Made a model too, kinda crude, but it shows that it works.
That simple model can explain it better than I ever could.
and yes, there is a lot of friction judging from how the model moved. I had to relieve tension in the individual legs to make the "truck" move over and over....probably best to have the truck in gear when using this method to help things along.
I don't know if I would do that. If I did use the trucks power, I'd be darned careful not to overrun the winching speed. It will lay the several parts behind the truck on the ground to be driven over. When I did this with the wrecker at work, driving was not needed for the pull, and was prohibited by the need to constantly watch and re-arrange the blocks around the extension cable hooks. And of course with the winch controls being outside of the truck, I would have had to trust one job or the other to the same (esteemed colleague) that drove the truck in there in the first place.
I believe you are familiar with capstan winches, no?
If you add up all of those "half of a wraps" around each screw... I bet it did bind up.
I don't have a little model made up to verify this, but given the diameter of that string, and the diameter of those screws, I would venture to guess that if you had used three eighths larger lag bolts instead of those screws (clean shiny ones, not old rusty ones) that the demo model would probably work on it's own at least well enough for no-load demonstration purposes.