Obama @ 86 & 83

[jheuristic]

By David Alexander
1 hour, 40 minutes ago


WASHINGTON (Reuters) - Democratic presidential contender Hillary
Clinton will win several state nominating contests in the coming
months but has little chance of becoming the party's candidate for the
November 2008 election, traders were betting on Tuesday.

Traders in the Dublin-based Intrade prediction market gave Democratic
front-runner Barack Obama an 86 percent chance of being the Democratic
presidential nominee, versus a 12.8 percent for Clinton, the New York
senator and former first lady.

Results were similar on the Iowa Electronic Markets at the University
of Iowa, with traders giving Obama an 82.9 percent chance of winning,
versus a 12.8 percent chance for Clinton.

Intrade traders were betting the Democratic nominee would ultimately
become president. They gave the Democrat a 59.1 percent chance of
winning, versus a 48.8 percent chance for the Republican. Iowa traders
gave the Democrat a 57.1 percent chance of winning, versus 46.3
percent for the Republican.

Prediction exchanges let traders buy and sell contracts on the
likelihood of future events. Contracts are structured so the prices
can be read as a percent likelihood of an event occurring. Studies of
prediction markets have shown they have an accuracy comparable to that
of public opinion polls.

Expectations that Illinois Sen. Obama would be the Democratic
presidential nominee have strengthened from 75 percent a month ago.

During that time Obama, who would be the first black U.S. president,
has weathered a political storm over controversial statements made by
the pastor of his church and has delivered a well-received speech on
race in America.

The strong view of Obama's ultimate success came despite expectations
that Clinton would win several important state nominating contests in
the coming months.

Intrade traders were betting Clinton would win the contest in
Pennsylvania on April 22, giving her a 66.1 percent chance, versus
32.8 percent for Obama. They gave her a 79 percent chance of winning
the West Virginia contest on May 13, versus 20.5 percent for Obama,
and a 70 percent chance of winning in Kentucky on May 20, versus 30.5
percent for Obama.

Traders were betting Obama would win the Indiana contest on May 6.
They gave him a 58 percent chance, versus 45 percent for Clinton.
Traders gave him an 88 percent chance of winning the May 20 Oregon
contest, versus 12 percent for Clinton, and an 82.5 percent chance of
winning the June 3 Montana contest, versus 17.5 percent chance for
Clinton.

(Editing by David Wiessler)

[Robin Hanson]

At 05:10 PM 4/8/2008, Reuters wrote:
>Intrade traders were betting the Democratic nominee would ultimately
>become president. They gave the Democrat a 59.1 percent chance of
>winning, versus a 48.8 percent chance for the Republican.

59.1+ 48.8 >> 100. I hope that's a typo ...

Robin Hanson rha...@gmu.edu <http://hanson.gmu.edu>http://hanson.gmu.edu
Research Associate, Future of Humanity Institute at Oxford University
Associate Professor of Economics, George Mason University
MSN 1D3, Carow Hall, Fairfax VA 22030-4444
703-993-2326 FAX: 703-993-2323

[J Maloney]

Hi -- Yeah, there are others like that, so not sure it's a typo... -j

[David Pennock]

I think it must be a typo. The republican contract has not been much
above 40 for weeks:

http://www.intrade.com/jsp/intrade/common/c_cd.jsp?conDetailID=173054&z=1207768430934

This brings up an interesting question: what is the "right" way to
report prediction market numbers? Should journalists report last trade
prices or bid-ask spreads? When phrased as a probability, as in this
article, should journalists by default normalize the numbers so they sum
to 100?

Another nicety of using a Hanson-style automated market maker: in that
case, all of these approaches amount to the same thing.

regards,
Dave

[Byrne Hobart]

This brings up an interesting question: what is the "right" way to
report prediction market numbers? Should journalists report last trade
prices or bid-ask spreads? When phrased as a probability, as in this
article, should journalists by default normalize the numbers so they sum
to 100?

I think the best way would be the utilitarian way: "For $4.00, an investor could purchase a contract which would yield $10.00 on X if Y." With perhaps some sort of link to an FAQ, or an indication that this isn't a direct reference to odds. They could report a range (e.g. "Yesterday, bettors were willing to pay between $3.70 and $4.10 for...").  

[Lucy Vega]

Split the spread, adjust for rapacious fees and other artifacts of silly market design (e.g. the several million $s in open offers on Intrade for non-candidates), and say, "Traders currently give A a B% chance to win the nomination and X a Y% chance."

By the way, what I find more interesting than the >100 pricing is the fact that there's a fair bit of cash backing Gore as Democratic Nominee (or just President) at roughly 30-1. Any of you drinking that Kool-Aid (or selling against it)?

And speaking of open offers on non-candidates, I'm not at all sure what to make of this... One pays ten cents at the expiration of a winning contract on Intrade, so offering to sell for less than that amount guarantees a loss unless one is selling to close out a long position. But, for instance in the case of 2008DEM.NOM.BLAGOJEVICH, there are 19,987 contracts offered for a penny, but only 5,486 contracts have ever traded. So at an absolute minimum the sellers are begging for a guaranteed loss of $1,305.09 for just this one "candidate." It can't be because of a non-fee-paying market maker, because Intrade says there are no non-fee-paying market makers on the Nominee markets. So, what gives?

- lucy

[vega...@gmail.com]

"So, what gives?"

Nobody?

Paging Dr. Delaney...

On Apr 9, 1:05 pm, "Lucy Vega" <vega.l...@gmail.com> wrote:
> Split the spread, adjust for rapacious fees and other artifacts of silly
> market design (e.g. the several million $s in open offers on Intrade for
> non-candidates), and say, "Traders currently give A a B% chance to win the
> nomination and X a Y% chance."
>
> By the way, what I find more interesting than the >100 pricing is the fact
> that there's a fair bit of cash backing Gore as Democratic Nominee (or just
> President) at roughly 30-1. Any of you drinking that Kool-Aid (or selling
> against it)?
>
> And speaking of open offers on non-candidates, I'm not at all sure what to
> make of this... One pays ten cents at the expiration of a winning contract
> on Intrade, so offering to sell for less than that amount guarantees a loss
> unless one is selling to close out a long position. But, for instance in the
> case of 2008DEM.NOM.BLAGOJEVICH, there are 19,987 contracts offered for a
> penny, but only 5,486 contracts have ever traded. So at an absolute minimum
> the sellers are begging for a guaranteed loss of $1,305.09 for just this one
> "candidate." It can't be because of a non-fee-paying market maker, because
> Intrade says there are no non-fee-paying market makers on the Nominee
> markets. So, what gives?
>
> - lucy
>

> On Wed, Apr 9, 2008 at 12:25 PM, Byrne Hobart <bhob...@gmail.com> wrote:
>
> > > <http://www.intrade.com/jsp/intrade/common/c_cd.jsp?conDetailID=173054...>This