Comments on: Mr Money, meet Ms. Mouth

By: Kimmitt

Kimmitt — Fri, 28 May 2004 07:17:19 +0000

Isn’t the point of the Iowa markets that they accurately reflect public opinion but far more cheaply?

By: dsquared

dsquared — Tue, 25 May 2004 21:53:48 +0000

Matthew: the positions are shown by the bars; a bar pointing up from the RHS zero axis means that I’m long KERR, a bar pointing down means I’m short. Bill: well yes, it’s all a bit of fun really. I put the caveats in a comment rather than the main story specifically to avoid making it look like I was trying to excuse failure of the system ahead of time. All your points are well taken, but I think that any similar market would also have them. Unless we are seriously considering, as a society, diverting material proportions of our savings out of investment in productive assets and into rinkydink “predictive” exercises like this one, this objection is always going to be there.

By: Bill Carone

Bill Carone — Mon, 24 May 2004 17:49:40 +0000

Dsquared, Do you have an argument that defeats the people who, disagreeing with the results of your study, simply will say “How unlucky/lucky of Dsquared.”? For example: “NB: I really am setting myself up for a fall on this one, by the way.” So you’ve preemptively explained away any negative result. Also, I don’t see how this addresses the specific issues with the IEM: small amounts of money bet, relatively large start-up costs for bettors, small number of bettors, etc. So your conclusions may be specific to the IEM as it exists now, and may not help you argue about other markets. For example, are the following reasonable? – Small potential profits mean that people will not spend much time and effort making their bets; prices could well be determined by people betting “for fun”. – This causes the market to be quite slow in incorporating new polling information. – This allows you to make small amounts of money, not worth your time and effort. – If just a few other people were doing what you are doing (or even similar things), then you couldn’t make money.

By: Jason

Jason — Mon, 24 May 2004 17:28:35 +0000

Kudos to you, win or lose. This is the best way to convince the market-mad people (myself included).

By: Matthew

Matthew — Mon, 24 May 2004 16:37:27 +0000

Dsquared, Without giving the details away can you explain what the chart shows? I thought it meant you took positions when the two lines cross, but the most recent one clearly isn’t.

By: dsquared

dsquared — Mon, 24 May 2004 16:07:23 +0000

heh. I’ll cut you in on a share of the profits if there are any … so far there haven’t been any moves at all.

By: Nasi Lemak

Nasi Lemak — Mon, 24 May 2004 15:51:04 +0000

I assumed you wanted the spreadsheet to make some recondite (yet wittily put) point about the fallibility of large populations, not to make your fortune!

By: dsquared

dsquared — Mon, 24 May 2004 12:48:14 +0000

NB: I really am setting myself up for a fall on this one, by the way. If you look at the second bar on the chart above, it’s driving nearly half of the entire backtesting performance, and it’s just by pure fluke that the system happened to switch over on the precise peak of one of the bigger volatility spikes during the period in which the KERR contract was settling down post-Iowa. I obviously can’t count on this happening in the future, so I’m just going to hope I can rely on 1) the KERR contract having persistent trends, 2) these trends running roughly in line with the poll data and 3) me being quick enough on the trading button to avoid getting my face ripped off by intraday movements and the bid-ask spread. My guess is that 3) is the biggest risk.