Happens to the best of us, I guess.
It’s funny how even the silly mistakes we make can lead into good discussion and new insights.
So like eating nightshades while popping a few pills of Tylenol? Sure, that could happen. But I think you can still do a thorough scientific investigation and get reliable results at the end of the day (although that might be a hell of long day).
landis seems to be saying that he gets migraines. He suspects that nightshades might be the culprit. There is a very simple test to see if there is a connection: go a few months without nightshades and see if the migraines disappear. If there is a confounding variable that happens to come up during these few months that might be having the effect of cancelling his migraines (ex. he, for some bizarre reason, decided to start a habit of popping T3s every morning at around the same time the experiment started–and this by sheer coincidence, landis not making the connection between that and the onset of the two months of his experiment), then we’ll find out as soon as that confounding variable disappears. Once it does, and his migraines come back, then he’ll know that a nightshade free diet does not cure his migraines and he’ll know to look for another cause. But if the confounding variable doesn’t ever disappear, well, what’s the problem? He may be wrong to conclude that his migraines are caused by the nightshades, but he’s free of migraines. So problem solved anyway.