• Quick note - the problem with Youtube videos not embedding on the forum appears to have been fixed, thanks to ZiprHead. If you do still see problems let me know.

Intrade

BenBurch said:
Which is what I was asking for. Are there any studies?
Click to expand...

I looked a bit, and didn't see any that specify 18 months in advance.

I would be willing to bet (so to speak) that they're at least as accurate as opinion polls that far ahead--and probably more so.
 
I knew there was something else bothering me in this post:

casebro said:
The point of the vegas line is "everybody knows who is going to win, but we need to set the odds to get betting going." So the bet becomes "how much will Team X win by?"
Click to expand...
That's not true. If Team X and Y are playing, and the spread is "Y by 3 points", the bet you make is whether or not you think Y will win by 3 points. (Imagine a number line. Taking one side of the bet, you'd get a filled in point at +3 extending to infinity to the right (signifying Y wins be at least 3 points. The other side of the bet is an open point at +3 extending to negative infinity to the left, signifying either Y will win by less than 3 points, the game will end in a tie or X will win by any number of points.)

Nobody knows that X will not win. In fact, many betters actually believe X will win.

As I said, the purpose of the point spread is strictly to keep the odds at even money to simplify betting. However, that point spread is determined by the betting "market". If you set the spread at "Y by 10 points" and nobody takes the bet, you should make the spread greater. That's the only comparison to the prediction market I was making. The spread will change to reflect the betting market just as the share price in prediction markets will change to reflect the buying and selling in the prediction market.

ETA: And, as I said, if you want to build any reflection of the "spread" in an election (that is the margin of victory), you'd have to create another market. One might be "Obama re-elected in 2012 by a landslide". I guarantee you the share price of that market would be considerably less than $5.72 (the current share price of the "Obama to be re-elected in 2012" market).
 
Last edited:
Isn't one of these 'markets' run by a college? I would think they would maintain stats. Seems there has been a finite number of 'stocks' traded... stats would be easy?
 
Brainster said:
You can't read it as a comparison to the percentage of the actual vote. Let me illustrate. Suppose that one week prior to the vote, polls showed Obama leading by 20 percentage points. What would the InTrade or IEM markets show?

They would show a much greater number, because the odds of Obama winning when that far ahead in the polls would be overwhelming; they would not be 60% but maybe 95%, so that his shares might be priced at $9.95 or so. Remember, the settlement price on Obama's shares is either $10.00 or $0.00; they don't settle at the percentage of the popular vote.
Click to expand...

This ^

Well said, Brainster.
 
casebro said:
Isn't one of these 'markets' run by a college? I would think they would maintain stats. Seems there has been a finite number of 'stocks' traded... stats would be easy?
Click to expand...

I think that'd be the University of Iowa's.

ETA: http://tippie.uiowa.edu/iem/archive/

It looks like in assessing their accuracy, they take the price of the stock at midnight the night before the election in question. I didn't see any information about 18 months out.
 
Last edited:
BenBurch said:
Right now the market gives a 58% chance of an Obama re-election;

http://www.intrade.com/v4/markets/contract/?contractId=743474

I know these markets tend to be very accurate a few months out, but has it been studied how accurate they are 18 months out?
Click to expand...

The thread started on the 2010 election in late 2009 initially stated the Democrats had a 75% chance of retaining both the House and Senate in 2010. We saw how that worked out.

So, in answer to your question: not very accurate 18 months out. A lot can happen in 18 months.
 
MattusMaximus said:
The thread started on the 2010 election in late 2009 initially stated the Democrats had a 75% chance of retaining both the House and Senate in 2010. We saw how that worked out.

So, in answer to your question: not very accurate 18 months out. A lot can happen in 18 months.
Click to expand...

And I would say in comparing opinion polls and prediction markets that relying on a prediction market 18 months out would be like trying to draw conclusions with just the first couple of data points in an opinion poll.
 
Dumb All Over said:
I am reminded of this thread wherein the polls and prediction markets guessed the Democrats had a 75% chance to hold the House in the 2010 mid-terms.
Click to expand...

