Yeah my tool is always used on a "roughed out inbetween" (even just a point, or two points if you rotate/scale). Sometimes I don't draw my inbetweens, I just click with my tool where i want my inbetween to be then on the two lighttable images and everything is at the right spot.
I understand what you mean, with paper you line up your two drawing and put your current sheet in the middle where you want it. But in tvp you have to align lighttable drawings on the current current image, not the other way around. Also if you want to calculate any position from the position of your out of peg, you would still have the problem of modifying your drawing that is not in place. But all you need is to use your current drawing as the "guide" to move the out of peg drawing, not your already existing drawings, it doesn't take much more time (in harmony I also work the same way as in tvp, I don't move current drawing usually).
I think it's far easier to visualise where the inbetween of one point will be (for example the nose) and just click on this spot, drawn or not, then on the nose on the two surrounding images
same for scale and rotate with a second reference point. And usually when I draw the nose out of peg, I can better see the arc for the ear for exemple, so I use my script again and I out of peg the ear directly cause it takes into account the already set out of pegs (you can do that for each part if you want to be precise, beginning with the most obvious reference points for your spacing like the clavicle then the chin, then the inside face details).
(you can do this without drawing the rough nose, just click on the right spot)
I actually never found useful to use the peg visualisation, I just directly place the drawings at the right position in the spacing.
Btw if you have suggestions on my OOP script feel free to tell me
on my side, if I take care of ending the flip properly with the button, I never have bugs or problems (except some undo weird behaviours, or if you accidentally flip normally or go to another layer, but you can always undo, and be sure to not use the gradient mode). Personally, I may be biased but I can't use the LT OOP anymore and always use my script for practically all my inbetweens cause it's much more precise and faster to use. I also agree with the center of rotation/scale that NEEDS to be movable, but my script is a nice workaround so right now i'm fine...
I thought at first of being able to flip and out of peg more drawings but actually I caught myself almost always using only two surrounding drawings.
HOWEVER i'd be happy to have a new alternate function for the lighttable that would register out of peg per image/instance position instead of lighttable position (like in toonboom), using the id of the instances. Could be useful to animate things like walks on the spot and be able to flip between drawings without having to reposition everything everytime we change to another drawing (you can do that by stacking multiple paper when doing traditional animation for exemple, but not in tvp).
I thought about scripting it (by adapting the out of peg/drawing when you move to another drawing with a special flip button) but it would be really limited or really complicated.