I think that tuck shrinking is not a good way to get the acute corners needed for an aero hitch box, but I might as well share what I learned about the technique. It was fun.
1) The tucking tool needs to have parallel rods with a spacing of about one half to one quarter of the rod diameter in between. Too much spacing, and the tucks are too oblique to result in a shrink. Too little spacing, and it's difficult to form the tucks without unduly stressing the sheet. I found this relationship experimentally with the rods under a clamp, but I couldn't clamp them tight enough so then I welded them up once I had them right.
2) A decent tuck looks like this; an acute angle between the sides of the tuck, and no sharp corners, because sharp corners will get worked into tears.
3) Unlike the stump method, start at the inside and work out. In other words, start with the longest tucks, and then work toward shorter tucks. Once you do shorter tucks, you get compound curvature that prevents you from doing long tucks, so once you move to shorter tucks you loose your opportunity to add more curvature on the inside.
4) This method with 0.032" aluminum is very time consuming. It can take 20 minutes to make one pass of 8 or so tuck/shrinks. Then you have to repeat that process 2-5 times before you can proceed to shorter tucks.
5) I succeeded in annealing with a propane (plumbing) torch, using a sharpie marker as the temperature indicator. The thing is that I'm not sure when is the best time to apply annealing and if it actually forestalls tearing long enough to be worthwhile. If I did it again, I would probably try annealing once for every two tuck shrink passes. Annealing is a 10-15 minute process for this piece. A) Preheat the propane bottle because my shop was cold, you get a lot more power with the bottle at 75°F than 50°F, just because the propane pressure is higher at that temperature. B) Preheat the area, keeping the flame moving across the whole area, for about 2-4 minutes. C) Dwell across a small area until the marker disappears. BTW, I don't believe that my little propane torch has enough power to melt aluminum of this thickness, but if I go too far it does seem to get a faint golden oxide layer of some sort.
EDIT: I'll double-check, but I think I'm using a 5000 series aluminum, which may not be ideal for this. If I did this again I would do some more research into the most 'shrinkable' and 'anneal-able' alloy.