Their system depends upon voting with the consensus or it punishes you. Because it uses voting to reach consensus it is fundamentally flawed in my opinion.
I haven't yet read this whitepaper yet. I'll take a look at it sometime soon.
But I think I should offer my two cents...
In contrast, it seems off putting in my opinion that people would trade on bitshares, just in the belief that someone else would will buy it at a higher price. This sounds like the greater fool theory. I buy an asset because I believe that other people are buying it too, and that nothing underlying the asset is what is fundamental driving the price. Why should anyone trade bithillary2016. Prediction markets because its strictly a betting market. The winner of the bet is guaranteed to get all of the winning pool. Its clear to me how prediction markets reflect fundamental event/ idea probabilities.
How do you suppose Bitshares Prediction Markets will reach consensus? What would happen the date after Bithillary2016? Why would anyone buy it at the higher price that I demand? If I am the last person to buy bithillary2016 at 90cents and we say she is guarenteed to win, who would buy for $1 after the election is over? They would lose 10cents per share because nobody else wants to buy it. Bitshares Prediction Market sounds like it can predict to a degree... just dont be the last person holding the bag.
Given that one is a continuously traded asset ( Bitshares) and the another (Bitshares PM) is traded until the end of a time horizon: I argue that you cannot develop them the same way. We already have proof in the Hollywood Exchange (which evidently Invictus had previously used as an example for why competitive trading markets work). Hollywood Exchange doesn't correlate to the stock price to budget earnings-- just by the bid and ask volume alone. The only reason Hollywood exchange seems to reflect the box office results is because the employees have to manually have to go in and multiply the price by the exzact weekend box office results on TV. Bytemaster, do you plan on doing that? Will you use centralized techniques to go in and make sure BitHillary2016 reflects 1 dollar as it should? Because if you don't (as already seen by HollyEX), it seems the bitshares prediction market experiment will trade on just "random expectations." And therefore fail to develop a strong correlation to the event.