This is a recent picture of traitor Mike Linksvayer (much better than the New York Times picture where he looked like an angry chimp):
—-
First, Mike Linksvayer sided with moi:
[Hollywood Stock Exchange] could be an indicator that the [TradeSports] price is wrong and in which direction, which is enough for a [TradeSports] trader to place an order.
Then, as soon as I turned my back, he sold me down the river (I should have known!!!!!!):
I agree with everything you [you = Sacha Peter] write in the last comment (and your original challenge to CFM to put his money where his mouth is).
—-
#1. HSX “-prices”- are predictive (see The effectiveness of pre-release advertising for motion pictures – PDF – by Anita Elberse and Bharat Anand – 2005-03-05):
Despite the fact that the simulation does not offer any real monetary incentives, collectively, HSX traders generally produce relatively good forecasts of actual box office returns (e.g. Elberse and Eliashberg 2003, Spann and Skiera 2003). According to Pennock, Lawrence, Giles and Nielsen (2001a- 2001b), who analyzed HSX’-s efficiency and forecast accuracy, arbitrage closure on HSX is quantitatively weaker, but qualitatively similar, relative to a real-money market. Moreover, in direct comparisons with expert judges, HSX forecasts perform competitively.
#2. You can investigate accuracy and precision scientifically- no need to back one’-s research paper with one’-s money.
#3. My original question was: “Should I use HSX info to speculate at TradeSports?”. Looking at HSX and TradeSports odds, maybe The Departed is correctly priced or underpriced at TradeSports, and maybe Babel is underpriced at TradeSports.
I agree with Chris on this one. After cutting through the irrelevancies, Sacha’s argument seems to be that HSX prices will be inferior because real-money incentives are lacking there.
First, this can be tested, and has been to some extent. Second, if play money markets perform best when information is cheap and abundant, HSX can be expected to perform well, ironically, because it acts as a poll. The cheap “information” possessed by traders is their aesthetic taste. How much HSX trading is driven by personal tastes versus Keynesian “beauty contest” logic? We typically disparage the former in the poll vs. market discussion, but this sort of trading is helping HSX at this point.
Sacha’s argument seems to be that HSX prices will be inferior because real-money incentives are lacking there.
No, my argument is that HSX prices cannot be directly compared to Tradesports prices. HSX might be “more right” than Tradesports or it might not be. My claim is that you can’t know due to the lack of arbitrage between markets.
Let’s take an extreme example. If HSX says that “Borat” is a 90% shot to win the award, and Tradesports is at 10%, what do you believe? How about 60% and 50%? You can use any two numbers and the result is the same – it tells you nothing.
CFM’s argument #2 is garbage. This is how finance academics are sold on “efficient market theory”. This is also why you can create trading systems and backtest it with 20 years of prior data to “prove” it a winning one, but when you implement it in the real world it doesn’t work. Prediction markets are no different.
So the argument is that empiricism is “garbage” and that it is useless to examine, for example, the mean squared error of each market over time in order to gauge their accuracies?? It seems that you must then argue that the Borat or any contract is sufficiently dissimilar to all previous HSX/TEN contracts, so that any past performance can be dismissed. Maybe this is what you meant by your reference to data-mining, but i don’t think data-mining is very relevant here, nor EMH.
(Just to be clear, comparing scaled claims to binaries is another issue.. we are now just talking about binaries for the sake of argument.)
(Btw, checking on my HSX account, i bought Borat at 32.85 in june, and it settled at 109. The Mel Gibson short didn’t work out so well!)