yeah, i would use caution in reading what moy is saying here. it seems to me the OP is talking not elite d1 ranks, while moy is. everywhere but elite d1, i feel potential is not optional. i don't care if you are the best coach in the game, or the worst, or anything in between - knowing players' potential is absolutely critical.
to the OP. what i think you are looking for is how to weight the tradeoff between ratings and potential. clearly, there is no absolute rule to follow here. to me, the best way to go is to project players out to juniors and recruit who will make the best end-of-senior junior (obviously if you are really planning in depth you might vary this - but this approach will get you 95-98% of the way there). a high potential category is 20+, low is 0-5, med is 6-19. if you break the potential down via evals (very smart, when two players have a lot of highs and its not clear who is better), high/high is 28+.
so, that gives you a range for the maximum value. you have to dock the player some based on how much potential he has and his work ethic, and playing time. for example, a guy with 5 high potential categories and 30 work ethic who wont play much as a freshman, will play 15 as a soph, and start as a junior, is probably going to be at least 5 points away from meeting his potential on avg in those high potential categories. so i would dock him some off his maximums.
hopefully that helps - let me know if you have any questions on that...