Untested BSG Loyalty Variant

Hello loyal Signpost-readers,
I know I owe you some session reports, reviews, and reactions on laptop gaming, but instead I'm presenting you with an untested variant I came up with for Battlestar Galactica.
Let me know what you think!
(cross-posted to boardgamegeek)

Note: I have not thought deeply about the full implications of this variant, but it seemed interesting enough to share.

This is just a random idea I had for a loyalty variant which focuses on the probability of players being a Cylon rather than a hard and fast number in the Loyalty Deck. Adjust paranoia accordingly.

It does require one person to be an impartial moderator with the ability to obscure die rolls and card draws, so maybe it's best for PBF games.
But you could work out a system for a face-to-face game with an especially patient moderator.

Here we go:

At the beginning of the game, shuffle the "You Are A Cylon" cards. If you're playing with Personal Goals/Final Five, then shuffle all of the You Are Not A Cylon (YANAC) cards together in a separate pile (otherwise it doesn't matter if you shuffle, since all YANAC cards are the same). (This variant does not use any Sympathizer card, no matter how many players).

For each player, secretly roll the die. On an 8, that player gets a card from the Cylon pile. Otherwise, that player gets a card from the YANAC pile. Roll twice for Baltar.

At the sleeper phase, do the same thing again (roll twice for Boomer).
If there were zero OR one Cylon pre-sleeper, then add +1 to this set of rolls. 
Don't give these Loyalty cards out until you have finished this set of rolls, because of this:
After this set of rolls, if there would still be zero Cylons, or if it's a 6+-player game and there would be only one Cylon, randomly choose one player who will receive a You Are A Cylon card instead.

Before the game begins, you can decide how Executing human characters works. Either Pegasus-style (no new loyalty card) or Exodus-style (roll a die as above for the replacement character).

You may even adjust probabilities up or down by changing the die type (with the top value of that die being the target for Cylonity, of course) -- d6 for a harder human game, d10 for an easier human game.

So you may end up with a 5-player-game with 3 Cylons, or even a 6-player game with 6 Cylons all wondering who they are screwing.. On average, though, you should have a similar experience to the base game with just a wee bit more uncertainty.

Is this worth an experimental game?