This is more than a "oh, it depends on your metagame comment" this is an attempt to gauge how much we need to take into account decks like Sui, Sligh, and the like. Fact is, I am more likely to face Trinistax or some other super expensive deck given my playtest partners than I am to face Sui. What do we do? Prepare for the ideal or realize we play in the real world? Settle this issue and I think it will be clear how many MisDs to play.
I don't think this is really the issue in this particular thread. I'm not going to completely disregard the 'regionalizing' of a deck for a particular metagame as a useful strategy. However, we're not building a deck for a tournament, we're trying to optimize GAT for a general (what Ric calls 'ideal') metagame. Rico already pointed out that the top non-control decks, even the current 'sligh' deck, FCG, are not effected by misdirection. Since only one or two of the top decks (Hulk & Landstill) are worth running misdirection against, you have to decide whether you want to have a good game 1 against these, or run something else (stifle, duress, annull) that has a more broad application.
So since non-control decks don't care about misdirection, and since in larger 'general' metagames control doesn't dominate like it used to, where does that leave us? You would think it means that misdirection is an obvious cut. Ok, then why do two of our leading GAT players (Ultima and Dave) argue so vehemently for its inclusion?
When I've run misdirection in high numbers main, its been because I thought enough control decks were competitive so that I wanted to make sure I win game 1, even at the expense of other matchups. I'm not sure if you can make the case for that right now. However, given the amount of hype that Hulk has gotten recently, perhaps they are worth it.
Although I really like some things about Ultima's build, he played against SO many control decks in NJ, that the misdirections couldn't help but be good. Not only do they force through early spells, but they cause the opponent to be reluctant when casting their best control match spells: ancestral recall and deep analysis.
A counter argument to this could be that since Ultima's list has so many ways to get rid of dead cards (brainstorm, FoW, thirst for knowledge), he can afford to run a deck highly metagamed against control. Perhaps this is how GAT should be built and played.