I realize there has been a lot stress testing on the synthetic ropes and soft shackles but I have not seen destructive test results on the actual rings . With a bow shackle, the bow will begin to deform under excessive loads before failing. Theoretically in a 4:1 systems with a 10K winch you could see forces of ~40,000 lbf and with a 16.5 ton winch, near 70,000 lbf before the winch stalls. Granted these forces are very rare and other things in the system may fail first but what are the ring's failure characteristics. Do they split, separate, fragment? Is it a gradual degradation or an immediate catastrophic failure?
The first thing you will see is some permanent deformation (elongation) of the ring. That is easy to check for with a set of calipers. The 6061-T6 material I use in my rings will show visible (plastic) deformation well before an ultimate failure above 100klbs.
It's easy to add the numbers up and get a bit concerned, but I have yet to see ANY ring failures across the industry. There is always a limit, but the practicality of being able to anchor these forces is a bigger concern to me. There are very few natural anchors or connections to a vehicle that would approach the ultimate failure of the ring.
On the technical side of things, Only with double rigging, 1 or 2 legs are going to be trying to stretch the ring from the inside. The other two legs around the outside have far less mechanical advantage on the ring.
It would be impossible to ultimately fail my recovery ring with 3/8 or even 7/16" good quality winch line. You just can't generate enough force before the winch line would fail ( and we shouldn't be rigging anywhere near to ultimate failure on the winch line in good practice). You should absolutely be thinking about the doubling up the soft shackle if you are rigging a 3+:1 mechanical advantage system however, and spreading that load out to multiple anchors and recovery points.
I wouldn't recommend double rigging my ring with a 16.5 TON winch , but I think that was a typo.