So you outright admit that star rankings are purely about how many factories you have on your offer sheet, without accounting at all for actual play on the field.
I didn't think I was defending star rankings, but...yes? That is what I'm saying.
I'm not sure what people are expecting out of recruiting sites. I count about 2,000 total recruits which could get offers from P5 schools. They just aren't going to watch a lot of film on all 2,000 players and then give a good prediction on whether a Safety from the 4A level in Colorado is better than a Safety from the 3A level in Florida.
Using offers should free-ride off of all the work of coaches and graduate assistants watching films and running camps. Offers are a definite, objective independent variable whereas watching film is entirely subjective. There are a few issues:
1. Not every school is looking for the same athlete.
2. The effect of early offers and "strong verbals." Will Bryan had interest from Georgia and Clemson. AJ Grey had interest from Georgia. Both were very early commits. Also, some places like Boise State will have their recruits hurt. If a guy is really good in Idaho, Montana, Utah or even Colorado, they may be unlikely to leave that area and so their offers won't correspond with their possible success.
3. Even using just offers weighted by weighted success, a lot of the rankings don't make much sense. Rivals and Scout made a lot of ND players 4*'s even though many of their other offers were from schools like Indiana, Purdue and Illinois. The impact of the number of subscribers and/or the subjective effect of an offer is clear. A Florida or USC offer still carries weight based on past success.
4. Many of the top recruits, especially using combine measurements, would also give only three years before turning pro.
One thing I've wanted to do is download a Rivals database of offers and seeing if an objective model could be created for a team's future success. Also seeing how an objective model's star rankings would compare with actual star rankings. The outliers for team success would probably be team's with atypical offenses, such as Tech's or Leach's offense where he's fine with short but fast receivers.