Yeah this is why this is horribly organized. As it stands BFL coul announce they won't be able to ship until November (guaranteeing the .fail stock wins) and people could still continue buying more .fail stocks right up until all were sold - meaning no real profit for anyone who actually invested at the start.
Shares were to be stop being sold the second it becomes October 1st. If BFL were to announce they failed before then, sales would have been halted as well.
Because of nefario, if BFL were to announce their failure, I cannot halt sales.
First we had zhou tong, then we had pirate, now we have nefario. Does this shit ever end?
Your OP doesn't allow you stop selling shares before october 1st - and your model doesn't allow people to react to new information becoming available.
Say BFL announced today "it's looking unlikely we'll be able to ship on time." That's not enough to settle the bet - but definitely adds a lot of weight to one side of it.
Would you then want to lock shares? If so then where do you draw the line? If not, then buying shares at any time before significant new information becoems available is a mug's game - with no benefit for early adopters (just an opportunity cost of tied up funds + the certainty they'll be gamed by others once more information becomes available).
Your model fails because it gives no benefit for early adoption - just massive penalties. You should have used a model like the one I described previously - where initial prices for each side of the bet are determined on current knowledge then the market allows adjustment of position as new information is revealed.