Smiley Face, Money Bags, Eagle

[Exsubmariner]

Hello all,
Some time back, the question came up about who between the Smiley face, money bags, and eagle icons made the most BB/100. I've been banging my head on my wall trying to think of a way to figure this out.
The best idea that I could come up with was to go to my PT databases and look at the player stats. I have about 12K players in my database. My idea was to order them by VP$IP and separate bands of 10%-15% and 15%-20% and 20%-25%, calculate the total BB won in each band then divide by the # of hands in each band. This should give a BB/100 for each band of VP$IP.
Before I waste a couple of hours of my time going through my database to find this answer, I want you all to tell me what a bad idea this is. Will it really yield any useful data? Will the sample sizes be inadequate? How do you account for one player being better after the flop than another? Does it all come out in the wash? Should I use different bands?
More importantly....I don't want to go through all this if there is already a general consensus out there or if anyone already has done the number crunching to find the answer. If so, does anyone know of any old threads where this is discussed?
If no one has done this, what is your opinion as to what icon will be the top BB/100 winner? Please explain your answer.
X

[Baseballer02]

Different players will have different tendencies and strategies in their play that work for them. I don't think you'll get any definitive answer as to which type of player has teh best rate.

[Nick C]

I think you'll find that the aggressive postflop players fare pretty well in general, including the LAGs.

In part, the reason for this will be that players who got lucky and ran well over their small individual samples will bet and raise with their hands. Thus, the players will appear more aggressive than they truly are (and sometimes looser preflop than they truly are as well).

Unless you have very large samples for the individual players you include, your results are going to mix cause and effect too much to be very meaningful.