Reply With QuoteIt's still amazing to me how many supporters Dubya and gang have. It must be mass insanity.
In a head-to-head, Kerry beats Bush 55 to 44.
With Nader in the mix, Kerry beats Bush 52 to 42 and Nader has 4. How can Nader cannibalize points from Bush?
Oh, and 60% of Americans polled weren't happy with the direction America was going.
It's still amazing to me how many supporters Dubya and gang have. It must be mass insanity.
last night at dinner i heard a brief diatribe about how clinton was the Worst President Evar. no one mentioned Bush at all.Originally Posted by Midnight Son
i'm pretty sure the "at least he's not a democrat" logic is going to hold true for an awful lot of dyed-in-the-wool conservatives.
Sadly, there are plenty of Democrats with the same line of thinking about any Republican. The polarity our two party system has caused is ridiculous. I'm not sure it had to turn out this way, but the way it has turned out - with people identifying themselves in xenophobic political tribes, not giving a shit about anything but "at least he's one of us," is ruining this country, along with the abandonment of media responsibility to educate and inform the citizenry, and a lack of even basic concern for the direction the country is going among said citizenry. The ship of state is sinking.
Amen.
Bravo. Very well said.
I'm not sure if you can blame the party or 24-7 media, where conflict=ratings.
I'll have more to say on this when I finish Brock's new book. It's eye-opening.
i dunno. bush's stock price versus kerry's is still freaking rock solid. i don't get how he never even breaks even with kerry even with the worst news.
http://128.255.244.60/graphs/graph_Pres04_VS.cfm
http://128.255.244.60/quotes/66.html
Probably because the IEM has the predictive power of a hubcap on the side of the road.
Looooook into the rims!
Different methodolgies. Polls ask "If the election were held today, how will you vote?" The IEM asks "When the election is held, how will everyone vote?" The way to read a stable market given unstable polls is that the investors feel the instability will eventually stabilize.
As far as hubcaps go, it's a pretty reliable one.
The 1996 U.S. Presidential was skewed in the final days because of a large influx of new traders. Despite the skew of the election eve close, the market over the week before the election was still more accurate than the polls.The market outperformed polls in 9 of 15 cases according to both measures (election eve closing prices and last week average prices). Across all elections, the average poll error was 1.93% while the average market error was 1.49% and 1.58% by the two measures. In a few cases (the 1988 and 1992 U.S. Presidential elections) the market dramatically outperformed polls. The worst outcome, the 1996 U.S. Presidential election, is a peculiar one that gets additional attention below. In the majority of other cases, the market does about as well as the average poll, sometimes worse but often better, even if by a small margin.
Election eve outcomes are to some extent less interesting than predictions over the full course of the campaign. We notice a general tendency for the market to be both closer to eventual election outcomes and more stable than polls over the course of election campaigns.
First, national polls are meaningless. The election will be decided in 11 states. Only polls that survey those 11 states matter.
Second, polarity between the parties is good, if a ridiculous statement. There have been complaints for years about how the two parties are virtually identical. This has led to voter apathy (at least, that is the argument.) If the country truly is as polarized as the currnet conventional wisdom holds, the voting participation should increase significantly this year.
The parties are VERY similar. Certain political factions on both sides get worked up about abortion, stem cell research, Iraq, etc. But in the end, there are members of both parties that cross those lines. Bush cut taxes like a good Republican, but he passed a massive Medicare drug expansion that makes Hillary smile. A Kerry administration would have cut taxes less (if at all) but he would have been stymied completely on any Medicare expansion. Neither candidate is calling for radical reform of anything. They are both just tweaking.
Daniel over at crookedtimber.org has this jihad going against the IEM that's far too complicated to go into. Suffice to say that he thinks it's a lagging indicator.
Of course they're a lagging indicator, investors have to have some time to determine whether an event has legs or not. As for Daniel's "jihad", it doesn't seem to be as damning as you claim.
This doesn't sound like "the IEM has the predictive power of a hubcap on the side of the road" to me.In particular, John and myself have defended the view that these markets do not appear to offer marginal information above and beyond published opinion polls.
