Wonked off.

As I have mentioned before, John Gasaway, f/k/a Big Ten Wonk, along with fellow stat-head Ken Pomeroy, have launched a promising new website called Basketball Prospectus. In many quarters, including from Inside the Hall and in ITH's comment threads, Gasaway is taking quite a bit of heat from IU fans about the this paragraph:

Gordon is widely assumed to be on track for a one-and-done career in Bloomington. If this is indeed the case, he’ll be doing well during his brief stint in an IU uniform to perform at the level set by Wilmont last season. No one paid much attention, but Wilmont took a lot of shots and had an outstanding year in 2007. In fact, the even more impressive year that Drew Neitzel had on offense for Michigan State last season could be summarized as simply “Wilmont plus assists.” This year Wilmont’s shots will fall to Gordon; Hoosier fans should be thrilled if Gordon hits about 40 percent of his threes and takes obsessively good care of the ball, as did Wilmont. Same performance, way more hype–it could happen.

I think Gasaway and Pomeroy have been godsends to college basketball. They have created, tweaked, and popularized various statistics that make much more sense than the conventional ways of looking at basketball stats. They have done for college basketball something similar to what Bill James and the Baseball Prospectus guys have done for baseball. Still, one very important thing to keep in mind, and I think Gasaway may have lost sight of it here, is that basketball isn't baseball. Baseball is particularly susceptible to statistical analysis because each at bat is an isolated "transaction," for lack of a better term. If David Eckstein gets behind in the count to a tough pitcher, he can't "pass" to Albert Pujols. While certainly weather, baserunners, defensive play, ballpark differences, and the like can have an effect, for the most part, it's the pitcher against the batter. Basketball isn't like that.

Basketball statistics are valuable, of course, but they aren't as valuable as baseball statistics. Unlike baseball, where there is an set batting order and limited number of at-bats for each player, in basketball, the most efficient way to score is to get the ball in the hands of the best players as often as possible. But there are diminishing returns. The 1990s Bulls couldn't give the ball to Michael Jordan and expect him to score on every single possession. Even great players require effective players around them. On the flip side, some players who aren't necessarily good enough to carry a team can be extremely effective and efficient in a limited role. Commenter Neal E from ITH has a pretty good take on it:

If any of you have heard coach Sampson talk about Wilmont, you will know what he thought of Rod’s “obsessive ball control.” He told us a story of the Northwestern game where Rod hit all those 3’s (9 I think, correct?). Before the game, Sampson was not very comfortable with Rod playing against their slap the ball away defense, and their aggressive play of the passing lanes. He could see the writing on the wall and it read a line of somewhere near 50 turnovers. He gave Rod one piece of advice: “Just shoot it. Every time you touch the ball, shoot it. Don’t pass it, don’t dribble it, JUST SHOOT IT.” His impersonation of Rod’s reaction is the best. “FOR REAL?!?!” I love Rod. But the man can’t pass and the man can’t dribble the lane. EJ can do both. If anything it’s like he is 2 different players. Rod Wilmont plus someone who can touch the ball without losing it.

Wilmont was fantastic last year, and it was to Sampson's credit that he convinced Wilmont to play a specific role. He played it well. As the commenter noted above, Wilmont was a limited player, a spot-up shooter. He wasn't a distributor and wasn't a shot creator. Wilmont's productivity last season was a direct reflection of the change at the top of the IU program. What Gasaway seems to miss is that even if Gordon's points per shot, rebounds, and the like are identical to Wilmont's, it is likely that Gordon will have done much more to make his teammates better. Again, I think it's important to consider statistics. It's possible for a player to be talented and exciting to watch without being terribly productive. Still, the Gordon to Wilmont comparison shows that using stats to analyze basketball is limited when compared to baseball.