The definition of this kind of award is vague enough so that half of the voters look at it one wrong way, and the other half look at it the other wrong way.
Olympic Fan pointed out in an earlier thread that there is nothing that requires a voter to look only at conference play, which means the whole regular season could be considered. Considering that the ACC named a Player of the Week and a Freshman of the Week as early as November 11, I'd say there's something to that.
Then there's the "best" vs. "most valuable" thing. Even if you prefer one over the other, neither is clearly defined. If there is no one best player, do you just pick the best player on the best team? What if there is a best player who plays on a poor team? Is it more pure to try to imagine which player's absence would most affect a team? And if so, how the Chapel Hill do you do that?
Finally, there's the supporting evidence. Are you a statistics person, or do you rely on the eye test? (Note: the eye test is old analyst-speak for "Get those numbers away from me, Poindexter. I know what I saw.")
Let me try: Entire regular season. Most valuable. His absence makes the team a complete wreck. Eye test. Congratulations to my ACC Player of the Year: PJ Hairston.