Even when he does let loose, it's largely off-base. His rant on the non-prediction of Dean's Iowa collapse is way off:
First off, as he notes, everyone was caught off-guard by Dean's free fall, not only in how poorly he'd do in Iowa, but also how badly he would recover. Even after he became the butt of jokes, Dean was still considered a front-runner, although no longer the only one. Secondly, the market is for the nomination, not the Iowa Caucus, and it made no claims as to how Dean would do in any one primary or caucus, even one in the home state of the market. While the caucus was an important determinant of get the nomination, it isn't the only one, and and even with a poor showing in Iowa, Dean still could have captured the nomination, so Dean's stability before the caucus is not totally without merit. Once Dean became the butt of everybody's jokes, however, it was clear that he wasn't going anywhere, thus the freefall after the caucus.Can we no longer hear about the “predictive power” of the Iowa Electronic Markets, please? They were bamboozled to exactly the same degree as the rest of us.
(UPDATE): A couple of people in comments have pointed out that this market is for the nomination, not the Iowa Caucus itself. Fair point, but sadly, no. Either the Iowa Caucus is an important determinant of who gets the nomination, or it isn’t.
Then his IEM quiz is just bizarre (SPOILER ahoy! Skip if you intend to take the quiz yourself):
Well, yeah, duh. The IEM isn't intended to be some super investment scheme. Counting the time value of money against returns is like counting your internet connection bill against returns. It's the cost to play. The bottom line is if you spent $2 for 1 Clinton contract and 1 Dole contract and held them to maturity, you would have gotten your $2 back.1. In the 1996 Presidential vote-share market, what should the sum of the values of the VS_CLINT (Clinton vote share) plus VS_DOLE (Dole vote share) contracts be?
...{snip wrong answers}...
The correct answer is 1.00 minus the time value of money. Because you have to pay your money upfront to receive the return after the election, a portfolio of 1 Dole plus 1 Clinton should sell at a discount to face value reflecting that fact.
Not surprising as the market is winner-take-all, based on popular vote, as noted in the prospectus. What's his point?2. What percentage return would you have made in the 2000 winner-takes-all market by buying the BUSH contract at the point when GORE was at its peak and holding to maturity?
You would have lost all your money.
...as noted in the prospectus. What's his point?3. You hold a porfolio in the current 2004 Presidential vote-share market long BU|KERR but short BU|CLINT. If George Bush were to announce tomorrow that he had decided to withdraw from the race, what would be your profit or loss?
You would neither win nor lose anything. BU is simply an identifier meaning “the Republican candidate”.
Hey, it's his argument, not mine. But it sure looks like "the IEM is no more predictive than the average of all polls" is accurate.
So are you saying that the average of all polls "has the predictive power of a hubcap on the side of the road?" Or are you changing your opinion?
[size=2]Edit: Just so I'm clear, I don't think changing your opinion is necessarily a bad thing.[/size]
If IEM is no more predictive than averaging public available polls, what information is it providing?
At worst, it's another data point by which to evaluate a complex data set. At best, realizing that the IEM has historically been better than the average of all polls 60% of the time, you can use that data to help evaluate the situation when the polling data shows instability but the IEM remains relatively stable.
Except that it hasn't. One obvious hypothetical is that people are taking the polls and just buying based on that. Seriously, what mechanism would allow the market to predict the election?
It hasn't what?
- Historically better than the average of all polls? It has (9 out of 15 times).
- Been relatively stable? Look at the 1996 Market trend, and note the wild swings in the Gallop results over the last few weeks of the campaign, the IEM was very stable in comparison (and this was the infamous "new investor skewed" market.)
As for what mechanism allow the IEM to be more predictive, the knowledge and profit motive of the investors.
i can't explain it in words, but in examples i heard in other explantions.
take 100 random people. take a jar of jelly beans. have all the people write down their guesses for the # of jellybeans. the average will be fairly accurate.
take 'who wants to be a millionaire'. they are right like 80 something percent of the time. (in america. in russia i heard the audience puts wrong answers on purpose) of course, i also heard that the audience mostly only knows popular culture questions.
the government actually tried to do something like this to predict terrorist attacks not too long ago. newspapers and the general population called it "ghoulish" and i think some guys lost their jobs.
