PrintNovember 4, 2008
Election Prediction
The final deluge of national tracking polls and state polls has arrived, and they reveal a most unusual picture. Barack Obama's lead in the national tracking polls has widened by almost 2% in the final days, now standing at 7.4% in the realclearpoltics.com average. Put quite simply, it is inconceivable that a candidate could win the national popular vote by over 7% (conceivably a 10 million vote margin if the turnout hits 140 million), and lose the Electoral College.
And yet, on a state by state level in the key battleground states, the race is much closer and appears to have tightened in the last few days. This is not to say that Obama is in any serious danger of losing. But the national margin may be less indicative of the state of the race than in prior years, due to a wide imbalance in what I and other call "wasted votes" or excess votes in landslide states. In 2004, George Bush won big margins in many Southern states and in Indiana, the upper Midwest and Rocky Mountain states. This year, John McCain could win almost all of the states in these regions, but even if he does, his margin of victory is likely to be smaller in every state he wins than the margin Bush obtained.
On the other hand, Obama is headed for enormous margins in many states: California, Illinois, New York, Maryland, New Jersey and Massachusetts among them, and far bigger margins than John Kerry won in some other states such as Michigan, Minnesota, Wisconsin, Washington and Oregon. In just three states: Illinois, California and New York, Obama will likely pile up a margin of between six and seven million votes. Think about it this way: if the final popular vote margin turns out to be 5%, and turnout is 140 million, then these three states could provide roughly 80% to 100% of Obama's total popular vote margin. If Obama wins by 7%, they would provide 60 to 70% of the total margin.
In the battleground states, almost all red states, the race is much closer. John McCain's problem is that he needs to win virtually every closely contested state to win, and that is unlikely. In my most recent article, I referred to tiers, and I still find that a helpful way to look at the race. Two red states -- Iowa and New Mexico, with 12 Electoral College votes between them, seem safely in Obama's corner. The most recent polling in Iowa (7) has given Obama a double digit lead in every survey. In New Mexico (5), it is a bit closer, but Obama's lead ranges from 7 to 10 points in the non-partisan surveys.
Assuming Obama holds all the Kerry-won blue states, these two red states takeaways bring Obama to 264 Electoral College votes. The next most likely red state to fall is Nevada (5). And Nevada would be enough, for with a 269-269 tie, the House of Representatives would likely give a victory to Obama. So if McCain cannot turn a blue state, he must win Nevada. The closest McCain has come in any recent Nevada poll is down 4, and he is further behind in several of them.
The two blue states still being contested both show solid Obama leads. Pennsylvania has clearly tightened in the last week, and many surveys now show Obama up by 4 to 6 points (he led by double digits only a week ago). In New Hampshire, other than Rasmussen which has Obama up 7, most polls have Obama up more than that.
McCain also trails in Virginia (13) and Colorado (9). Both states, I think, will be closer than some of the Obama partisans believe is the case. Rasmussen has had Obama up 4 in both states, in each of the last two surveys. There are other polls showing Obama up by 4 to 6 in Virginia, and 5 in Colorado. I think Colorado is a bit safer for Obama than Virginia.
I feel fairly confident in predicting that Obama will win Iowa, New Mexico, Nevada, Virginia, Colorado, Pennsylvania, and New Hampshire. If Obama wins no other red states, he will be at 291. What this mean is that if McCain wins all the other closely contested states -- Florida (27), Ohio (20), North Carolina (15), Indiana (11), Missouri (11), Georgia (15), Montana (3), North Dakota (3), he will fall short. Simply put, for McCain to win, he needs to hold all those state just listed and also win either Nevada or New Hampshire, and either Pennsylvania or both Colorado and Virginia. To say this is a tall task is to greatly understate the odds.
I think Obama has an edge in Ohio due to organization, and help from the highly partisan Secretary of State Jennifer Brunner, who has done everything possible to allow fraudulent votes to come in, and keep Republican votes out. I am less certain about Florida, Missouri and North Carolina. I think McCain will win narrowly in Indiana and Georgia. If forced to make a pick, I will give North Carolina and Missouri to McCain and Florida to Obama.
Net, the Electoral College comes out 338-200 for Obama. I think the real range for Obama's victory is from 291-247 in the best case for McCain to 364-174 in the worst case. If Virginia is too close to call tomorrow night, then McCain may make it a long night. If Indiana is too close to call, or falls to Obama, it will be a landslide in the Electoral College, similar to Clinton's wins in 1992 and 1996 (370 and 379 Electoral College votes, respectively). All things being equal (whatever that means), Bill Clinton won by near 6% in 1992, and by over 8% in 1996, and if Obama's margin falls in that range, a big Electoral College win should not be a big surprise. If the Electoral College race turns out to be close, despite a significant popular vote margin for Obama, it is because of the distortion of the wasted votes described earlier. If McCain loses by 3-5%, and escapes with a narrow Electoral College win, I do not think I would want to be a policeman in Grant Park Tuesday night.
As for the Senate, I think the Democrats will pick up 7 seats to get to 58, if you count Joe Lieberman on the Democrats' side. The 7 pickups are in Virginia, New Mexico, Colorado, New Hampshire, Alaska, North Carolina, and Oregon. If Republicans have a good night, and win some close ones (they lost them all in 2006 except for Tennessee), they could hold onto Oregon (Gordon Smith), and possibly North Carolina (Dole). Stevens in Alaska and Sununu in New Hampshire are behind by more than the margin of error, and have a slim shot at holding on.
To get to 60, the Democrats would have to win Minnesota and Georgia. I think they will fall short in both. In two close races, both Norm Coleman in Minnesota and Gordon Smith in Oregon, despite moderate voting records, have been targeted with tens of millions in negative ads portraying them as Bush clones, the same strategy the Obama campaign has used to falsely tar John McCain with $300 million in ads. In the case of Franken, his entire campaign has been a savage smear job against Coleman, and Franken's election to the Senate, were it to occur, would make Minnesota even more of a national laughing stock that when it elected Jesse Ventura governor.
Richard Baehr is chief political correspondent of American Thinker.
