I wouldn't put much stock in my opinion because I have never played Othello, or studied how it might be programmed. Also, I strongly suspect you're a better programmer than I am.
You know the game, and clearly, you know a good deal about programming this game. I'd say take your own intuition on this decision, and go with it.
I'd look at two other things, in deciding:
1). Is there any way that you could make a little test set of positions that it normally fails on? They nearly always use several quick test sets in chess programming, just to see if an idea they have is plausible, before they put a good deal of time into it.
2). What do the best Othello programs do? Not that you want to copy it, but as a general guide to good play, that would be good to consider.
