I'd file this under "systematic measurement error", and the team specifically said that something like it was the most likely cause for the results. They also said that if neutrinos really were FTL, we'd have observed this effect much earlier (for example when we started measuring supernova radiation bursts). So, the article's hypothesis is a very good candidate for a real flaw in the experiment's design.
The scientists probably didn't take this into account because they assumed their setup already compensated for all kinds of clock drift... I wouldn't call this an epic failure just yet, because even if this turns out to be the root cause it's a pretty standard type of mistake as far as science experiments go. That's why we have independent reviews, and that's why they are necessary.
They also said that if neutrinos really were FTL, we'd have observed this effect much earlier (for example when we started measuring supernova radiation bursts).
I am seriously doubtful that these neutrinos are traveling faster than light, but I've also been skeptical of this argument since it was presented. It would be pretty hard to have observed this if they weren't looking for it because they would have to sift through old data. For example, for SN 1987A the FTL neutrino event would have preceded the visible light neutrino event by over four years.
They could go through all of the old data, to be sure, but they would have to specifically have had to be looking for something which they think fundamentally is impossible in order to find something like this. So, if these events were even recorded at all, it's certainly not clear to me that they would have done this (at the time of the FTL announcement, that is- of course, they should be doing it now).
it's a pretty standard type of mistake as far as science experiments go
Yes, but when one makes it this far it is big news. No matter what I give props to the scientists to publicly invite the scrutiny they did. That took balls.
The scientists probably didn't take this into account because they assumed their setup already compensated for all kinds of clock drift... I wouldn't call this an epic failure just yet, because even if this turns out to be the root cause it's a pretty standard type of mistake as far as science experiments go. That's why we have independent reviews, and that's why they are necessary.