from Amazon
Thursday, September 06, 2007
Baseball Prospectus is dead.
Or so says co-founder Gary Huckabay. (The article id number happens to be 6666.)
Here’s what he says:
When I’m talking about ‘baseball analysis’ here, I’m talking about the rigorous review of player performance data. I’m not talking about the inclusion of pitch velocity and location data that’s now coming available, and I’m not talking about the integration of scouting data with performance data. I’m strictly talking about activities like developing value metrics, forecasting, and all the other stuff we do with the massive yarn-ball of data we’ve all put together over the years.
On its surface, it’s a bullsh-t statement to make. However, he’s modified the definition and context of “baseball analysis” to such an extent, that it’s used as shock value. Sort of like me saying Baseball Prospectus is dead. When you use his definition of “baseball analysis”, a term which I will rename as “quatlu analysis”, it’s still a bullsh-t statement to make. If there’s very little to be gained from forecasting beyond the massive yarn-ball of data (a position that Marcel the Monkey agrees with), then why such black box, behind the curtain, IP position of PECOTA? It sounds hypocritical. I don’t know if it is. It just sounds like it.
Analysis and information are only interesting and useful if they inform a decision
That’s about a myopic position as you will find. People who bought our book or come to this site are not necessarily here because they have a decision to be informed. They come here because baseball is a pastime, a hobby, and it’s fun. Even most readers to his own site go there for fun. Gary’s position is strictly from some business GM standpoint. If we were to hold Baseball Prospectus to the standards and mission statement that Gary is evoking, then Baseball Prospectus should close up shop.
He could be better in saying “Professional Baseball Analysis is dead”. And he’s still wrong, as evidenced by the signing of Carlos Lee for 6/100, a move that I have called the worst-signing of the 2007 free agent pre-season. And, when I evaluated all the free agent signings, I did not spend countless hours looking at all the little stats. I did it in almost Gladwell’s blink of an eye. That his current .299/.355/.524 line matches so closely to Marcel’s blink forecast of .287/.346/.509 provides me comfort for my claim. What is clear to me is that there’s a segment in the baseball professional world, a strong mover and shaker segment, that doesn’t get it. They don’t get regression toward the mean, they don’t get sampling, they don’t get positional scarcity, they don’t get how to process massive yarn-ball of data. And even if they have people who do get it, they don’t listen to those people. And professional player agents aren’t any better, with the fantastic signing (by the Phillies) of Chase Utley, the best arbitration signing of the 2007 pre-season.
Even Baseball Prospectus itself is in conflict with itself. On the one hand, you have WARP, which sets a replacement level (akin to the positional scarcity I noted earlier) to such an absurdly low level, as to call into question the legitimacy of the stat. There is no such thing, at all, as replacement level hitting plus replacement level fielding. What does exist, in reality, is replacement level players. On the other hand, you have BP’s little publicized measure like SuperVORP, which treats replacement level properly. VORP itself has its own problems, because it values the walk far too low, and makes no attempt to correct itself. For a site that takes such pride in giving things to decimal places, how can it get the basis wrong?
It is the height of arrogance to say that everything that can be done with massive yarn-ball of data has been done, when they’ve been doing it wrong. And then to say that it has no extra information for the professionals, when the professionals are not even looking at it. For a sizable number of teams, we are still in 1982.
There’s several more columns I could write about the damage inflicted on careers by the bullsh!t macho attitude that dominates locker rooms, but for now, let’s focus on the Dallas Greens of the world. Pitchers’ arms are fragile, and they get damaged by overuse. There’s no ambiguity on this point, and as much as color commentators are doing their best Abe Simpson impersonations when they talk about the evils of the 100-pitch limit, front offices have made the connection between overuse and injuries.
I have shown evidence on this site that pitchers today, and in the past (outside of WWII), have thrown the same number of pitches in their careers. You take top pitchers in their 20s from the past and today, and you will find that those pitchers will throw the same number of pitches in their 30s. This is fact.
But generally speaking, most of that improvement is small beans compared to the two biggies listed above. And that’s the deal. That’s why baseball analysis is pinin’ for the fjords. It’s a simple issue of a bad business case.
Yet another bullsh-t statement to make. A few years ago, Gary was asked about advancements in sabermetrics, and he noted replacement level. It is an enormous thing to know if your player can play 3B or if he can only play 1B/LF. There are countless other things that don’t take a back seat to “the big two”, one of which is simply wrong to begin with.
Gary thinks that quatlu analysis is dead, and he’s wrong. If it is dead, then Baseball Prospectus is dead, outside of Dan Fox. That’s what Gary Huckabay just said. Or at least, that’s what I heard him say.
Generally speaking, I don’t agree with overall assessments by anyone, as if one person has such a handle on all the myriad of issues that they can assess the landscape to such a degree as to proclaim something as dead. I don’t want to hear from some bullsh-t financial analysts telling me how he thinks the world economy will progress, or the real-estate market will change. The market will speak for itself, in terms far clearer and more real than any single person can. People are great for focusing on the bits and pieces. Grand overall assessments and proclamations? That’s what dictators do. Unless you are forced to, who listens to them?
I would hope that everyone will read, and then discard, everything that Gary has written in that article. And then, everyone should discard everything that I said in this blog post. It’s just two bullsh-t articles that adds no value, advances no cause, and informs no decision. The two articles together cancel out, lying dead, and you should be back to where you were before reading all this: reading and enjoying real baseball analysis by Dan Fox, Nate Silver, et al. That’s what I’ll be doing. Baseball analysis is dead. Long live baseball analysis.
What I don’t understand is why he wrote the article in the first place? I don’t necessarily agree with everything you’ve said in response, but I certainly don’t agree with Huckaby either. Because of the goofy definition of Baseball Analysis...what the hell is he trying to say? Quatlu analysis indeed. I don’t understand why he wrote it, what he had to gain by it (or what he was trying to convey) and I certainly don’t understand how we (the readers) are supposed to react to it.