Doesn't that message support the reading "drat, I can't filter out enough noise to make it statistically significant, so I don't have anything to publish"? (Genuine question.)
Publish or perish. This all sounds very normal to me. I don't see references to making data fit, rather it sounds as if he is looking for a ways to extract some useful information from it.
Anyway, you know what: people believe what they want to believe. This discussion is completely pointless.
