When you think about it, really, it makes sense. What is the market for an honest assessment of the talent? Coaches aren't going to pay for it. They aren't going to (and shouldn't) trust anyone outside of their own assessors. So the only person they're evaluating for is the subscriber, and until there's some sort of accountability mechanism, who is gonna even know that the sites were that far off?
It would be interesting if someone had the time and inclination and math skills to create some sort of database with some number (similar to MLB's WAR) and track each player's WAR vs. his expected WAR (eWAR?) based on his rankings from each service. Then you could have an easy, one-number-to-look-at bias filter.
If a site is consistently negative on WAR minus eWAR for a school--boom. There's your bias. (Or if they're always plus on the blue bloods).
Probably someone is already doing something like this. I seldom have an original thought. I'm getting too far into the details here but my main point is that there's no tracking system to hold predictions/evaluations accountable, so until they begin to look ridiculous, they're just gonna keep following the $.