Obama assassination : 1 in 55?
September 1, 2008 2:00 PM
Does the fact that the betting markets still give Hillary Clinton a 1 in 55 chance of being President in 2008, mean that that is a reasonable estimate of the chance that Obama is assassinated before the election? What else could justify those odds now?
Third party candidacy. I can see it now: Clinton/Lieberman. No, wait…Clinton/Clinton.
posted by dinger at 2:07 PM on September 1, 2008
posted by dinger at 2:07 PM on September 1, 2008
As the Clinton campaign awkwardly pointed out during the primary, Bobby Kennedy was the only person assassinated whilst campaigning.
44 Presidents say 100 candidates to make the math easy. 1 out 100 is the historical trend.
So it looks like the oddsmakers are (thankfully) betting on something else.
posted by three blind mice at 2:36 PM on September 1, 2008
44 Presidents say 100 candidates to make the math easy. 1 out 100 is the historical trend.
So it looks like the oddsmakers are (thankfully) betting on something else.
posted by three blind mice at 2:36 PM on September 1, 2008
Remember that the cost to trade is not zero. There is both fees and time/opportunity cost of capital.
If the price is 1.8, to bet that she will not be president you need to commit $9.82 for the $10 contract. But you pay $0.05 if you are a price-taker and are charged another $0.10 if you are correct. You could win between 3 and 8 cents on a $9.82 commitment.
The math
Price taker:
$9.86 = $9.90 * e^(-rT) * P(x)
Price maker:
$9.82 = $9.90 * e^(-rT) * P(x)
This contract would expire in about 4 or 5 months (inauguration day). I'll be charitable, and call it 3 months. 90d LIBOR annualized is around 3%. The discount factor e^(-rT) is e^(-.03*.25) = 0.9925.
0.9925*$9.90 = $9.826
The price taker is irrational, since they could put their money to work in a savings account and do better. The price maker's risk-neutral probability is more like:
$9.82/$9.826 = 99.94% probability that she will not be president.
posted by milkrate at 2:41 PM on September 1, 2008
If the price is 1.8, to bet that she will not be president you need to commit $9.82 for the $10 contract. But you pay $0.05 if you are a price-taker and are charged another $0.10 if you are correct. You could win between 3 and 8 cents on a $9.82 commitment.
The math
Price taker:
$9.86 = $9.90 * e^(-rT) * P(x)
Price maker:
$9.82 = $9.90 * e^(-rT) * P(x)
This contract would expire in about 4 or 5 months (inauguration day). I'll be charitable, and call it 3 months. 90d LIBOR annualized is around 3%. The discount factor e^(-rT) is e^(-.03*.25) = 0.9925.
0.9925*$9.90 = $9.826
The price taker is irrational, since they could put their money to work in a savings account and do better. The price maker's risk-neutral probability is more like:
$9.82/$9.826 = 99.94% probability that she will not be president.
posted by milkrate at 2:41 PM on September 1, 2008
What else could justify those odds now?The only thing "justifying" those odds is some subset of people still willing to bet on Clinton being elected. Oddsmakers don't make up odds based on their perception of the likelihood of an event -- it's pretty much all supply and demand based on the number of people betting each side of a proposition. The odds are set in order to (roughly) equalize the dollars bet either way.
posted by Doofus Magoo at 2:44 PM on September 1, 2008
I think that some people were playing a longshot on an upset in the vote at the convention last week. Basically the possibility that superdelegates would switch from Obama to Clinton.
posted by Class Goat at 3:15 PM on September 1, 2008
posted by Class Goat at 3:15 PM on September 1, 2008
44 Presidents say 100 candidates to make the math easy. 1 out 100 is the historical trend.
I'm not sure it works that way, esp something as complicated and unpredictable as that.
Heck, the historical trend of vice presidents committing an embarrassing spelling error during a visit to an elementary school is around 1 in 50, but I don't think one could say that Cheney's chance of doing the same during his tenure were about 2%.
(Although if Palin gets elected, her odds of accidentally shooting a guy in the face would be a little better than Biden's...)
posted by TheSecretDecoderRing at 3:22 PM on September 1, 2008
I'm not sure it works that way, esp something as complicated and unpredictable as that.
Heck, the historical trend of vice presidents committing an embarrassing spelling error during a visit to an elementary school is around 1 in 50, but I don't think one could say that Cheney's chance of doing the same during his tenure were about 2%.
(Although if Palin gets elected, her odds of accidentally shooting a guy in the face would be a little better than Biden's...)
posted by TheSecretDecoderRing at 3:22 PM on September 1, 2008
milkrate got it right. Intrade prices on the periphery are not directly translatable to percentages. You'll see a similar situation on contracts very near 100.
posted by 0xFCAF at 4:07 PM on September 1, 2008
posted by 0xFCAF at 4:07 PM on September 1, 2008
I noticed I have the price taker price wrong (it should be $9.87 in the example, since the fee is 5 cents), but it doesn't change the math or the conclusion that it is not profitable to continue betting against it. Fees are listed here.
posted by milkrate at 4:46 PM on September 1, 2008
posted by milkrate at 4:46 PM on September 1, 2008
Those people who are "betting" on Hillary aren't really betting on her to win the presidency. Rather, they are betting on her odds to go up (by some Obama bad news) - and then selling.
posted by whatisish at 4:52 PM on September 1, 2008
posted by whatisish at 4:52 PM on September 1, 2008
A tangential question - how is what intrade does not classified as straight up online gambling?
posted by TravellingDen at 9:35 PM on September 1, 2008
posted by TravellingDen at 9:35 PM on September 1, 2008
whatisish: That's just a dressed-up version of what happens at a regular bookie - it's not a stretch to make placing a series of bets based on what is currently being offered as buying and selling options on a future.
posted by TravellingDen at 9:38 PM on September 1, 2008
posted by TravellingDen at 9:38 PM on September 1, 2008
Slightly off-topic question, possibly connected to TravellingDen's: why does my work block the linked site?
posted by fiercecupcake at 9:16 AM on September 2, 2008
posted by fiercecupcake at 9:16 AM on September 2, 2008
« Older How do I get my Samsung Juke phone to work with my... | Where can I buy a copy of the G.C.S.E. Science... Newer »
This thread is closed to new comments.
posted by TheSecretDecoderRing at 2:06 PM on September 1, 2008