There has apparently been much discussion and research into the exact positioning of the nails in crucifixion. It is argued that it is not feasible to nail through the palms, as the hands cannot support one’s body weight; there are no structures in the hands to prevent the nails from ripping through the flesh once the cross is put upright. It has been shown that nailing through the wrists is feasible, as is nailing between the radius and ulna.
What about if one is crucified upside-down, as Peter supposedly was? Is a single nail through both feet sufficient to support the body, or would they rip just like the hands? What about one nail through each ankle, binding the feet to the sides of the cross?
In the depictions I’ve seen of Peter’s crucifiction, his hands are generally nailed to the crossbeam, but his feet are just tied and hoisted by rope up the upright. To nail the feet, one would have to start with the entire cross flat on the ground, and then pull the whole thing upright, which strikes me as being rather more difficult.
From what I understand of the standard procedure (with the disclaimer that I Am Not A Roman Soldier): The cross is in two pieces. The upright is planted firmly in the ground, and stays there more or less permanently. It has some sort of winch or hoist attached to it. The crossbar is a separate piece, and is what you’d compel the condemned to carry, if you wanted to humiliate him. Once the poor sod has schleped the crossbar to whereever you keep the upright, you nail or otherwise secure him to the crossbar, on the ground. Then, you tie the winch-rope to the crossbar, and hoist it up onto the upright. For a Petran-style execution, the only change needed would be, instead of tying to the crossbar, you tie to the condemned’s feet.