What Will We Know by 7 PM?; Today's Polls and Final Election Projection: Obama 349, McCain 189
What Will We Know by 7 PM?
Andrew Gelman of Columbia University has taken a recent set of our simulations to look at what may happen conditional on the outcomes of the first states to close their polls at 6 and 7 PM. The bottom line? If those states go roughly as expected (meaning, say, an Obama win in Virginia and a close race in Indiana), we can conclude with almost literal 100 percent certainty that Obama will win the election:
Now, what if the vote margin in Virginia, Indiana, Georgia, South Carolina, and Kentucky were to equal the expected -5.7%? We pipe this assumption through our model by calculating, for each of our 10,000 simulations, the average vote margin in these five states, and then restricting our analysis to the subset of simulations for which this vote margin is within 1 percentage point of its expected value (that is, between -6.7% and -4.7%). Out of our 10,000 simulations, 2800 fall in this range; that is, we predict there is a 28% chance that McCain’s average vote margin in these five states will be between 4.7% and 6.7%. What is of more interest is what happens if this occurs. Considering this subset of simulations, Obama’s expected national popular vote margin is +4.7%, his expected electoral vote total is 343, and the conditional probability of an Obama victory is 100%: he wins the electoral college in all 2800 simulations in this condition.Andrew's full paper can be found here. (And don't forget to buy his book).ADVERTISEMENT
Today's Polls and Final Election Projection: Obama 349, McCain 189
It's Tuesday, November 4th, 2008, Election Day in America. The last polls have straggled in, and show little sign of mercy for John McCain. Barack Obama appears poised for a decisive electoral victory.
Our model projects that Obama will win all states won by John Kerry in 2004, in addition to Iowa, New Mexico, Colorado, Ohio, Virginia, Nevada, Florida and North Carolina, while narrowly losing Missouri and Indiana. These states total 353 electoral votes. Our official projection, which looks at these outcomes probabilistically -- for instance, assigns North Carolina's 15 electoral votes to Obama 59 percent of the time -- comes up with an incrementally more conservative projection of 348.6 electoral votes.
We also project Obama to win the popular vote by 6.1 points; his lead is slightly larger than that in the polls now, but our model accounts for the fact that candidates with large leads in the polls typically underperform their numbers by a small margin on Election Day.
This race appears to have stabilized as of about the time of the second debate in Nashville, Tennessee on October 8th. Since that time, Obama has maintained a national lead of between 6 and 8 points, with little discernible momentum for either candidate. Just as noteworthy is the fact that the number of undecided voters is now very small, representing not much more than 2-3 percent of the electorate. Undecided voters who committed over the past several weeks appear to have broken roughly equally between the two candidates.
Our model forecasts a small third-party vote of between 1 and 2 points total; it is not likely to be a decisive factor in this election except perhaps in Montana, where Ron Paul is on the ballot and may garner 4-5 percent of the vote.
Any forecasting system is only as good as its inputs, and so if the polls are systematically wrong, our projection is subject to error as well. Nevertheless, even as we account for other cycles in which the polling numbers materially missed the national popular vote margin (such as in 1980), a McCain win appears highly unlikely. It is also possible, of course, that the polls are shy in Obama's direction rather than McCain's, in which case a double-digit win is possible.
Nor does McCain appear to have much chance of winning the Electoral College while losing the popular vote; in fact, our model thinks that Obama is slightly more likely to do so. McCain diverted many of his resources to Pennsylvania, a state where he narrowed Obama's margins somewhat, but which our model concludes that Obama is now virtually certain to win. This may have allowed Obama to consolidate his margins in other battleground states, particularly Western states like Colorado and Nevada to which McCain has devoted little recent attention.
Thank you for placing your trust in FiveThirtyEight.com over the course of the past several months. We hope that you will join us both on the website and on HDNet tonight, where I'll be providing election coverage for Dan Rather's team. FiveThirtyEight intends to continue to apply our unique approach to politics after the election, and we hope to have several announcements about our future plans in the coming days and weeks.
