Crashburn Alley

A Response to Blogger Ed Hardiman

by Bill Baer on December 4th, 2008
Posted in MLB, Philadelphia Phillies, Sabermetrics | Print | 14 Comments »

Preface: Ed Hardiman is a blogger within the FOX Sports blogosphere. As one who used to blog there as well, I’ve had many an encounter with the man and have nothing but great things to say about his writing, especially considering he is a fellow Phillies fan. However, I can’t say great things about his criticisms of Sabermetrics. A crafty Google search might lead you to some others, but his most prominent criticism was written back in July of 2007 called Slobbermetrics.

He’s sarcastic is some areas, serious in others, but his general point is that Sabermetrics are not a valuable tool in baseball analysis. In our many discussions there and elsewhere, he was, for the most part, rather civil and you can probably see that in the comments that follow (my handle there was UltraMegaOK1988).

With that said, I want it to be understood that I have a great deal of respect for Ed. I just happen to think he’s dangerously uneducated on the ideas he argues against.

Should you read this, Ed, feel free to clarify any claims of yours you think I may have misrepresented or took out of context, or to just continue the debate.

Blogger ian2813 interviewed Ed. Question 9 is the one in question.

He asked Ed,

You gained some notoriety on the internet last year thanks to your “Slobbermetrics” article. In all seriousness though, what do you really think of Sabermetrics?

I’ll take Ed’s response bit-by-bit and respond to it, his words in Italics, mine to follow.

As a viable interpretation of the past it’s OK, as a predictor of the future, not so hot.

Untrue, as Vegas Watch points out. The top-three best predictors were PECOTA (from Baseball Prospectus), and Rob Neyer and Keith Law, two very Sabermetric-friendly analysts who work for ESPN.

When you think of Steve Phillips in regards to Sabermetrics, the word “antithetical” comes to mind. So it’s no surprise that he finished next-to-last with Buster Olney, another “antithetical” under him.

Moneyball hasn’t changed baseball. Influenced it perhaps and improved Boston, but what has it done for the A’s?

A) Moneyball != Sabermetric bible. This is what happens when you take something the media says and you run with it. Moneyball is not about worshiping computer numbers; it’s about finding what’s undervalued in the baseball market and capitalizing on it — a great strategy for teams with marginal payrolls like the Oakland Athletics. There are some attributes that work for larger-market teams as well, but to say that Moneyball “improved Boston” is disingenuous. The Red Sox are one of the more Sabermetric-friendly organizations however, as evidenced by their hiring of Bill James. By the way, since hiring James, the Red Sox have won two World Series.

B) Billy Beane became the general manager of the Athletics in 1997. In the 11 seasons since, the A’s have won four division titles and one Wild Card berth. They never advanced past the first round** — prompting him to say, “My shit doesn’t work in the playoffs” — but four straight playoff berths (including 90+ wins five straight seasons) is impressive nonetheless. Under Beane, the A’s have averaged about 89 wins per season. With the payroll he’s had for most of the years, that’s particularly impressive.

** Correction: Thanks to reader Ernie, who in the comments below pointed out that the A’s did, in fact, advance past the first round of the playoffs in 2006 against the Minnesota Twins. They were swept in the ALCS by the Detroit Tigers.

I think Charlie Lau’s contact hitting sytem has had a far greater impact than Sabermetrics.

This statement means absolutely nothing unless it’s supported by evidence. I realize it’s an interview and it’s not the place to unleash a thesis on Lau vs. Sabermetrics, but I’d be surprised if Ed’s actually done research and compared the results of the two, instead of simply throwing a statement out there to discredit Sabermetrics.

It’s like believing in fortune tellers.

No, it’s not. It’s the exact opposite, in fact.

Fortune tellers use the intangible or supernatural to explain past and future phenomena. Sabermetrics use the tangible — logged observations — to explain the past and give an educated statement about what is likely to occur in the future (emphasis on the word likely). For instance, a Minor Leaguer with poor plate discipline and marginal on-base and slugging, is not likely to translate well as a Major Leaguer. He definitely could prove us wrong and turn into the next Albert Pujols, but even if he does, it’s not a discredit to Sabermetrics, since it’s really about probabilities.

The problem a lot of anti-Sabermetrics people have is that, as long as Sabermetrics don’t explain everything, it’s not good for anything.

If you exclude the failures it looks great, when you factor them in it resembles flipping a coin.

