A little of each.

if you played press all your life, like me, you'd probably be more in the BLK is not that huge camp. i think i am biased for 2 reasons. 1, i played pre-potential, that is when i was really studying and learning the game. i won most of my titles (all but the first... 5 maybe?) during potential, but it was mostly right at the start of potential, and i wasn't still studying the game like i had been except for the first few months when potential came out.

anyway, pre-potential, you had a lot tighter correlation between ratings. i cared greatly for reb, and blk and reb go hand in hand. i won at least a half dozen championships before i even started to glance at blocking, i totally ignored it, completely and utterly. like i would be able to tell you within 2 points, every rating of every player on my team, except blk and dur. it was effectively dur to me. that belief was wrong, but didn't hurt me much because i loved reb so much, and great reb = good or better blk.

the second bias comes from playing press forever, i think i have won twice without it, but i think they were my 2 more recent or 2/3 most recent wins.

however, i think rebounding is right up there as one of the most important attributes, from a defensive standpoint, and i maintain that any regression against blk that does not account for reb is effectively meaningless. so i hope you don't let that regression bias you too much, like i always let my experience bais me. i am much less against blk being very important than i used to be, but i still am probably pulled a bit in the wrong direction. still, the reality remains that ignoring blk will not hurt new coaches nearly as much as ignoring a long list of other ratings, theres no way you can ignore ath, spd, def, reb, even lp/per like i ignored blk for so long, and have it not really hurt you. anyway, somewhere between all these things pointing us in the wrong direction, is the actual right answer, of how important blk is. and my guess is, its fairly important, but not the most important, defensively. but still, im not sure we are THAT far apart in how important blk is. do you think its the most important rating on defense? or just one of the more important ones?

Based on the regression it's most important for all positions, but I don't think that's true for guards certainly, and it's hard to say for posts...  I don't see how REB would be a confounder given that I'm only talking in terms of OFG%.  I collected the data by hand and tried to ignore putbacks, breakaways, etc.  I'm sure a few slipped through, but hopefully not too many...

I lost very little regression power by limiting my equation to only the highest- and lowest-order terms, IE: FG% = A + B*(ATH^C)*(SPD^D)*(DEF^E)*(BLK*F), and of C-F BLK was the largest coefficient.  ATH and DEF were of very similar values for all positions, SPD was smaller for all positions but not all that much smaller for guards.  In fact, all of them were in the same ballpark for guards, which actually leads me to believe that if you could ever find a guard with poor DEF but good BLK he might have some serious defensive value, but I've never found the guys to test that with and I'm not sure if they would ever exist.