Monday, August 20, 2007

D-Backs: Random Walk or Success By Design?

One of man's great faults is that he must attach an explanation to everything, even that which can be explained by randomness. Market pundits make a job out of this, as do talking heads on baseball broadcasts. Sometimes, a team defies expectations by so much that even smart analysts can be fooled. This year, it's the Diamondbacks.

If a coin comes up heads nine times out of ten, it's easy to ascribe this to luck, because we "know" the coin is a .500 proposition. On the other hand, a baseball team has an unknown and variable intrinsic winning percentage. We can estimate a team's quality by using their current record, but W-L records are subject to heavy variance, which is why runs scored and runs allowed are more accurate predictors of future performance. Some even take this a step further, breaking a team's performance down into individual elements.

As regular readers have certainly guessed by now, I bring this up because the Diamondbacks currently lead the NL West by four games despite being outscored by 20 runs on the season (and 35 in the BP Adjusted Standings linked above). Naturally, everyone has proffered his own theory on why this is so. Like the many explanations of the crime drop in U.S. cities in the 90's, most of these contain some element of truth but explain only a small part of the deviation.

Here's a simple way to show that Chris Jaffe's analysis cannot explain the full 11 games separating the Baby Backs from their Pythagorean W-L: Let's wipe 30 runs off the slate for the Zona mop-up men, giving them roughly a 4.50 ERA for the season. This gives their bullpen a more typical split between the back and front ends--and our Pythagorean record is still 7 games short of the D-Backs' pace. Remember, that's AFTER we make 30 runs vanish into thin air.

Can the bullpen explain the rest of the difference merely by being an above average collection of pitching talent? No, and it's not particularly close. If you think a good bullpen automatically makes you invincible in close games, I refer you to the 2005 Indians, who had the best bullpen ERA and best team in baseball, yet missed the playoffs.

Even if we assign credit everywhere we possibly can, there will be at least five wins unaccounted for. Basically, the best-case scenario is that Arizona is a 66-59 squad, and I personally think the truth is even bleaker than that.

I'm not trying to sell the D-Backs short; Bob Melvin deserves credit for his bullpen usage, and the team has hit well when it counts. But anyone who thinks this is the best team in the NL--or even close to it--is going to be sorely disappointed when October rolls around.

No comments: