As we noted last week, there is some confusion in the polling data that is creating a bimodal projection as to who will ultimately win the election. The national polls seem to show Romney with a steady lead, and with Obama failing to break 47% as an incumbent. On the other hand, a number of state polls show the race tied or Obama slightly ahead in the key battleground states.
However, when you examine the party breakdown of the polls, they are really reflecting the same reality. There is almost a linear correlation between the level of Obama’s support and the size of the Democrat advantage in the given sample (duh?). Most of the national polls show a reasonable D/R split (not overwhelmingly optimistic, but somewhere between 2004 and 2008 turnout), while most of the state polls that show Obama ahead have a sample that is more advantageous to Democrats than 2008! Does anyone really believe that the turnout will be worse for us than 2008 when the enthusiasm level is completely reversed from 4 years ago?
The one common theme from all the polls is that Romney is winning Independents and other key swing constituencies that are needed to win. Some polls have him winning them by historic margins. As such, the only way Obama can win is if the D/R split is more favorable for him than 2008. And that is exactly what PPP and others are predicting. However, if he performs only slightly worse than 2008 in terms of D vs. R turnout, there is no way he can win, given the strong support for Romney among Independents.
Take a look at this chart put together by Josh Jordan of National Review, and tell me how Obama can win. (note that some of this is a day or two outdated, but the overall picture remains the same):
As you can see, Romney is winning Independents by a whopping 15.4% margin. That is also factoring in the IBD and Ipsos polls, which have shown Obama ahead the entire time, even when he was clearly losing. Moreover, these polls are not overly optimistic on the R/D turnout model for us. The average of the abovementioned polls is D+5. That is a lot closer to the D+7 turnout of 2008 than the D+0 turnout of 2004. It is very likely that with the enthusiasm gap, Republicans will further bridge the gap. In fact, Gallup conducted a survey of almost 10,000 voters (less than 1% margin of error), and found an R+1 advantage in the turnout model. Rasmussen found the same result, although they weight their individual polls halfway between 2004 and 2008 turnout.
The pattern is clear: the more unrealistically optimistic the turnout model of the poll is for Obama, the better the result for him overall. That’s why the national polls that only show Romney with a 1-point lead have a skewed D/R split. The ABC/Washington Post poll moved from D+4 to D+7 over the weekend. Not surprisingly, Romney’s lead declined from 3 points to just 1 point.
The state polls are really portraying the same political landscape as the national polls. The only difference is that more of them show a turnout model that is more Democratic than 2008. They are also dramatically overstating the results of early voting, which automatically gives Democrats a distorted advantage.
Let’s a take a look at an example from over the weekend:
PPP Florida: It’s clear that Florida is all but over. Romney will win by 5-7 points. Yet PPP has Obama up 1. How so? The sample is D+5, even though it was only D+3 in 2008 and R+1 in 2004. Republicans have the best ground game of all in Florida. So does that mean PPP is wrong? No. They are correct if Democrats enjoy an unrealistic turnout. Yet, even PPP admits that Romney is up 7 points with Independents and 8 with Hispanics.
PPP also shows Obama up 4 in Ohio with a crazy turnout model. Yet, Rasmussen shows Romney up 2 (and up 3 among those certain to vote). That’s because their sample shows a realistic split between Ds and Rs.
In conclusion, who would you like to be going into Election Day? Would you rather be the one who is leading among Independents by 15-17 points? Or would you rather be the one who is leading in a hypothetical turnout model that has never happened, even during better times?
This analysis is all encapsulated in the latest report from the GW/Battleground poll, which shows Obama leading nationally by 1-point. If you dig deeper, Romney is far ahead:
In sum, this data indicates this election remains very close on the surface, but the political
environment and the composition of the likely electorate favor Governor Romney. These
factors come into play with our “vote election model” – which takes into account variables
like vote intensity, voters who say they are definite in their vote, and demographics like age
and education. In that snapshot of today’s vote model, Mitt Romney leads Barack Obama by
five-points – 52% to 47%. While that gap can certainly be closed by the ground game of the
Democrats, reports from the field would indicate that not to be the case, and Mitt Romney
may well be heading to a decisive victory.
Folks, Obama’s goose is cooked. Honestly, Republicans are so energized to vote for Romney, that organizations like ours are worried about shutting off that energy come November 7. We need to stay an arm’s length from Romney after the election, so we can hold his feet to the fire. The enthusiasm leading up to the election is heartening (and undeniable), but we must be vigilant of Romney’s policy decisions from day 1. And at this point, we are indeed looking at a President Mitt Romney.
Paid for by Madison Project. Not authorized by any candidate or committee.
© 2017 Madison Project. All rights reserved.
Site by A3K Advertising, Inc.