Jay has done the heavy lifting on this model and has posted the details of the 2008 forecast on his web page. The results are summarized in this map.
Based on the September Poll Model, we project an Obama Electoral College win with 336 votes to McCain's 202 votes. Projecting the national popular vote based on predicted state margins and assumed state-level contributions to the national vote, we get Obama with 51.8% of the two-party vote.
Coming attraction: We will soon be unveiling an October Model that tracks changes in projected electoral votes as new polls come out.
Update: Missouri but not Virginia? Several readers have asked in comments why we are calling Missouri but not Virginia for Obama. Good question. The short answer is because that's what the model projects. Sometimes when you run a model like this you get results that make you scratch your head and wonder what's going on. But if you look under the hood of the model, it's somewhat easier to understand.
As described above, the model considers two state-level variables, the average poll result from September and a lagged vote variable from previous elections (averaged over four prior elections). On the lagged vote variable, Democrats have done better in Missouri than in Virginia, so this would push it toward the Obama column. On the poll variable, Obama did better in Virginia than in Missouri but not by enough, given Virginia's voting history to call the state in his favor. While Obama was leading in Virginia polls by the end of the month, he trailed McCain earlier in the month, and the month-long average was just about dead even. In the end, the data from September polls and lagged votes together pushed Missouri just into the Obama column and Virginia just into the McCain column.
No doubt part of the surprise here is due to Obama's strong performance in recent polls in Virginia. But remember, his biggest gains have come in the first week and a half of October and are not incorporated in this model. As promised, an October model is coming soon to a blog near you!
14 comments:
The real issue is not how well Obama or McCain might do state-by-state, but that we shouldn't have battleground states and spectator states in the first place. Every vote in every state should be politically relevant in a presidential election. And, every vote should be equal. We should have a national popular vote for President in which the White House goes to the candidate who gets the most popular votes in all 50 states.
The National Popular Vote bill would guarantee the Presidency to the candidate who receives the most popular votes in all 50 states (and DC). The bill would take effect only when enacted, in identical form, by states possessing a majority of the electoral vote -- that is, enough electoral votes to elect a President (270 of 538). When the bill comes into effect, all the electoral votes from those states would be awarded to the presidential candidate who receives the most popular votes in all 50 states (and DC).
Because of state-by-state enacted rules for winner-take-all awarding of their electoral votes, recent candidates with limited funds have concentrated their attention on a handful of closely divided "battleground" states. In 2004 two-thirds of the visits and money were focused in just six states; 88% on 9 states, and 99% of the money went to just 16 states. Two-thirds of the states and people have been merely spectators to the presidential election.
Another shortcoming of the current system is that a candidate can win the Presidency without winning the most popular votes nationwide.
The National Popular Vote bill has passed 21 state legislative chambers, including one house in Arkansas, Colorado, Maine, North Carolina, and Washington, and both houses in California, Hawaii, Illinois, New Jersey, Maryland, Massachusetts, Rhode Island, and Vermont. The bill has been enacted by Hawaii, Illinois, New Jersey, and Maryland. These four states possess 50 electoral votes-- 19% of the 270 necessary to bring the law into effect.
See http://www.NationalPopularVote.com
susan
weird you guys have him winning Missouri but not Virginia
Surprising to say the least
Obama will win Virginia, he has over 40 field offices all over the state, with voters excited about his campaign.
I find the whole MO - VA thing weird, too.
I, too, find it unlikely that Obama will win Missouri but lose Virginia.
I strongly disagree that Obama would win Missouri while losing Virginia or North Carolina. THat's ludicrous. Moreover, I don't recall September polls for Missouri being favorable to Obama. North Carolina and Virginia are MUCH more likely to go for Obama than Missouri. Finally, could you list the polls you used for your research, and did you average out polls during the entire month of September, or took the average in mid September or some other point in the month?
Susan, we have "battleground" and "spectator" states because some states are a near certainty to vote for one candidate and some states are legitimate tossups. It would be foolish, for example, for McCain to spend time and money on his home state or on Texas. It would be just as foolish for Obama to focus on Chicago or Massachusetts.
Abolishing the Electoral College might actually increase the focus on a few states. If nationwide popular is the deciding facotr, then candidates have every reason to focus time and money on the most urban and most populated states. States like Alaska, Delware, Hawaii, the plains states and others would be cast into permanent campaign invisiblity.
More, fueled by the need to find favor in urban areas, elected politicians would have every incentive to divert resources to those areas.
Do, while I agree that going to a nationwide popular vote seems, at first, to have a lot going for it, on a closer look it brings with it some baggage that we may not want to lug around.
Looks like other comments agree with what I wanted to say. Missouri before Virginia--I have trouble with that one. NC could be a surprise. Who knows perhaps
another state will pop out for Obama that we haven't be focusing on--like West Virginia....Polls have gotten quite a bit closer there.
I'm pleased to see that your numbers on the popular vote agree with fivethirtyeight.com. They currently have him winning 51.9% of the popular vote - but that includes all polls up to and including today (most likely all the polls from today).
Additionally, they MO as a toss-up with Obama winning 54% of the time (they do 10,000 runs of the model) and VA as a likely DEM with him winning 89% of the time.
I too, find it difficult to imagine Obama carrying one state but not the other.
One would assume your October projections will take recent polling into account.
Just to back up what Tom has said about Virginia and Missouri. Over the past four elections the average Democratic share of the 2PPV in Missouri was 51.18%. That is no doubt due to Clinton's strong showing in that state in '92 and '96. In Virginia, the average was 47.03. In other words, Virginia has been much more consistently Republican, and the model is picking that up.
The thing you also have to keep in mind is that these state win projections are simply based on the point estimates from the model, which is why we also report the state win probabilities. The model estimates that Obama will win Missouri with just 50.48% of the vote, certainly within typical margins of error, and that is why it only has a predicted win probability of 57%. In other words, it would still be considered a tossup.
We'll have to see what happens in the October forecast. I should point out that in 2004, our September forecast mispredicted Bush wins in 3 states (NH, PA and WI). The October forecast correctly put those in the Kerry column.
Tom/Jay,
What does the model look like when you run it without the past elections variable or scale that back to just the last two presidential elections instead of four?
I guess another way of putting this is to ask how much of the variation in the DV you find that variable to be explaining?
Thanks to both of you for posting this.
Josh,
Removing the prior vote variable from the model shrinks the adjusted R-Square from .917 to .894 and increases the standard error of the estimate from 2.43 to 2.75.
Not a huge impact, but enough to make a big difference at the margins.
It's always nice when you can not only be informed, but also entertained!
Thanks for share this information, It’s great to see good information being shared.Wonderful article, very well explained.
Post a Comment