Actually, it's a lot stronger than most think. LWS has a spreadsheet he posted showing where all of the 5 stars in a given year ended up. 26 out of 27 got at least a free agent shot in the NFL. The 4 stars had a much lower percentage and the three stars were much lower than the four. It was all exactly as you would expect based on the rankings.
I think that year was probably an outlier because it would seem you would be likely to get more than one guy in a class of five stars who had a career ending injury or decided he would rather sell crack and smoke weed all day.
But the correlation holds no matter what metric you use... All conference, starter etc... And no matter what time period you look at.
It is tempting to look at a list like the one above and say there is no correlation. But when you realize there are a hundred quarterbacks that are 2-4 star for every 5 star qb and if there were no correlation you would expect to only have one 5 star pan out for every 100 qbs. If you had 2 pan out then you would have a very strong correlation of twice what chance would dictate. 26 out of 27 for all positions is just ridiculous and shows why coaches all want these guys. They don't get paid millions because they can't spot talent.
People are just not very good with statistics. But they are champions of rationalizing the failures of their team. That is why they are called fans --- to denote they are not rational when it comes to certain subjects. No where is this more evident than in recruiting.
About the only thing more accurate than the rankings is the correlation between posts like this one appearing on message boards and the number of losses suffered by the teams those boards support. You can be damn sure Bama isn't worrying over having too many highly rated guys.
You can rest assured when we finally turn things around and recruiting is going well we will no longer see posts like this one.
