Some Support for the Tie Prediction

A couple of weeks back, I predicted a tie in the US Presidential elections. I even posted the likely electoral map. Now, it turns out that the website FiveThirtyEight.com, which runs simulations based on polling results finds that -

As you may have noticed from our scenario chart, the probability of a tie has increased dramatically in recent days and now stands at 3.2 percent. This is partly because, as we draw closer to election day with the race remaining tight, the probability of any one candidate running away with the election diminishes -- meaning that all "close" electoral permutations, including ties, become more likely.

However, there is one specific scenario that is driving this outcome. That is the scenario wherein Barack Obama wins the Kerry states plus Iowa, New Mexico and Colorado, but loses New Hampshire. Of the 320 times that our simulation ended in a tie, this particular scenario was responsible 294 times. Indeed, we presently have Obama winning precisely the Kerry states plus Iowa, New Mexico and Colorado, so all that would be needed to make a tie occur is to flip New Hampshire back to McCain, and entirely reasonable possibility.

So the scenario I talked about last week, is the most likely tie scenario. It would be so awesome if it happened.