Talkin’ Run Differentials

Nate Silver has a cool article (subscription required) about the different types of strong and weak teams, and it uses the Brew Crew as an example in contrast to the Yankees.
The Crew is characterized as a Strong, Flat team, which basically means that we have a bit of star power, but that our main strength is that there have been very few negative performances. It’s true that Prince and J.J. are breaking out — indeed, Jiggly Puff has been an absolute monster this year. But, we’ve been succeeding by having most of our guys check in at slightly above average, and only a couple check in at significantly below (Graffanino and Bush); the latter are countered by Fielder, Hardy, and KoKo.
The Yankees, on the other hand, are checking in as an Average, Sharp team, because while they’re getting awesome years from guys like A-Rod, Jorge Posada, and Derek Jeter (the “sharp” part), they’re also getting very bad years from Kei Igawa, Bobby Abreu, Doug Mientkiewicz, and Melky Cabrera. The star power barely makes up for the anti-star (the average part).
The conventional wisdom is that a balanced lineup like that of the Brewers will help them to whether the dog days better. If Prince goes into a slump, say, one of any number of guys are liable to go off in his place on any given day. Most recently, Gabe Gross.
I guess there’s something to this, but it makes me want to see some exploration of streaks and slumps to see if some players can really be defined as “streaky”. Like clutch hitting, it might take some work, but you could probably just take a guy with a reputation for being streaky who’s had a fairly long career, and set parameters for what streaky is. For example, we might say that if more than 40% of a player’s VORP is accumulated in blocks of games that account for less than 40% of his games played, then he is streaky. I just pulled these percentages out of thin air as a hypothecal, but once the numbers are tweaked, this might be an interesting study. You could then begin to identify different levels of streakiness, figure out what percentage of a player’s production has come from “streaks”, and who knows, maybe even begin to predict when a player is likely to streak or slump based on career trends, comparable players, etc. If I was more mathematically savvy, I’d do it.
Anyway, in the midst of a rough patch, this is actually a bright spot for the Brewers. Now that Graffanino’s playing time will be drastically reduced in favor of what should at least be a moderately above average contribution from Ryan Braun, the team has shored up it’s one offensive sore spot. I see this as a fluke. The team’s going to come around and start scoring more runs.
Also, big ups to Nate Silver for mentioning Edward Tuft, who really makes beautiful books. It’s nice when I can combine my baseball nerdom with my other types of nerdom.