The scenario we're positing is that perhaps, in fully-solved chess, white is actually in zugzwang, where every initial move is actually bad for them.

If there were some way for white to hand the first move to black, then clearly this would be the best solution.

But all the take-back solutions (e.g. move the knight forward and back again) actually give black two moves, which is a different situation.

And the stutter-step solution of white moving a pawn only one space instead of two doesn't work because that's simply refuting the premise that white is in zugzwang.