Hi Allen,
I've not seen this knot ...
I thought of proposing the use in "reverse" of this knot illustrated by roo on its website, in the wake of the knot shown in the video linked by Tony(asemery)some posts before,for be understood as an attempt to achieve a compromise between the behavior of the knot in the proposed video, and what subsequently explicated by by knot4u in regard to the greater simplicity of the mechanism and the most facilitates regarding the untying of some friction hitches cited by him.
I must say though, that,experimenting the use,in this way,of the original HFP Slippery 8 Loop with a rope of small diameter (2.5 mm polyester),that also this knot can jam, and also sometimes is reversed in a unexpectedly way, trapping the tail between the wrapped object and the wrapping loop,making it dramatically difficult to untie(I think I was able to remedy, at least in part, to these defects,by doubling, during the execution of the Figure of Eight,one of the two loops that make it up,and that is,that envelops the Standing Part; by making this amendment, or by running a slipped version of the original Figure of Eight, thus transforming the final product in a quick release knot,the above problem is not presented more).
The photo mentioned above shows how the buntline holds.
I must say that I agree with this: after reading your post, I too have seen that the Buntline Hitch behaves in much the same way: by the immediate fixed tightening on the object with a single pull,and by leaving the possibility to re-enlarge the loop by pulling on the other end of the rope, a little bit as with the Canadian-Arbor knot shown in the video from Youtube.but even in the case of the use of Buntline Hitch, there remains the problem that the knot can tighten too, preventing the fact that the realease mechanism, which in fact in my opinion still inside it contains, can remain practicable!
Bye!