Not anymore scientific than driving different cars around and comparing their performance to each other, which includes raw data (0-60 time, horsepower, braking distance) and subjective data (this seat feels more comfortable, this car feels higher/lower quality, ect.) and then ranking which cars are the "best". If you want to do Science, you have to follow the scientific method, and that ain't exactly the Scientific Method. There are so many unaccounted for variables that can effect blade performance that it would be impossible to definitively verify which swords cut best simply by cutting up criminals. It's ultimately an artificial scenario that is not all that similar to real-life when you think about it; when swords were actually used in battle they were often subject to stresses unrelated to just cutting naked flesh and also had to perform under harsh conditions, variable weather/temperatures and less than ideal maintenance.
Besides, by the time Yamada Asaemon put the list together in the early 1800's, Japan had been at peace for 200+ years and swords were more symbolic than practical weapons of war. Smiths had essentially forgotten how to make blades like those from the times of the Sengoku Jidai and before... that's why we distinguish between Koto and Shinto Nihonto, after all.
Not saying it's totally useless information (just like car tests/reviews), it's not the "be all, end all" to determine which swords are "best". It's still cool to have a sword from a smith that was rated Saijo o wazamono and it certainly vastly increases a blade's value in today's market.