I would advise you to go back to your course components on non-parametric and Baysean statistics.
Because a study isn't ideal (and I am not saying that small studies like this are ideal) does not mean that they are devoid of useful information. If you don't understand statistics well enough to use that information, then that's fine. But your incomprehension does not give you the right to go around accusing other people of fraud in a public place.
That sounds like I would say if I was running a fraudulent study.
Wow, it's up to 200% effective!?!?!?!?!
That's the level of accuracy we are dealing with here folks.
I'm not so sure. I'm thinking that Hanlon's Razor [wikipedia.org] should be applied here.
The corollary? Nothing is provable; everything is permissible.
Your adage is one of a pushover. You are a sucker, and you promote being one.
If you have a procedure with 10 parameters, you probably missed some.