OK, for anyone who is interested in the data on grip strength (or at least one internet dude's data*), I got out my grip dyno. First, I measured a "baseline" grip strength using the 3 non-index fingers on my right hand, to approximate gripping force available when running a gun with a normal strong-hand grip; my index finger was held out straight.
Then I held both the index and middle finger off the grip, to simulate a middle-finger-on-trigger grip. The force I was able to exert was 45% of the 3-finger grip. In other words, I lost more than half of my available grip strength. That's terrible. Giving up a middle finger's gripping power appears to be a terrible idea if you care about recoil control.
Next, in order to consider the case of tiny guns that simply don't permit 3-finger grips at all, I tested a grip holding both the index and little/pinky finger off the grip. This was 50% of the regular 3-finger force. Still terrible, although marginally better than the pinky-and-ring-only grip. So, for pistols that are so small you can't get all 3 non-index fingers on the grip, letting the pinky dangle is about 10% better than moving the index finger up in terms of straight grip strength. That's without counting any recoil control coming from curling the pinky under the grip, which I think would add to the measured advantage of the pinky-drop that uses the middle finger for gripping, not trigger. But that's at least in a zone where maybe for some rare individuals or weird guns maybe, maybe there's a plausible claim that recoil control might be kinda in the same ballpark with the middle finger running the trigger. Maybe.
For any gun where you can get your whole hand on the grip, it is not close. If you care about gripping ability on the gun (which is the primary determinant of recoil control), use all 3 fingers to grip and use the index finger like a sensible primate. Forget this middle finger triggering nonsense.
Note: This data would have no bearing on rifles or shotguns where strong hand grip strength plays little or no role in recoil control. And if you truly didn't care about recoil control with a handgun, this might also be irrelevant.
* I'm an early-40's male with no disability regarding hands or hand strength. I have moderately strong hands for a member of the general populace, and average to maybe slightly above average grip strength for a fairly avid handgun shooter. Others may have different experiences, but I think I am likely to be a reasonable approximation of "typical" for purposes of this experiment.