Look at the criteria they rated the prospects on, and it will make more sense. The overall rating of each prospect are based on the individual rating within five categories (skill, physical tools, two-way game, projection as a pro and “intangiblesâ€).
Given three of these categories favour a player like Phaneuf over a player like Crosby, it starts to make sense. Especially if it was a 1-5 scale (or something like that). In skill, Crosby is clearly a 5, and putting a player like Phaneuf at 4 probably doesn't accurately represent the gap in the skill between the two. But it's also tough to put a player like Phaneuf at "3" given he does have some solid skills, and on a scale of 1-5 that would make him just average. So it makes it tough for those doing the ratings.
Overall, I would bet 90% of scouts would take Crosby if they could just pick one CHL player to build around. But the mere fact a small percentage would take Phaneuf speaks volumes about what his upside is looking like. He looks like another Jovocop to me.
Overall though, I don't think the ratings are that bad. Comparing players in different draft years (it includes players from 2002-2005) is tough to do, but interesting none the less. The fact the ratings don't jibe with how you see certain players perhaps should make you take a second look at your own rankings.