The test was interesting but testing to failure has little to do with bows. Failure is starting to occur when the bow starts taking set. Just because the hickory broke forst does not mean it was the first to fail. The white oak may have crushed long before the hickory actually broke. Bubby made a bow of hickory backed paduak. The bow came out nice because he used the wood within its limits. P aduak is known to chrysal if overstrained. So I would say testing to destruction won't give you anything useful. Don't feel bad I used to do it all the time LOL.
I've never been able to build a bow that hasn't taken set... But then, I had been under the impression that as long as the set wasn't excessive to the point that it slowed the bow down to a point of being an unacceptable shooter, then it really doesn' harm the bow. I mean, isn't someone's quote: "set happens..."
Sure, our goal is to avoid as much set as possible - we heat treat, glue in reflex, etc., but if we're looking to make a bow without any set, aren't we striving for that which is unobtainable?
And for that matter, what does set really tell us? Was some aspect of the wood itself being "not right?" Was it a poor design? Was it a poor execution of construction? Or is it just something that happens?