That's fair, and it does prove that different knots affect strength differently. That's a base principle that is useful to know even if it doesn't get you far on its own.
With regards to your application: while it lacks the "finesse" of tying knots that give you just enough strength, simply getting stronger twine would solve this problem without the complexity.
In general, there's a whole lot of ways knots can fail. If you can solve the problem by just using stronger cord, that frees up mental cycles for dressing, avoiding slippage, effective communication, ease of untying, etc. If you're just tying up boxes do whatever you feel like doing. But for climbing, I'd rather have the rope be more than strong enough to do its job, and have those brain cycles for focusing on preventing other system failures.
I'll admit that I've mostly used this "useless knot" test methodology you're criticizing here. But in its defense, it is an easy to do micro-benchmark with consistent results.
I guess I could setup a more cohesive test to measure the strength in more "realistic" scenarios. That's really all it comes down to: the strength of the rope in whatever specific situation you're doing.
I'll note that 40lb-strength rated cotton twine is just strong enough to be useful, and just weak enough that a bad knot or two will cause the twine to snap under moderate usage. (Ex: Package Tie to support a box containing 20lb).