I posted in the thread that I assume spurred this one. Knot-choice (and knot advocacy) is something to be extremely careful about.
You need to be able to consistently and easily inspect any knot that you use (whether you tie it or not) and reliably tie it correctly. You also
must test the heck out of it at ground level.
This is why the Figure 8 on a bight/figure-8 retrace is the go-to, with an ample tail (6 in plus) and optionally a backup knot. I use the backup knot primarily as an extra/easy way to check and confirm that I have plenty of tail, and increase the security of the knot against loosening when tied, especially in stiffer rope like HTP that requires a much firmer dressing to really set the knot. It's a strong, reliable, easy to tie and inspect knot.
I personally believe that telling people to tie a figure 8 follow through with 6 in of tail on their tether (and link to proper instructions!) is more productive from a safety perspective than just telling them to buy a manufactured/spliced set of lines - simply because there's a better chance at getting people to do something that doesn't cost them extra money and wait time, vs. blowing 60-100 bucks on a pair of ropes.
I don't know of any cases where people have screwed up this knot...which I CANNOT say for the following ones.
The next knot that's non-negotiable is a stopper knot. Normally a double-overhand with plenty of tail.
https://www.animatedknots.com/double-overhand-stopper-knot Especially on a stiff rope like HTP...you need to set, re-set, and inspect the heck outta this. It'll save your life, but only if it's tied and secure.
I use and prefer scaffold/poacher knots on my hitches, and have generally preferred a distel hitch.
I tie the distel first, with extra length in the cord. Make damn sure that there's a secure stopper knot on my tether. I test the heck out of it. I weight it slow, weight it fast. Weight it heavy, weight it light. Weight it with tension on the tag, and without. Weight it with a dry line, and a wet one.
This is important. I've had hitches that I considered secure at the time. Then conditions shifted slightly, and my stopper knot saved me from a serious fall.
10000x if you are messing around with self-tied prusiks, alternative hitches, etc...do not skimp on testing!
poacher/scaffold I tie directly around my carabiners (
https://www.animatedknots.com/poachers-knot), with an over-long section of cord. I'll weight it to set the hitch and test the knot, then loosen the knot, slide it up closer to the hitch, move the excess cord through, redress the knot, and reset/retest it. I test like this until I have the carabiner-to-hitch cords as short as possible (this helps to keep the distel dressed -
a poorly dressed hitch will not grab - and also improves performance with a tender), until I've thoroughly tested every conceivable scenario. Then I cut the ends, leaving long tails, and test some more.
Note: there are alternative ways to tie the poacher and scaffold, that can result in a
SLIP KNOT rather than one of the most secure end loops. Do not screw this knot up.
The last knot that I use is an overhand on a bight for webbing bridge. Keeping a long tail is critical on this as it will loosen under cyclic loading. Again a barrel backup is useful primarily for enforcing that you always have this tail.