....What MAY be significant, however, is that the cue tip is likely to deviate significantly from a straight-through path when the shooter twists his wrist. This could significantly alter the direction of the force applied to the cue ball, and might well explain the success that some have claimed for the technique, as jalapus and JoeyA describe above. This is also consistent with Earl's description of "swiping" the cue ball when applying maximum English.
I have some doubts and think the simplest argument against it is based on the effective tip offset, as distinguished from the geometric or apparent offset, and the coefficient of friction between tip and ball. When applying english (sidespin), it's the spin/speed ratio that mainly determines the "action" off a cushion. But this ratio is directly a function of the effective offset, which is limited by the coefficient of friction (COF) between tip and ball. If you can push closer to that limit by swiping, you should be able to do the same by just lining up farther out on the cueball and executing a normal stroke.
Some time ago it was surmised by some of us that the geometric miscue limit is greater for a high squirt cue than a low squirt one, based on the same argument. They both are set by the maximum effective offset, which, if given the same tips, chalk, etc, should be the same for each type of cue. At least, it seemed plausible. This is apparently true, or true enough, according to a comparison done by Dr. Dave here:
http://billiards.colostate.edu/high_speed_videos/new/HSVB-47.htm
Can swooping possibly raise the maximum COF? It doesn't seem likely, but....
Another argument against can be made based on how fast you have to get the tip going sideways before contact just to make a significant dent in squirt when striking at a large offset. I'm a little rusty on the exact numbers, but I believe you'd have to end up pushing the butt more sideways than forward during your business stroke in order to generate enough lateral tip movement to make a major reduction (at typical cueball speeds). And this was based on a hypothetical where the center of mass of the cue is moving forward (aligned with the long axis of the cue) at the moment of impact, although the cue is still rotating about a vertical axis. I think it's more realistic to assume that the center of mass would be moving in the opposite (lateral) direction of the tip. This probably makes matters worse (further reduces the pre-impact angular momentum of the stick about the cueball's center, for instance), but the net effect is still fuzzy in my mind.
The point is that if you were to achieve nothing else beyond significant squirt reduction, which itself seems problematic, you're not going to generate that much more spin/speed, around 4-6%, and that only if you managed to completely eliminate it (squirt).
Jim
I see Patrick has beaten me to the punch again...drats! (but well said)