Oh, let me count the ways. Almost all of this, by the way, is lifted from Mark Bluemthnal's outstanding Exit Poll FAQ. For the long version, see over there.
1. Exit polls have a much larger intrinsic margin for error than regular polls. This is because of what are known as cluster sampling techniques. Exit polls are not conducted at all precincts, but only at some fraction thereof. Although these precincts are selected at random and are supposed to be reflective of their states as a whole, this introduces another opportunity for error to occur (say, for instance, that a particular precinct has been canvassed especially heavily by one of the campaigns). This makes the margins for error somewhere between 50-90% higher than they would be for comparable telephone surveys.
2. Exit polls have consistently overstated the Democratic share of the vote. Many of you will recall this happening in 2004, when leaked exit polls suggested that John Kerry would have a much better day than he actually had. But this phenomenon was hardly unique to 2004. In 2000, for instance, exit polls had Al Gore winning states like Alabama and Georgia (!). If you go back and watch The War Room, you'll find George Stephanopolous and James Carville gloating over exit polls showing Bill Clinton winning states like Indiana and Texas, which of course he did not win.
3. Exit polls were particularly bad in this year's primaries. They overstated Barack Obama's performance by an average of about 7 points.
4. Exit polls challenge the definition of a random sample. Although the exit polls have theoretically established procedures to collect a random sample -- essentially, having the interviewer approach every nth person who leaves the polling place -- in practice this is hard to execute at a busy polling place, particularly when the pollster may be standing many yards away from the polling place itself because of electioneering laws.
5. Democrats may be more likely to participate in exit polls. Related to items #1 and #4 above, Scott Rasmussen has found that Democrats supporters are more likely to agree to participate in exit polls, probably because they are more enthusiastic about this election.
6. Exit polls may have problems calibrating results from early voting. Contrary to the conventional wisdom, exit polls will attempt account for people who voted before election day in most (although not all) states by means of a random telephone sample of such voters. However, this requires the polling firms to guess at the ratio of early voters to regular ones, and sometimes they do not guess correctly. In Florida in 2000, for instance, there was a significant underestimation of the absentee vote, which that year was a substantially Republican vote, leading to an overestimation of Al Gore's share of the vote, and contributing to the infamous miscall of the state.
7. Exit polls may also miss late voters. By "late" voters I mean persons who come to their polling place in the last couple of hours of the day, after the exit polls are out of the field. Although there is no clear consensus about which types of voters tend to vote later rather than earlier, this adds another way in which the sample may be nonrandom, particularly in precincts with long lines or extended voting hours.
8. "Leaked" exit poll results may not be the genuine article. Sometimes, sources like Matt Drudge and Jim Geraghty have gotten their hands on the actual exit polls collected by the network pools. At other times, they may be reporting data from "first-wave" exit polls, which contain extremely small sample sizes and are not calibrated for their demographics. And at other places on the Internet (though likely not from Gergahty and Drudge, who actually have reasonably good track records), you may see numbers that are completely fabricated.
9. A high-turnout election may make demographic weighting difficult. Just as regular, telephone polls are having difficulty this cycle estimating turnout demographics -- will younger voters and minorities show up in greater numbers? -- the same challenges await exit pollsters. Remember, an exit poll is not a definitive record of what happened at the polling place; it is at best a random sampling.
10. You'll know the actual results soon enough anyway. Have patience, my friends, and consider yourselves lucky: in France, it is illegal to conduct a poll of any kind within 48 hours of the election. But exit polls are really more trouble than they're worth, at least as a predictive tool. An independent panel created by CNN in the wake of the Florida disaster in 2000 recommended that the network completely ignore exit polls when calling particular states. I suggest that you do the same.
posted by Whitney Tilson at 5:36 PM
293 comments
Post a Comment