Again, evidence is needed here. This is another statement meant only to discredit Sabermetrics. Ed’s intent should not be to discredit Sabermetrics but to provide valid, objective criticisms so that they may be improved. Instead, as ideologues do, Ed is simply trying to squash a differing belief system.

Why wouldn’t a Kansas City just hire the brightest SABR guy if it was so infallible?

This is a strawman argument. No educated follower of Sabermetrics thinks that they are infallible. They have proven to be much more reliable and much more strongly correlated with reality than any other system of analysis out there. But it doesn’t mean it’s perfect and it doesn’t mean it can’t improve.

Arguments like Ed’s need to be squelched and legitimate criticisms need to be highlighted. As a science, Sabermetrics are self-critical and open to change, so if there are any out there (and there have got to be thousands), write something up. Baseball Think Factory is a great place to flesh out arguments.

I’m OK with people who live and die by it, but don’t try to cram it down our throats or claim it’s the last word in understanding baseball.

This is an example of judging the idea by the ideologue. If I call up Ed and yell at him,”Two plus two is four and that’s final!” does that make my statement less correct or less valid? While you may be less inclined to believe what I’m saying, and less likely to have an open mind towards my ideas, it doesn’t affect the validity of the idea.

It’s math that can’t pass any rigorous testing.

Math disagrees with you. If you’ve taken a statistics class in high school or college, surely you’re familiar with the correlation coefficient and r-squared. Baseball Prospectus explains the concept well.

If the player doesn’t conform he’s underperformed?

It depends… either way, it’s a strawman argument.

What does “conform” refer to — projections? Projections are “best guesses” often based on large amounts of recent data. The projections are by no means perfect nor meant to be accepted as the holy grail.

How about the math underperfomed and individual humans can’t be expressed mathematically?

Why can’t they? Numbers are simply representations of ideas, usually quantity. As much as we humans like to think we’re some kind of special, we’re not. We can absolutely be accounted for mathematically.

That humans can’t be expressed mathematically is a very poorly-thought-out philosophical idea.

What about the 2008 Mets? Sabermetrically, that team was better than the Phillies. Just grade out the pitching staffs and the Phillies shouldn’t have been a speck in the Mets’ rearview mirror.

The Mets were not Sabermetrically better than the Phillies. The Phils had an equivalent offense, a much better bullpen, much better defense, a much better bullpen, and a much better bench. The only area where the Mets topped the Phils was with their starting pitching.

In short, the Phillies were better than the Mets any way you want to look at it. And the Phillies also had some unbelievably aberrant seasons, particularly from Chase Utley defensively (+47 according to John Dewan’s Fielding Bible, after a couple years of being in the +20′s), Jamie Moyer, and the entire bullpen (including Brad Lidge). They will significantly regress next season.

The Cubbies should’ve beaten the Dodgers SABR-metrically, what happened?

Actually, a lot of Sabermetric-users picked the Dodgers to beat the Cubs. In the playoffs, it’s two good teams going up against each other where the probability of one winning is barely greater than 50%.

That’s why I take it with a grain of salt.

Ed takes Sabermetrics with a grain of salt because…

• Sabermetrically, the Cubs were “should’ve” beaten the Dodgers, even though we’re talking about percentage points in probability.
• There has been no refutation of his strawman argument that Sabermetrics said the Cubs “should’ve” beaten the Dodgers.

So, yeah, there you have it. I respect anyone who disagrees with the tenets of Sabermetrics, but is it really asking too much to read up on the subject before going on a two-year-long (which is about how long I’ve “known” Ed) diatribe against it?

The way to refute Sabermetrics is to find objective evidence that it is not as good an indicator as some other method of analysis. As yet, there has been no such evidence and mountains of evidence supporting Sabermetrics.

If you don’t like Sabermetrics because it’s nerdy or you don’t like numbers or you enjoy the intangible aspects of the game, that’s great, and I respect your mindset. Just don’t go around spouting that your ideology trumps another especially when you have provided absolutely no evidence.

In conclusion, the key here is evidence. Any ideology should be subject to change based on conclusive evidence. That’s not to say that the evidence we have now will forever be the best evidence, but it’s the best we have right now and that should reflect the mindset. Think of it like a court case: we find your fingerprints on a gun, and the bullets in the gun match the one found in a murder victim. To preserve your innocence, you have to come up with evidence as to why your fingerprints were on the gun yet you didn’t fire the bullet. (i.e. “I keep the gun for protection, and at the time of the murder, I was out bowling with Carrot Top.”)