I’ve not watched the video and I wasn’t particularly going to bother but I’m not sure about that. I never thought trees were unequivocally safe anymore than I’m about to accept they’re unequivocally not.
We know that when people start to take avalanche training there’s a statistical blip where they become involved in more incidents (The Role of Training in Recreational Avalanche Accidents in the United States, Ian McCammon, ISSW 2000 etc) and there’s a variety of reasons offered for that but one of the main ones, in my opinion, is related to making these sort of unequivocal statements. At these earlier stages of experience there’s a tendency, hardwired by evolution, to try and formulate heuristics or rules of thumb. I think this occurs because people are naturally struggling to to assimilate large amounts of new information and trying to understand how to apply that new knowledge. So, they establish these heuristics which are sometimes wrong, but more normally incomplete, and then later entirely discard them. When they’re discarded it’s no safer, if the heuristic was incomplete it probably did reflect something of value which is now also discarded.
Sorry, that’s longwinded of saying that trees, their placement and density on a slope is just one piece of information, it’s a characteristic of that slope and one that needs throwing in a mix of other data. The comparative weight you might give that one piece of information is going to vary on the day. It’s also an integral part of several decision support frameworks in the form of avalanche paths of course with the obvious paths stripped of trees. Again analysis of incidents says that at the early-acquired skill level missing obvious clues like avalanche paths is quite common.
Which is still a long winded way of saying that we ought to avoid things like “XXXX is a myth” if it’s really a case that “XXX isn’t always true and has these limitations”