Good find. Forgot about that one.

Just goes to show that a year or even half a year is a long time in politics. IIRC by the summer of 2010 it was becoming quite evident that republicans would win the midterms.
 
A subjective question: How often/how much, do the Markets differ from opinion polls?

I see the markets as another opinion poll, but with a different subject selection process. Just as the polls subjects vary from party affiliation, random citizens, those likely to vote, etc. Rather than random citizens, markets get more serious subjects- those willing to put up $$$.

I wonder if any data is kept on 'shareholders'? Party affiliation, demographics? Possible selection bias towards ivory tower occupants? That could account for the wrong prediction of the dems holding congress.
 
casebro said:
A subjective question: How often/how much, do the Markets differ from opinion polls?
Click to expand...
The link I posted above to Iowa University has a report on that question. (But they take the final "prediction" as the value of the market at midnight before the election and compare it to final pre-election opinion polls.)

[ETA: And how is that a "subjective" question?]

I see the markets as another opinion poll, but with a different subject selection process.
Click to expand...
Not at all.

It doesn't ask who you would vote for as in an opinion poll. It essentially asks who you are willing to bet will win.* (Your own opinion on who you would like to win is irrelevant.) There is no subject selection process (or rather it is all self-selection), and participation is proportional to the amount of money invested. If you feel strongly that the prediction is a good investment at the current price, you can buy as many shares as are available.

A prediction market is not at all another kind of opinion poll.

*ETA: That's not really accurate either. It essentially asks if you think the prediction is a good investment. A given market is simply a prediction ("Obama will win the 2012 election" for example). Right now I think that particular market, for example, is likely to go up, so I think it's still a good buy at roughly $5 per share.
 
Last edited:
Can you bet actual money on Intrade? Is it legal in the US? If you win money, is it taxable?
 
casebro said:
That could account for the wrong prediction of the dems holding congress.
Click to expand...

I think what accounts for that is that at 18 months out, the market hadn't yet settled down. (If for example that market had just opened, investors could have bought a lot of shares at a bargain price just for the anticipated spike in stock value, planning to sell before the price dropped.)

ETA: That is, markets can have "bubbles" where stocks are overvalued (in relation to the intrinsic value of the prediction).
 
Last edited:
Number Six said:
Can you bet actual money on Intrade? Is it legal in the US? If you win money, is it taxable?
Click to expand...

You can't make or lose a lot of money. It's not a bet, though, it's an investment, so it is as legal as any other investment. Some other prediction markets use play money.
 
The Central Scrutinizer said:
Sounds like a buying opportunity.
Click to expand...

I agree and would do so but for 2 things. You have to jump through a lot of hoops in order to invest with Intrade (including sending scanned images of documents to them). Since the absolute maximum is $10 per share (and that won't happen), and the maximum number of shares you can buy is pretty low, you can't make much of a return. (You also can't lose very much.)

Expect a significant jump if and when a clear GOP frontrunner emerges or at least by the time the GOP names its candidate.
 
JoeTheJuggler said:
I agree and would do so but for 2 things. You have to jump through a lot of hoops in order to invest with Intrade (including sending scanned images of documents to them). Since the absolute maximum is $10 per share (and that won't happen), and the maximum number of shares you can buy is pretty low, you can't make much of a return. (You also can't lose very much.)

Expect a significant jump if and when a clear GOP frontrunner emerges or at least by the time the GOP names its candidate.
Click to expand...

I'll take your word for it. I have no idea what Intrade is, and can't be bothered to find out. :)
 
JoeTheJuggler said:
Sinking? Today's change was plus 14 cents (+2.9% change).
Click to expand...

Dead cat bounce.

In general, it's been pretty flat, hovering right around $5 which is pretty much what you'd expect so far in advance.
Click to expand...

Wrong. In general it's been above 60% ($6.00) for most of this year, which as discussed above is because most presidents get reelected for a second term. In Ben Burch's opening post on July 14, it was still as high as 58% ($5.80).
 
You must log in or register to reply here.

Back
Top Bottom