As an aside, I am amazed that chairs can be sold without some redundancy, in case of breakage. 350 pounds is not exceptional these days, that is only 160Kg, which is only 90 Kg more than what I weigh with clothes, cup of tea and plate of food... 80Kg is reasonable for a healthy person and two people should be supportable by the chair.

However, chairs get left outside and they can rot. This can lead to collapse even if a 50Kg person gets on. This can be incredibly dangerous. Hence every chair should have a secondary support structure, in some cases this can be wire under tension, other times it might have to be steel tubing.

There are thousands of chair designs, none of them built for the ultimate eventuality of catastrophic failure.

No. There's a *big* difference between the load limit printed on something and it's true expected failure point. As somebody could be hurt by the chair failing I would expect the real strength to be in the ballpark of 3,000 pounds or even more. Figuring the permissible % of expected yield point is handled by the engineers, normally not by the end users.

Load ratings near the failure point are only done when the failure will not cause a problem, or when the failure is actually a desirable property (breakaway tethers of various types.)