In modern science, an idea or hypothesis must be tested and ideally produce predictive value to eventually be elevated to a theory.

“It's the ones, that say, if this idea is correct, you should find this, if you look in that direction, and then we do it, we find it, hey, you're onto something. Give me another prediction. Oops, that failed, okay. Should you modify your hypothesis, and by the way, if your predictions keep coming ri”

Crossover designs in human trials are valuable because they increase statistical power and require fewer participants, though they carry risks of carryover effects.

“there's a real statistical power and I don't use that in the beta sense of the word but a statistical gift that comes from being able to do a crossover in that you can now leverage a student t test for example as a very powerful statistical tool that allows you to use fewer subjects”

The three most critical factors for assessing science are the collected data, the methodology used, and the logical framework connecting them, while other elements are tangential.

“in science three things matter. The data, the methods used to collect the data which give them their probitative value which shows what they mean and the logic connecting the data to conclusions.”