Word! Well, we gonna agree to disagree here, keegsy. I'd rather have more brackets up front, see where the population opts to play, and filter from there. Folks are gonna play where they feel they can be the most successful & have the most fun - why limit the menu when the depth of menu is most important, during a test run?

Yea, ya might be right re: pt 2 - however, the best A-B test will be to keep things contained as much as possible for a few seasons. Confining players allows for a stable data set for each bracket, makes it easier to track retention and makes it possible to tie behavior (such as participation over time) to bracket design. I could be wrong, but definitely trying to suggest as much of a scientific method here as possible.

At any rate, cheers hombre! Thanks for the thoughts