BB with a 20 man roster is tough! Do you do FAs?
FFPC does 26 without FAs. I like that format as you get what you get, so to speak.
No FAs in MFL10s... pretty tough.
I was wondering about the question of 2QBs with one taken earlier or 3QBs taken late. I quickly looked at the numbers in 2012 and, though there is nothing scientific or precise about this exercise here's what I found with a very quick look:
In the MFL 10 format here are the starting QB scores the following combinations would have yielded:
Brees and Luck: 416.8
Rodgers and Romo: 415.5
Peyton and Roethlisberger: 386.1
Cam and Eli: 384.7
Ryan and Eli: 379.3
Dalton, Freeman, Flacco: 389.1
Dalton, Roethlisberger, Schaub: 381
Freeman, Schaub, Bradford: 378.3
Bradford, Palmer, Rivers: 360.8
Weeden, Tannehill, Ponder: 315.1
Not sure that this means much, but it does seem like unless you're landing a truly elite QB, you can do just fine waiting on QB and picking up three guys in the QB12-24 range. That is, as long as you don't pick three stinkers. This year I feel pretty good about platooning three of Dalton (ADP in the 10th), Flacco (11th), Roethlisberger (10th), Rivers (12th), Freeman (12th), Bradford (13th), Schaub (13th), Cutler (12th), Tannehill (14th), Alex Smith (15th), Palmer (13th), and even Weeden (22nd). In the first 8-9 rounds I'd rather just land the extra starting WR or committee RB, as those positions tend to dry up much earlier than QB. That said, I can also see the appeal of taking a guy like Peyton in the 5th (as he's been there quite regularly), not take backup until the 12th or 13th and having an extra player at WR, RB or TE. Either strategy could work if you pick the right players...
Run the numbers for Brees/Rodgers paired with the guys in your second group. If you go stud at QB, you generally will be fine waiting until 20+ QBs go off the board before drafting a second.
Using your example, my opinion is that
QB 1st + QB 22nd + WR/RB 10th + WR/RB 13th + WR/RB 15th > WR/RB 1st + WR/RB 22nd + QB 10th + QB 13th + QB 15th in a bb format.
I was just winging those combos before, so I'll do these with actual ADP from 2012. According to MFL, QB22 was Locker (who got injured), after him were Wilson (who's situation was a bit odd do to his being a rookie) and Alex Smith (who didn't play a full season) so I'll do this for two guys, Wilson and Sam Bradford who was next up at QB25.
Rodgers (ADP of QB1) and Wilson (QB23): 416.6
Brees (QB3) and Wilson (QB23): 428.5
Rodgers (QB1) and Bradford (QB25): 396.9
Brees (QB3) and Bradford: 403.6
Wilson's week 15 game (41.4 points) really boosted the scores here, and I'd say that he would definitely be considered a 'hit' on a backup. If you 'missed', say drafting Locker, these would be the results:
Rodgers (QB1) and Locker (QB22): 393.1
Brees (QB3) and Locker (QB22): 376.4 (this included a zero during Brees' week 4 bye)
And a few more 3 QB combos using ADP:
Flacco (QB17), Freeman (QB18) and Bradford (QB25): 401.9
Dalton (QB21), Fitzpatrick (QB21), Wilson (QB23): 388.1
Luck (QB16), Dalton (QB21), and Wilson (QB23): 407.7
These and the ones I ran above (in my previous post) are more or less random (I just generally picks guys in that range I might have considered). If you picked poorly (like the Weeden, Tannehill, Ponder combo above) you'd surely be out of luck. But if you picked well I think you'd be OK with the 3QB combo (as in the case of Flacco, Freeman, Bradford or Luck, Dalton, Wilson). Honestly, either strategy requires you to 'pick the right players' to succeed, but I feel pretty certain that if you pick the right players with the 3QB strategy, you can be better off by going RB, WR, or QB early. I'd love to do some comparisons on that front, but that would simply require more time than I have and I'm no math whiz. I've already tried both strategies on MFL10s, for example, drafting Cam (QB3)/Weeden (QB27) and Eli (QB14)/Culter (QB21)/Weeden (QB29) in two of my leagues.