Daily Poll Watch: My favorite analyst says likelihood of Obama winning ranges from 89 to 97 percent
On several occasions in recent days, I’ve argued here that the best way to track the presidential race at this late stage is to pay more attention to polls in the individual states — especially the so-called battleground states — than to national polls.
And the best way to follow state polls is to read the analyses offered by Sam Wang at the Princeton Election Consortium and Nate Silver at FiveThirtyEight. Their methods differ somewhat, but they’re both good at cutting through the mainstream media nonsense peddled by journalists who don’t understand polling trends and how they relate to the Electoral College.
I’ve been a big fan of Nate Silver since the presidential race of four years ago. But lately, I’ve become an even greater disciple of Sam Wang. His methodology is arcane (see HERE), but his record of accuracy is peerless.
Beginning today, Wang is featuring two numbers at the top of his Web site dealing with the probability that President Obama will win re-election. He EXPLAINS:
We give two probabilities, which are built on the same assumptions that went into calculating the “strike zones” in the history graph. The “Random Drift” number is a minimum (conservative) probability, and the “Prediction” is a maximum probability. In the coming 10 days, the two numbers will converge…
Both predictions (“random drift” and “prediction”) start from a current snapshot of polling conditions, the Meta-Analysis of State Polls which forms the core of this site. The snapshot is listed in the top line above. It is currently Obama 297 EV, Romney 241 EV, Meta-margin Obama +1.96%. This predicts what would happen in an election held today.
To calculate this snapshot, we (a) use recent polls for each state (3 polls or 7 days, whichever is greater) to calculate the probability that one candidate is ahead, (b) calculate the exact distribution of all 2^51 = 2.3 quadrillion outcomes, measured in terms of electoral votes (EV), and (c) take the median of the distribution to get the expected EV count.
In addition, we calculate the amount by which polls must swing overall to create a perfect toss-up. This quantity is just like a two-candidate margin that people are used to seeing in polls, so we call it the Meta-Margin. Both the EV estimator and Meta-Margin are extremely precise, and performed very well on Election Eve in 2004 and 2008.
If you read Wang’s entire post, I strongly urge that you also read the comments following therefrom. They’re mostly pretty intelligent.