Please excuse my rather "raw" knowledge of knot terminology. Definition of "cuckold" and "collar" noted.
Excused. In any case, knots nomenclature--both of how to speak of parts of a knot,
of steps in tying, ... , and of names for knots (and how many knots for each name !)
is a royally confused area! So, welcome to the fray.
"Collar" comes (to me) from Charles Warner (author,
A Fresh Approach to Knotting
and Ropework) and seems an apt name for those lobes formed in Ashley's #1452
and Carrick Bend and others, and of in the Bowline.
As regards strength, still, I like the idea of wringing out every ounce of it (if only for psychological reasons).
Though one doesn't buy climbing ropes based on strength (as that usually isn't specified).
And the pertinent
strength might not be what is measured by slow-pull testing,
but by some other method. !? E.g., as I've noted above, I wonder at whether the bowline's
non-jamming behavior might result in a movement of rope back'n'forth between loaded
and relaxed states such that it garners more chafing wear than is desireable!? The test
for this, I think--and suggested--, is to do a series of (reasonably loaded, not super heavy)
of drop tests, and to check for such wear. (Sounds like "sport climbing", yes?!)
More practically, I would conjecture a knot of greater strength would impart less wear and tear on a rope,
although your point may also apply here to a fair degree.
Yes, I've wondered as much. AND whether the use of a variety of tie-in
knots--presumably thereby putting hard-bend stresses at various points--,
would be a help. But the likely hard-wear spot of climbing-rope ends (which
leads to some users buying long ropes and chopping ends maybe twice in
the course of the rope's life qua
lead rope (before retiring to TR use))
comes at the hard bend over the top 'biner, not at the knot so much (or that
this is another point of wear, not so easily varied by knot variance).
--dl*
====