Do we really need to go through all this again?
As you can see
here, the properties of this kind of system really do change over time—more rapidly at first, less as time goes on. InnerFidelity had an
article on this (referenced in an above post), showing the effect for headphones, the Q701 (do note that the listening test is pretty much a garbage invalid comparison that does not demonstrate what is claimed, so it is not evidence one way or the other). There is a very small change in the response and distortion characteristics.
The effect of headphone placement on the head, pads wearing over time, etc. may be significantly greater.
The effect of people getting used to a headphone's sound is probably even greater than all of the above.
It doesn't seem like peoples' actual claims of break-in effects are substantiated by any of the evidence, because they claim differences much larger than seem to be apparent, types of differences that probably do not occur. As pointed out, it is suspicious that the majority of reports indicate that headphones sound better rather than worse, but this does not by itself rule out the possibility that there is a real (and perceived as positive) change. It's all the other evidence. That said, loosening the mechanical damping by breaking in the elastic materials may have an effect not that different than increasing electrical damping factor, which in theory should be generally positive for most headphones—though any changes are realistically probably still in the "hardly different at all" or "can't tell a difference" ranges.
As for the OP's scenario, break in however you want, as long as you want (0 hours is an option), whatever floats your boat. Just don't use deafeningly loud levels, like above or close to the max power input handling specified by the manufacturer. If it's some reasonable-quality headphone and the level is high enough such that you can stand to keep it on your head, it should be okay.