John,
I totally disagree with your opinions.
I agree that the pull of the club handle is directed inside-left after impact, which means that it is obviously happening in a through-the-screen manner (= 3rd dimension). However, I cannot perceive that one can be “holding shaft flex” in that 3rd dimension if the clubshaft has bypassed the lead arm from an angular rotational perspective and if the clubhead end of the club is ahead of the club handle (where the pull is being exerted) in that 3rd dimension.
Imagine a child pulling a toy car along level ground using a piece of string. If the string is taut, then that is equivalent to the concept of lag tension. If the car moved faster than the child’s hand pulling the car, then the string would be lax, and not taut, and lag tension could not exist. If the child pulled the car up an inclined path that is angled 45 degrees (relative to level ground) then the same principle applies. The string would only be taut (implying the presence of lag tension) if the child’s hand (holding the string) was ahead of the toy car. If the toy car gained momentum (for some reason) that allowed it to travel faster than the child’s hand up the inclined plane, then the string would become lax and lag tension would not be present. The same principle applies to the clubshaft that is moving up-and-inside along the ~45 degree angle of the swingplane in the early followthrough. If the clubshaft gains a lot of momentum through impact so that it travels faster than the lead hand during the early followthrough so that the clubhead end of the club is ahead of the club handle in the third dimensional inclined plane of the swingplane, then lag tension cannot exist. That is what I believe is happening in those early followthrough capture images that you posted of Knudson and Hogan.
The clubshaft was traveling faster than their leading arm/hand through impact (from an angular rotational perspective) and it bypassed their lead arm (from an angular rotational perspective). Some biomechanical phenomenon had to happen to make it biomechanically possible for the clubshaft to bypass the lead arm as a result of its gained momentum through impact - either lead wrist bending and/or lead forearm counterclockwise rotation. If you look at Knudson’ lower radial bone in his lower lead forearm it does not look like it has rotated counterclockwise, but his lead wrist is obviously bending. In Hogan’s image, one can clearly see that his lower radial bone in his lower lead forearm has rotated a lot counterclockwise between impact and that early followthrough image, and he may have little lead wrist bending happening during that early followthrough time period. However, in both of those two pro golfers, the clubhead end of the clubshaft is traveling faster than the lead hand holding the club handle end of the club - as perceived from the perspective of angular velocity - and therefore I cannot envisage how lag tension can be present in those capture images.
Addendum added later:
Here is a copy of John Erickson’s image showing how he holds shaft flex pre-impact and post-impact.
Here is an enlarged image of the early followthrough time period.
Note that his hands are ahead of the clubhead at impact and also during his early followthrough as perceived from an angular rotational perspective where the clubshaft is moving up-and-inside parallel to the swingplane in the early post-impact time period. Because his hands are ahead and because the clubshaft has not bypassed his lead arm from an angular rotational perspective, the conditions for a “holding shaft flex” scenario are being met. Note that he does not allow his lead wrist to rapidly breakdown (extend) and he does not allow his lead forearm to rapidly rotate counterclockwise in his early followthrough and that prevents the clubshaft from bypassing his lead arm, which will cause the clubhead end of the club to get ahead of his hands from an angular rotational perspective.
Those conditions are not being met in those early followthrough capture images of Knudson/Hogan that John posted.
Jeff.