I have long been in the "WE times Potential" (where I assign each potential level a certain number) camp. It works.
The X-Factor here, though, is that high WE tends to be a self-fulfilling prophecy of sorts. To rephrase, high WE begets high(er) WE much more rapidly. And higher WE means your "WE times Potential" number increases faster. Yeah, it makes for a moving target, but it's something to keep in mind if you're trying to break a close tie between two players. I haven't modeled it mathematically, so I just trust my gut. You'll form your own gut feelings on these things over time, if you pay close attention.
Also, don't overlook the "potential is a spectrum, not discreet values" point that has been made in this thread. I summarily reject the notion that there is any overlap of the potential levels - that is to say, a "high average" recruit's potential is NOT EVER higher than a "low above average" recruit's potential. However, there is substance to the notion that a "high average" recruit can *basically* grow at the same rate as a "low above average" recruit, all other things being equal.