People love polls, no matter how often they prove worthless.  Consider the polls before the New Hampshire primary that put Obama up by 20-some odd points, only to suffer a double digit loss). Right now, the various polls have Obama and McCain scattered all over the place.

Gallup’s latest survey has Obama up 46-43 percent. Quinnipiac gives the Democrat a slightly larger edge at 50-41. Rasmussen found the race a close 47-45. And Newsweek posted recent results at 44-41. Most of those are statistical ties, given the margin of error. (The same Newsweek poll four years ago had Kerry leading 47-44). The change is stark considering that just a few weeks ago Newsweek had Obama up 15 points, 51-36. Three weeks before that they found the two in a dead heat 46-46.

Given these quick switches, is there any good way of projecting the race? If there is, it involves a principle: follow the money. Any pundit can get on CNN or Fox and say why McCain or Obama will win in November. They can camp out on weak reasoning to draw the public along. But to put money down on who you think will win takes a little more conviction.

Intrade, an online “prediction market,” allows you to buy and sell “belief” in a particular candidate. If you think McCain will win, you can put money down on it, though its more like buying stock than gambling because you “buy and sell,” not just place a one-time bet.

Prediction markets have a track record of incredible accuracy. The reason is simple: people don’t throw money away. If you have the inside scoop, markets like Intrade are where you can profit from your knowledge. Currently the lines have the Obama up 65.1-30.7. If there is any poll McCain should be worried about, it is this one.

Rasmussen also has a prediction market, powered by Intrade. With them, Romney is the favorite to get the Vice Presidential nomination, though at a weak 24 percent. You can also bet the index of a select field of candidates.

Prediction markets are telling in other ways too. On Intrade, only 30 percent believe we will go into a recession (and no we are not in one yet, just a slow economy).  Only 14 percent believe the US and/or Israel will strike Iran. If that number spikes up at any point you can bet it’s because people “in the know” are preparing for war—and Tehran should take it as a strong warning.