There is definitely a potential that the study design offered a measurement that wasn’t useful in determining how well these drugs do or do not work. But, in y

There is definitely a potential that the study design offered a measurement that wasn’t useful in determining how well these drugs do or do not work. But, in your example, you measured the length of time you put into playing the piano, something the study did address and note that people on the drugs did do—by increasing their time on task. The important part though was that while taking longer, the participants did not perform any better when on the actual drugs and worse than those on the placebos. So, following this logic in your own claims, it’s possible that while you played for 8 hours, it wasn’t your best work.