Wednesday, August 20, 2008

The Electoral College Map (8/20/08)

Well, things are getting interesting. [Haven't they been all along?] Recent poll movement toward McCain continues with the incorporation of twelve new polls in eleven states. Blue state margins are tightening, especially in states like Iowa and Minnesota and the toss up states have taken on a decidedly McCain tilt of late. Recently, both Colorado and Nevada have moved toward the Arizona senator. Also, after a pro-Obama blip in the polling in Florida, the Sunshine state has returned to polling numbers that reflect the weighted average we have here. Which is to say, a toss up favoring McCain.

New Polls (Aug. 17-20)
(With Leaners/ Without Leaners)
Survey USA
Univ. of Iowa
Survey USA
New York
North Carolina
Public Policy Polling

In Ohio, though, the back and forth continues. The tie in the latest PPP poll and another McCain advantage in this month's Rasmussen poll of the Buckeye state (one that largely mirrors the firm's poll of the state last month), swings Ohio back over to McCain's side. The switch of those 20 electoral votes brings to fruition the closer feeling the race has taken on of late. In fairness, though, I should mention that Ohio reclaimed its position as the closest state in our averages. After yielding the title to Nevada after last week's Rasmussen poll in the Silver state, Ohio not only shifted over to McCain, but did so by the smallest of margins: 0.07 points. Due to the weight being placed on the most recent poll, a subsequent result that favored McCain by less than 1.5 points or a poll favoring Obama would shift the state back to the Illinois senator. Needless to say, Ohio continues to be on the Watch List (below).

[One other note, given the updated discussion yesterday regarding Rasmussen's reporting of both "leaners" numbers and those without leaners: If the without leaners data are used, Obama still leads in Ohio even with this new poll included.]

Changes (Aug. 17-20)
OhioToss Up ObamaToss Up McCain

That 20 electoral vote shift now brings the tally in the electoral college to 278-260 in Obama's favor. Despite the shift, the election appears to be hinging on the results in the four closest states in our average: Colorado, Nevada, Ohio and Virginia. But given the current dynamics of the race -- seemingly favoring McCain -- you can begin to envision more of those light blue states being brought more seriously into play. That isn't to say that McCain isn't making efforts in any of those states -- he is -- but they are certainly more attainable if the winds are blowing in his direction. Colorado and Nevada are already marked, but states like Michigan and New Hampshire are also worth increased attention if McCain is pushing into the blue states. I exclude Pennsylvania from that discussion because unlike any of the other toss up states favoring Obama, the Keystone state is actually trending toward Obama. Even the lower margins in the most recent polls of Pennsylvania are running above where the weighted average has the state charted. That being the case, the average typically inches up every time there is a poll that runs above the established average.
[Click Map to Enlarge]

At the same time, if the political winds were to shift [What, two wind references in one post? I know, I have a problem.] back in Obama's direction -- and they are likely to do so at least a little with the Democrats' convention next week -- the current trend could reverse itself. McCain, though, seems to be in a good spot now. Obviously, polling is moving in his direction, but what happens when the dynamics of campaign spending are altered. Seth Masket over at Enik Rising has a post up asking that very same question. McCain's time of spending furiously is almost at its end. Following the GOP convention, the Arizona senator will be party to the spending cap placed on him by the federal matching funds system. Currently, he's spending the uncapped money collected for use prior to the general election campaign. Granted, he'll have some help from the coordinated efforts of the Republican National Committee, but will we see any drop for McCain after his convention (Well, not directly after it, but in the period afterward.)? Put differently, will Obama be able to use his decision to opt out of federal funding to his advantage to regain his footing in the race? That's definitely something to consider (Perhaps for the comments section.).

The Electoral College Spectrum*
*Follow the link for a detailed explanation on how to read the Electoral College Spectrum.
**The numbers in the parentheses refer to the number of electoral votes a candidate would have if he won all the states ranked prior to that state. If, for example, McCain won all the states up to and including New Hampshire (all Obama's toss up states, but Michigan), he would have 299 electoral votes. Both candidates numbers are only totaled through their rival's toss up states. In those cases, Obama's number is on the left and McCain's is on the right in italics.
***Colorado is the state where Obama crosses (or McCain would cross) the 270 electoral vote threshold to win the presidential election. That state is referred to as the victory line

Even with the shift in McCain's direction, the Electoral College Spectrum hasn't changed all that much (nor has the map for that matter). What has happened is that you begin to see the state of play differently. We've moved in short order from talking about how far Obama could potentially push into those pink states to which ones McCain may now be able to pull off. That said, surprisingly tight margins recently in Iowa and Minnesota didn't pull either into the toss up category. And those same four states -- Colorado, Nevada, Ohio and Virginia -- remain the most vital components of either candidate amassing 270 electoral votes. Of course, as Allen -- from Election Projection -- aptly said earlier today, Obama still has more paths to victory. That is certainly true, but if things continue on their current trajectory, that may change as well.

