1) The primary purpose of external bearings in BBs is to allow the use of a larger OD BB spindles, which provide increased stiffness for the BB unit. Wider bearing stance does have a slight effect on stiffness, but this is a secondary design artifact
2) This stiffness is desireable, as the crank/bb unit is (obviously) a primary recipient/converter of muscular force. Stiffness in this unit results in increased efficiency in the transmission of muscular energy to the bicycle drivetrain.
3) The efficiency of this system can and should be optimized.
4) This optimization can be considered as a discrete operation, seperate from the considerations of frame design, other than as they effect the standardization of the design parameters.
Further:
"frame flex" needs to be usefully defined to be considered, and does not occur in nearly as simple a manner as would be readily apparent, especially as regards drivetrain function.
"frame flex in the BB" is essentially non-existant, if by BB we are referring to the BB shell. If by BB we are referring to an undefined bottom-ish area of the bike, than it does in fact occur, may well be the source of inefficiencies, and it's consideration is of absolutely no value in a discussion of the relative efficiencies of different crank/bb configurations.
EG: "Who cares how stiff my crankset is - I am riding in a 6 foot pool of peanut butter."
"Who cares how stiff my crankset is - I am riding a late 80's Vitus"
"Who cares how stiff my crankset is - It's mounted on a bike I never ride"
All good questions; all also completely irrelevent to the discussion at hand.
.
.
Tech writer/support on this here site. FIST school instructor and certified bike fitter. Formerly at Diamondback Bikes, LeMond Fitness, FSA, TiCycles, etc.
Coaching and bike fit -
http://source-e.net/ Cyclocross blog -
https://crosssports.net/ BJJ instruction -
https://ballardbjj.com/