When Evaluating Recruiting Services, Don't Forget To Divide

This is not an attempt to pick on Bruce Feldman, who's a nice guy and not one of the many members of the media who hates the recruiting gurus reflexively*. Except it kind of is, because Feldman's latest blog entry is an evaluation of the Super Bowl starters based on their star ratings. (Feldman makes educated guesses at players too old for the star system.) The conclusion is that the Super Bowl is pretty light on stars:

So if we take some educated guesses the Colts starting 22 would be made up of: one five-star (Manning); maybe four four-stars (center Jeff Saturday; RB Joseph Addai -- he was ranked by recruiting guru Tom Lemming as the nation's No. 68 overall recruit; WR Reggie Wayne -- labeled as a USA Today honorable mention all-American and Hayden); eight three-stars … and nine two stars or less.

To be fair, Feldman doesn't say this is anything more than "interesting" and stays away from any pronouncements about how useful or useless recruiting sites are. But I'm taking issue with the surprise itself.

When Feldman counts up the Saints he finds two five star-ish types, eight four stars, a mere two three stars, and then a great wave of ten random players who were virtually ignored. Four and five star sorts—the anointed—are then 14 of the 44 Super Bowl starters, or 32 percent.

Is this low? Given that this elite slice of high school recruits represents only about 10 percent of the players Rivals and Scout rank, not so much. Athlon did an excellent study on recruiting rankings applied to the NFL draft and found strong evidence they're worthwhile. Here's a chart showing the results from the first three rounds of the 2008 NFL draft:

No. in each class	25-30	275-325	700-800	1,600-1,800
No. drafted in '08	12	30	29	17
Percent drafted	40-48	9-11	3.6	Less than 1

Okay, yes, more than half of the players in the top three rounds were (relatively) anonymous recruits out of high school. But the percentages are remarkable. While lower-ranked recruits are far less likely to make the NFL or get drafted high on a per capita basis, their sheer number can obscure that.

The number of starters in the Super Bowl is a noisy, somewhat silly metric and shouldn't be taken too seriously either way, but even a surprisingly low count of touted recruits is actually a confirmation that recruiting gurus have some value. This is especially true when many touted recruits' careers end short of the NFL for reasons other than their natural talent, like injury or a tendency to shoot guns in the air. Rankings remain far from perfect but also far from useless.

*(For an example of the form, check any of the Wizard Of Odds's many posts on how recruiting rankings are all a big sham. The Wiz is a blogger now but was a longtime employee of the Los Angeles Times and is decidedly old school.)

This post originally appeared on the Sporting Blog. For more, see The Sporting Blog Archives.