The IEM is not intended to be random people making guesses. It's intended to be comprised of informed traders making profit-motivated decisions.
If they don't have unique information available to them to make those decisions on, the market has no predictive power. Not objecting in general; I'm going with Dan's implicit knowledge argument.
But argh, we're just rehashing the same stuff on there.
Many of the investors do have unique information available to them. I've known campaign staffers getting in on the market. Many professors and researchers who particpate, while their information may not be unique within their field, have information and knowledge the general public doesn't. Even those investors who rely on publicly available sources may be able to make a better prediction than a random poll, simply because they are more involved and self-informed than the average Joe.
Since you feel we're just rehashing stuff, and, I guess, ending the discussion, I'll take the opportunity to note that although you opened your discussion of the IEM with, "the IEM has the predictive power of a hubcap on the side of the road," you've since said that "'the IEM is no more predictive than the average of all polls' is accurate." Which is a great deal more predictive than a hubcap on the side of the road, so I have to assume that you've changed your opinion of the predictive power of the IEM (from "hubcap", to "average of the polls"), even though you won't admit it.
I'm also not sure why it matters what information the IEM provides. If nothing else, it's just another data point to rely on or ignore as you choose. It was set up as an experiment in teaching market mechanics and, as a secondary goal, predicting elections. It has performed admirably in both regards. I'm not sure how some random blogger's (I'm sorry, but "Daniel at crookedtimber.org" is not an authorative source. Honestly, has he even graduated? From high school?) jihad refutes the historical results the IEM has posted.
He's Paul Davies's kid, actually. Investment banker of some sort who used to have a mid-level position at the UK central bank.
But anyway, his argument should stand on the merits; it's not as simple as I made it, getting into market theory I haven't read:
http://www.crookedtimber.org/archives/001901.html
Some predictive markets work: the IBM ones where engineers and marketers both try to predict future sales of a product, for example. Some don't.
Edit: Here's the most interesting one:
http://www.crookedtimber.org/archives/000340.html
I still have yet to see a single reason why I should give his arguments any presumed weight.
Absolutely, let's see how his arguments stand on their merits. I read that link earlier earlier, but it doesn't have any arguments (other than when he says, "...these markets do not appear to offer marginal information above and beyond published opinion polls."). It sets up a hypothesis (that the IEM is expolitable) and gives a very high-level overview of the experiment that he'll run. His results aren't in yet, so we're all still just waiting on pins and needles to find out if his hypothesis is right. He doesn't even say that the IEM doesn't work, just that he thinks it's expolitible for profit.
This one is interesting, because it criticizes the Policy Analysis Market that DARPA propsed last year, not the IEM. Digging into his argument, he says that the market can't create knowledge about an event, it only aggregates knowledge. Which is no indictment of the IEM, since they are just looking to aggregate knowledge about an election, not determine whether or not the election will happen. If anything, this piece lends credibility to the IEM, becuase the IEM has more in common with the examples he cites where the markets work than it does with the PAM.
As for his other arguments elsewhere on the crookedtimber.org site, I've addressed them earlier in this thread. I've read every article on that site that has Iowa or IEM in it, and overall, I think he'd disagree with your assertion that "...the IEM has the predictive power of a hubcap on the side of the road."
In the end, his argument is "...these markets do not appear to offer marginal information above and beyond published opinion polls." Which is a far, far cry from "...the IEM has the predictive power of a hubcap on the side of the road."
What is the predictive power of an IEM if:
1. It can't beat opinion polls.
2. People are using those opinion polls as inputs to their IEM bets.
What information is being added by the IEM?
You can turn your first arguement completely around, what's the predictive power of opinion polls if they can't beat the IEM?
As for what information is added by the IEM, the "tacit knowledge" if the investors, who are decidedly not just using opinion polling data.
Um, polls existed before the IEM. Polls are an input to the IEM, not the other way around.
Posting Permissions