That's a good point. From one perspective, it looked like the direction was obvious, but I don't think anyone here would plan a fall from a single photograph. It's a hindsight issue...it's easy to tell which way the lean is once the tree is falling.
Now that things have settled, I'm going to have another go at what I was trying to do in the first place. If it causes a stink, I hope someone will be able to pick something useful out of the ensuing mess.
I have nothing new to offer, but if there is something here that someone has not heard before, it is worth posting. This is just the way I do things, and I hope someone with more experience will offer some alternatives. There may be a few people on this site that have nothing to learn about falling, but I'm not one of them.
A big problem here was cutting through the holding wood. One way to avoid this is to dog-in where you want the back cut to end (or farther ahead for larger trees, with a "reset" once the first part of the backcut is parallel to the face).
For the sake of discussion, let's assume the lean was known, and this was a fall 90 degrees from that lean. from the description, the diameter of the tree made it necessary to do the backcut in 2 parts. Also from my interpretation (because the holding wood was severed), it seems the first part of the backcut was made on the side away from the lean. The lean was misread, so it wasn't a conscious decision, but the end result brings up a useful point when the lean is recognized.
In my view, the first part of the backcut should be on the side of the lean, so you have maximum holding against the lean. This should go a long way towards avoiding getting the bar pinched. I would start the second part of the backcut at a dogged in position maybe 45 degrees from being parallel to the face. The idea is to taper the holding wood so that it is thickest opposite the lean to swing the tree towards the face. If the tree doesn't start to go, then continue deepening the back cut so that the holding wood remains tapered.
In some cases, I have experimented with putting a Dutchman on the lean side, which closes as the tree shifts stopping the leanward movement of the tree if there is sufficient holding wood opposite the lean. This is supposed to shift the weight towards the face. My question is, would having a Dutchman in place make it more likely or less likely for the bar to get pinched while making the backcut?
I realize that all of the above is painfully obvious, but I'll gladly risk being obvious on the off chance that someone is not familiar with it.