 100 quarterback rating--a stat whose complexity puts most Sabermetrics to shame--surely .300 EqA or 150 ERA+ isn't too much to swallow.


Revel in the deviations.

Sometimes, it seems, when results don't follow the best projections or deviate from the strongest indicators, they're treated by Sabermatricians with hostile skepticism, bordering on bitterness. This is a no-no.

Nothing alienates the traditional fan like a perceived attempt to digitize the game they love. While the numerical element can enhance enjoyment of baseball, very few people would cite its calculable nature as a primary reason why they love the sport. Even for me--someone who derives pleasure from that scientific perspective of the game--baseball's quantifiable and instantaneously random nature can be discomforting. Those of us that equate baseball with pure pleasure prefer not to see it as we see meteorology--even when the two behave almost identically.

Projections and indicators are all you need if you're a bookie or a GM, but for the fan, the deviations are all we have left. The MLB playoffs are either a stage for human talent and drama or a temple to the god of random error. That's left for each fan to decide.


Embrace the voodoo stuff.

Are we sick of hearing what a gritty player David Eckstein is? Sure. Are we tired of hearing that Darrin Erstad played football in college, and that makes him a tough-as-nails, team-first guy? Absolutely. Would we rather our analysts stay tuned to the on-field talent rather than harping on what a great/terrible manager Ozzie Guillen makes? Of course.

But will these sorts of discussions ever cease completely? That's doubtful. And for that matter, are these hazy subjects--i.e. effort, team chemistry, managing strategy--entirely dispensable elements of the game? Well, no. They may not amount to much compared to, say, an injured superstar, but each has some, albeit tiny, impact on the measureable results.

Do motivated, happy players who like their teammates play better than disgruntled ones? One would have to assume so... at least a little. Say a team committed itself to running super hard on every ball in play and followed through on that commitment for an entire season. Wouldn't that team see a small net positive effect on their offensive production? Well, maybe not. Maybe that would only lead to fatigue and injury. But see? These are the types of value-oriented debates that could bridge the stathead mentality and the achievement ideology were the atmosphere more inclusive.

The old guard analysts will tell us how excellent basestealers can impact the game simply by rattling opposing pitchers or how young starters benefit from veteran mates in the rotation. These are anecdotal truisms that seemingly most major leaguers swear by, so it doesn't make much sense to brush them off altogether. The measureable effects may be overblown, but the self-imposed challenge remains to quantify that.

A veteran teammate will never be the sole cause for a breakout season by a young pitcher, but can it be a contributing factor? If one wants to argue no, I say the burden of proof lies with the person refuting the anecdotes, not the person telling them.

The point is this: It's tempting to smother the voodoo stuff with a blanket of Sabermetric appeals that don't adequately disprove the conventional wisdom. But doing so reaches no one except those that see things your way already.


Admit the limitations.

This one hardly needs to be said, but some Sabermatricians--not a lot, but some--still approach baseball topics with too much of an absolutist mentality.

Not everything can be quantified. Not everything has a control group. All metrics work within certain ranges better than they work within others. (For the record, I note the irony in criticizing absolutism and then proceeding to write three consecutive absolutist statements.) As the scope of discussion narrows, that grows even truer.

Unfortunately, too many people adopt a Sabermetric mantra and treat it like a Smart People badge without ever understanding its limitations or acknowledging the proper degree of uncertainty.


Know your damned shit.

To spin off my last point, credibility is only established through understanding. As much as we rag on fans who love sac bunts because their selective memory underrepresents their failures, those who put blind faith in Sabermetrics are no more credible. The former simply has a keener sense for small-scale events than for their aggregate effects, and they can hardly be faulted when their intuition distorts reality by five or ten percent.

On the other hand, the person who lacks that independent understanding of baseball but drops Sabermetric terms as if he invented them only comes off as a pretentious hack bitch. And because he probably doesn't understand the inticacies of the terms he spews, that's exactly what he is. The danger with such holistic evaluations lies in ignoring their inner components and watering down our perspective as a result.

That is to say, Sabermetrics are nothing but numbers with funny acronymns until we can tie them to something tangibly baseball-related. That's true of traditional stats as well, but even more so of Sabermetrics because of their inherent complexity and indirect link to on-field "moments"--the runs, hits, homers, etc.

The primary questions, then, are not, "Who has the higher EqA?" or "What is Justin Morneau's VORP?" but rather, "What are typical ranges for EqA?" "What traditional stat or ability does EqA tend to overrate compared to VORP?" "What key assumptions are inherent in the RC27 formula?" "Why was WARP created? What shortcoming of traditional stats was it designed to overcome?" And so on, and so on.

This may sound overly dogmatic--and it is--but the person who touts Sabermetrics without asking these questions is no better than the guy who blindly swears by the hit-and run. Just latching on with the faster yacht doesn't mean you're not an anchor. (And that, my friends, is called poetry.)


Don't go creating a new monster.

Following my own logic, a sudden boom in the popularity of Sabermetrics would only mean more of those individuals I criticized in that last paragraph. We would have fans everywhere passing judgment on teams and players based on evalations they don't fully comprehend.

But then maybe Sabermetric measures really are that fantastic. Maybe they're so effective that a blind Sabermetric comparison is better than any independent judgment based on some combination of traditional stats and intuition.

Going back to the QB rating analogy, football fans rarely object to comparing quaterbacks in this manner. Or an even better example, how many of us know the first thing about seismology? Nevertheless, when a big earthquake hits, the first thing reported is the quake's impact on the Richter scale. We have no idea what this truly means--except that a higher Richter scale rating means a "bigger" earthquake--but we go by it anyway.

Hell, come to think of it, what percentage of casual baseball fans could compute ERA or slugging percentage?

I guess it all comes back to a simple question: How much blind confidence should a fan put into a highly-refined yet esoteric "better" and "worse" comparisons?

I guess that all depends on the confidence of the creators themselves.