One of the reasons for the observations of THCV could be that there was just very little THCV in the sample and it mostly evaporated already before reaching the boiling point, giving a peak lower than the BP.I am familiar with that video, it's awesome content and good to see again. But I guess it leaves more questions than answers considering the metrics are so wildly different in the results that S&B found. I suspect this could simply come down to thermocouple location and/or the way the display temperature is measured versus the air temperature, but there's probably plenty of additional variables as well.
Few observations though: THCV is falling way off on the cannabinoid chart which is intrguing as most sources cite THCV as having one of the highest boiling point cannabinoids, at 428F. So its interesting to see it fall off before D9-THC?
Another interesting one is how much terpenes are destroyed by starting at a hotter bag. A lot of people mention trying to cache a bowl with various vapes in one hit under the premise of "getting all the terps" but you can see how having a vape hot enough to crush a bowl in one hit will realistically destroy a lot of the flavor compounds.
There is also a possibility of boiling points changing somewhat in a substance mix.