Well this is very telling. I have always said that tradesports seems to be ahead of the curve with its predictions because it really is a very efficient and active traded exchange. Currently, they are now making markets on which states Bush is going to win in the next election and also who is going to win the caucuses and the democratic primary. Currently the mkt for Bush to win re-election is about 65 offer. That's pretty steep. That's predicting a landslide win for him. In fact when you go state by state, he is killing whoever the democratic candidate is by an enormous margin. These are the only states in which Bush is currently not ahead in. They are California, Connecticut, Hawaii, Illinois, Maryland, Massachusetts, New Jersey, New York, Rhode Island and Vermont. That's it. Ten out of fifty states. And Illinois is pretty close. In the other states Bush has 95 bids and higher in almost 75% of the states in which he is ahead. This is just unreal. As for the Democratic primary, Dean has a huge lead with a 70 offer now to win. No other candidate is close. I think this is very telling. I have found tradesports to be incredibly accurate with its market and if it is any indication, Bush is going to win by a landslide with the electoral votes. Even Florida where the democrats are spending a lot of money and really counting on this time around, Bush has a 68 bid and 70 offer there. This is not looking good for George Soros and the democrats.
I am really glad someone is documenting all of this as it happens. I wonder if tradesports shares the historical data...?
This kind of stuff is very interesting. I have not followed tradesport but I gather the accuracy of the predictive value of the current odds decays over time, as events evolve. Do you have any anecdotal examples to share on other event outcomes that have been traded on Tradesport? I recall some weebsite trading on Hussein's death by July, and in February it was 70% certain.
I'm not sure what you are referring to as historical data but you can click on the chart of the contract and see the history of the action just like a stock chart.
Yes, I must say it is quite accurate. I watched the markets on the oscars very closely and they accurately predicted all the upsets and the winners. I've watched the markets on the weekend box office totals and right on the money. Same with the grammies and emmy's. The sports markets are obviously very tight and very efficient. Also notice the markets on court trials. These have been frighteningly accurate. For example they have the Koby Bryant case at a 94 offer for it going to trial and only a 30 offer that he is found guilty of rape. I think this is right on the money even though I think Kobe is guilty as sin. So yes, I would say they are very accurate which is why these political contracts are very interesting because they are basically saying this election is already over.
Don't you think though that the election contracts and the kobe and oscar contracts are very different? I mean there isn't really much that can happen, in terms of news, with the oscars and the kobe trial for it to have an impact on the contracts and change the outcome. For the election though, who knows what kind of stunts we will see in the next year by both parties to change the outcome?
Yes I agree there is going to be more volatility with the election contracts but let me point something out. These contracts trade a lot like options. And the trading price in a way represents the delta of an option contract. So contracts that trade around 50 or between 40 and 60 tend to be very volatile. But contracts that are trading at 75 or higher or 25 and lower are not very volatile at all just like in real options. It's amazing at how the pricing structures are so similar. The delta of the option in many ways accurately depicts the odds of the contract finishing in the money or in the case of the tradesports contracts, finishing with the right outcome. So my point being after all this dribble is that many of these contracts are priced so that there won't be any volatility which is the big surprise. Thereby representing the absolute certainty in the outcome in which it is predicting. I find this really interesting. So my point about all of this was that these contracts are not just depicting that Bush is going to win the election or certain states but rather the absolute certainty behind it. Keep in mind these contracts are very actively traded and there is very real money behind these contracts. If someone thought they were mispriced they could make a fortune off of them but what I have noticed so far with my experience with tradesports is that most of these contracts are very fairly priced and stunningly accurate. Even the arbitrage that exists between different months and similar events is very tight. This is really interesting to look at. Because for the savy trader, you can easily extrapolate various outcomes from different things happening that can create virtual risk free trades. But these relationships are very tight showing that someone or some group is actively arbing them. Maybe I'm the only one noticing this.
Let me give you an example of a perfect arb here. The florida/overall election spread. I think most people would agree that the outcome of Florida will possibly decide the fate of the election. Remember Florida was the tiebreaking state in the last election. And the state is neither democratic nor republican. Let's take a look at this. Since the last election Florida has gained two more electoral votes. They now have 27 vs 25. Texas also picked up 2 which will go to Bush. California picked up one which will go to the left and New York lost 2 which will go to the left. So between the other three big states (NY, CA, and TX) republicans gain a 3 pt edge. So this makes Florida even more important since it picked up 2 votes. So a very good spread would be to bet that Bush would win which is at a 68 offer and then sell florida short at 66 bid. This would lock in 2 pts which is very very tight. Basically what you are saying is that if Bush loses Florida he will lose the election and if Bush wins Florida he will win in the overall election. So your long at 68 and short at 66. You lock in 2 pts either way. This spread was wider but is now very tight and someone is trying to capture the edge on both sides of this spread by making a very wide market. Size is bid at 65 for Bush to win and size is offered at 73 for Florida. He is sitting there trying to arb 8 pts. I don't know I find this very interesting.
Very cool idea Mav. I'm gonna look around and see if there is a more liquid book. How about that Univ. of Iowa thing that started some of this stuff?