How I Made Some Easy Sheckle on Osama Bin Ladin

I've been asked to explain this SO many times that I've finally decided to write it down.

One by-product of society's current obsession with gambling and auction style commerce are "current event" markets. Just apply the protocols of sports betting to other uncertain topical outcomes like elections, judicial rulings, weather, or military incursions and you've got yourself a near endless supply of wagerable events!

Trading contracts are assigned to event that have binary outcomes due at a predefined expiration date. At expiration the contract is either worth $0 or $100, depending on the outcome. While the event is still uncertain the contract price will fluctuate according to the probability of that event occurring.

Let's take an example:

Bird Flu <> 31Mar06
This contract will be valued at $100 if by 31mar06 a confirmed case of bird flu is reported in the United States. If no cases are reported the contract will expire at $0.

Currently the market for this contract is 12.5 @ 15

This market is implying that there is a 13.75 percent chance that this event will occur (just average the 2 numbers above). The market above also tells you that someone is attempting to "buy" the contract @ 12.5 while someone else is trying to "sell" the contract @ 15. Just as in the stock market or any similar dutch style auction there are always buyers and sellers, and when a buyer and seller agree to a price there is a trade. Remember that in this type of market participants wager WITH EACH OTHER. There are no bookies, oddsmakers, or intermediaries involved here. The exchange makes money by charging a commission each time you roundtrip a contract.

Here's how it breaks out:
You can buy 1 contract @ 15 (there is someone attempting to sell a contract @ 15)If no bird flu occurs you lose $15 (the contract expires @ $0)
If bird flu occurs you win $85 (the contract expires @ $100 so you've won $100 minus the initial price)

You can sell 1 contract @ 12.5. Note that you don't have to previously own a contract in order to "sell" or "go short" it.
If no bird flu occurs you win $12.5
If bird flu occurs you lose $87.5

Make sense?
Now to Osama Bin Ladin...
Osama Bin Ladin <> (assume that today is 20Mar03)
A. 31Jun03: 37 @ 38
B. 31Dec03: 38 @ 40
C. 31Jun04: 39 @ 41
So what's wrong with this market? Remember that if at anytime Osama Bin Ladin is captured all of these contracts will be worth $100. But if he is not captured by each expiration that contract will expire at $0. The trade opportunity here was pretty obvious - sell contract A and buy contract C in equal amounts. Let's say today is 20Mar03 and you are able to sell $1000 worth of contract A and buy $1000 worth of contract C.
You sell $1000 of A @ 37 (risking $630 to win $370 if Osama is not captured)
You buy $1000 of C @ 41 (risking $410 to win $590 if Osama is captured)

As 31Jun03 approaches contract A will slowly leak towards zero if Osama is not captured while contract C will retain most of its value since expiry is still so far in the future.

So here is a snapshot of the market on 31Jun03...

Osama Bin Ladin <> (assume that today is 31Jun03)
A. 31Jun03: 0 @ 0
B. 31Dec03: 25 @ 27
C. 31Jun04: 36 @ 38
So we've made $370 by betting that Osama would not be captured by 31Jun03 (by selling contract A @ 37). And on contract C that we bought @ 41... Well, it's now only worth 36 but we can sell it out @ 36 at a loss of only $50. You can repeat this trade by selling contract B and buying contract C, but it's no longer as attractive because the term structure has correctly "steepened" back out. Originally the term structure was too "flat" assigning too little probability for the same event but with more time until expiry.