I mean: I don't trust any actor on any market.
If more than half of the actors decide to cooperate,
they trivially win in this case.
If the NSA decides to fuck a prediction, they can buy
ether and their bots will do the rest.
Does it matter whether the crowds can do accurate predictions if
the main purpose is for special actors with the skills and interest to
be encouraged to create outcomes that are being wagered upon. I've not
looked closely at Augur but back in Cpunk days stumped for PMs that
included forwards, similar to financial futures, but where the
participants could create contracts which narrowed the prediction
windows to months, weeks or even days within the greater expiry period (e.g., Dec. 31, 2015) but were not required to specify
which month, week or day . By enabling
such sub-contract narrowing the benefit for an accurate prediction
scales with its temporal ratio with the general contract thus increasing
leverage and benefit to special actors.