I decided to take a look through some past Rivals 100 lists to see just how accurate recruiting websites are at predicting the success of college football players. To do that I looked at a random year (2002) to see how well the top 20 recruits fared. Then a took a look at a random All-America Team (2005) to see how many were 5 star, 4 star, and 3 star recruits.

By looking through this data it seems as if 5 star recruits are quite overrated. Obviously, it is a small sample size (I would have done more but it took me effing forever) but the results are pretty intriguing. First the 2002 top 100 class.

I classified each player as with these titles: NFL Success ( which means they were drafted in first 3 rounds), College Success (which means they were good in college but late round pick or not drafted), OK (which means, well, they were ok in college) and Bust. Now, because I am looking at exclusively 5 star recruits who have lots of hype and expectations, I judged them harshly.

So, in the 2002 Rivals top 20 (which included Vince Young, Haloti Ngata) there were only 5 NFL Successes (Young, Ngata, Lorenzo Booker, Trent Edwards, Justin Blalock). There were 6 College Successes in this list as well. On the surface, this seems like a good amount of success. But, take into account that these are the top 20 players in the nation and only HALF of them became good college players. The rest fell into mediocrity. There were 2 OK players and 7 outright busts. If half of the best 20 players in the nation fall into mediocrity, then the '5 star' moniker really is a tad overrated.

Now onto the 2005 All-American list. Obviously, these results are a little tougher to analyze because there are about 50 5 star recruits total, 100 or so 4 stars, and a million 3 stars. Obviously the results will be skewed but let's take a look anyway.

Out of the 53 offensive and defensive all-americans, 8 were 5 star recruits, 19 were 4 star recruits, 14 were 3 star recruits, 2 were 2 star recruits and there were 7 players unranked. Again, on the surface it seems as if 8 5 star recruits is a lot, but it is also intriguing that there were actually more 2 star or lower (non-ranked) recruits (9) on the team than 5 star recruits.  To me, this data tells me that there is value to be found in all the tiers of recruiting and the bulk of it is not found in the 5 star recruits at all. 3 star recruits are very well-represented on this team. The most value, it seems, is found in 4 star recruits. For whatever reason, the success rate seems higher. Perhaps 4 star recruits, because they are not ranked as the best of the best, tend to have a bit stronger motivation and work ethic because they weren't ranked as highly as they believed they should have been. Who knows.

This data tells me that the 5 star recruit is quite overrated. It does not guarantee an All-American, it does not even guarantee a good college career. Hell, in 2002, the bust/mediocre rate was almost 50 percent among the top 20 players in the country. Plenty of 3 star recruits go on to become All-Americans and top draft picks. The biggest drop off in talent, to me, is between 3 and 2 star recruits. There are just as many 2 star as 3 star recruits, yet the 3 star recruits had 14 players on the All-American team while 2 star recruits had only 2.

Obviously this is all a very small sample size and not really the most scientific of analysis but it at least gives a general idea of how well the 5 star recruits do.

The lack of 5 star recruits in this year's UMD class does not worry me. There are plenty of 4 star and sleeper 3 star recruits to make up for the lack of perceived talent.