Is dirty power the culprit?

Are you playing the same source material when comparing?

Seems like an obvious thing but, if not, then this is a likely culprit. Not all recordings are created equal.

Mood/disposition towards listening is another likely factor.

Power? Maybe. That might be hard to prove though.

DOn't some power conditioning devices provide displays that indicate a measure of power "cleanliness" both in and out? If so, then maybe this could provide some real evidence that power is the culprit.

Also, any electronic device can have intermittent problems that prevent them from operating to spec at all times. Intermittent problems are the hardest to detect and reolve sometimes. You have to be able to swap pieces in and out one by one over a period of time and listen and compare to isolate a problem.

Simplicity can be your friend in resolving these kinds of issues. The more components in the signal chain, the more places there are where something can go wrong. Electronic gadgets are not perfect and do not always perform up to spec forever after all.

So if the problem is indeed dirty/variable quality power, aren't there meters or other gadgets out there that can measure and confirm this rather than guess or assume and possibly avoid a (potentially expensive) wild goose chase?

Leaving the amp on is a good idea for a test to see if that makes things sound more consistent. Problem is it is not practical with tubes on a regular basis because you will shorten tube life. Works good with SS amps though. I do it all the time.