I think it would be a better comparison if somebody would do a test with more control, (you could do it even) for example instead of swinging a hammer to hammer the knife through a metal pipe, drop say a 10 lb weight from directly above (a fixed position) it from exactly 10in high every time, controlling angle and force. For the horizontal load stress test (where he stands/jumps on the blade stuck in the wall) would be much more informational to load it up with weight, stuck in the wall/vice the same amount of force, you would also be able to calculate the force on the knife (there will be more force on the knife if the 50 lbs weight is on the end of the handle than if its at the base of the blade, etc. Of course it would be a lot better if the sample size is larger than one, since the knife being tested could be in the top 1% of knives made (strongest, best heat treat) or it could be in the bottom 1% (poor heat treat, bubble in the steel, etc) Imagine if the busse tested in knife tests was in the lower 10% of that busse model, and the average busse in that model would have done significantly better.