The Watch List*
from Toss Up McCain
to McCain lean
Floridafrom Toss Up McCain
to McCain lean
Georgiafrom McCain leanto Strong McCain
Mississippifrom Strong McCainto McCain lean
Nevadafrom Toss Up Obamato Toss Up McCain
New Mexicofrom Obama leanto Toss Up Obama
North Carolinafrom Toss Up McCain
to McCain lean
Ohiofrom Toss Up McCainto Toss Up Obama
Virginiafrom Toss Up McCainto Toss Up Obama
Washingtonfrom Strong Obamato Obama lean
Wisconsinfrom Obama leanto Toss Up Obama
*Weighted Average within a fraction of a point of changing categories.

As for the Watch List, there are a couple of alterations to Sunday's list to note. Minnesota has shifted from being on the list as a state to potenially shift from Strong Obama to an Obama lean to now being completely off the list. However, if you look a the Spectrum above, you'll see that the North Star state has not shifted at all. Truth be told, Minnesota is a victim of the definition of what's included on the Watch. It is no longer within a fraction of a point of switching back to that Strong Obama distinction, but it is within exactly one point of it (...tied with seldom-polled Delaware). Ohio is the only other change. As I mentioned, the Buckeye state remains on the list but is now slated for a potential move toward Obama instead of a move in McCain's direction.

[Note: I purposely avoided the VP topic here. If you'd like to weigh in on the latest speculation, please follow the link to the VP thread immediately below in "Recent Posts". Thanks.]

Recent Posts:
On VP Predictions: Timing and Choices

Is Rasmussen's Inclusion of "Leaners" Affecting the Electoral College Outlook Now? An Update

The New Ohio Poll and McCain's VP Choice


Anonymous said...

Ohio has a lot of "Reagan Democrats" who may reject Obama.
The changing demographics in Nevada and Colorado with growing Mexican population there is what is probably going to sink McCain, even tho he wrote the amnesty bill.
Mexicans for the most part vote for democrats, and that is why the republican party is foolish when they try to pander to the Mexicans.

I agree with your updated map, I think this will be the result on election day.
There might be some states you have as weak McCain that may turn out to be moderate McCain and vice versa, and also with Obama, but the 278-260 result I agree with.

Anonymous said...

The tendency is definitely to look at those McCain lean states as Strong McCain states. All four -- Georgia, South Carolina, South Dakota and Texas -- went overwhelmingly for Bush in 2004. And both South Carolina and South Dakota have been underpolled leaving everyone wanting more proof that they will actually turn out the way it appears now.

Robert said...

Did you see that Obama is trying to do something about frontloading at the Denver convention? See

Anonymous said...

I did.

I'm working on something regarding that now. Should be up in a bit.

Meanwhile, here's that link from Rob.

Unknown said...

Just tracked down a fascinating source of pollster induced error.

SurveyUSA, as part of their screen, only counts responses from voters who are registered. Rasmussen doesn't say, but it's quite possible they do the same.

It's obvious that this undercounts voters under 22, but that's doubtless well known, and they probably jigger the weightings to compensate.

But do they jigger individually for each state? Seven states allow same-day registration: Minnesota, Maine, New Hampshire, North Dakota, Idaho, Wisconsin, and Wyoming. If they don't jigger separately for these states, then they're likely to be seriously underreporting young voters and new voters, and thus Obama voters.

Does this hold up? Hard to say, particularly since many of those states are lightly polled. But there does seem to be some suggestion of a gap between SurveyUSA and Rasmussen results and those of some of the other pollsters for those states.

Something to think about, anyway...

Anonymous said...

This is something that Nate has dealt with in a couple of posts and has been in the back of my mind all along. The younger portion of the electorate appears to be energized this time around but perhaps are not being accurately represented in polls.

This is an idea that popped up a few weeks ago in the context of the poll (Gallup, I think) that showed McCain ahead almost simultaneously with their daily tracker poll that had Obama up 9.

The firms can weight all they want to, but this is an issue that could keep them tweaking that weighting until election day. And we probably won't really know until the postmortems are done on the election.

Unknown said...

Yes, the issue has come up--but has it been discussed in the context of the particular states that do election-day registration? If SurveyUSA uses the same tweaks for Minnesota that they do for a "conventional" registration state like Ohio, they are making a mistake.

I bring this up now because the Minnesota Public Radio poll makes this claim, singling out SurveyUSA.

Anonymous said...

Yeah, of all the states you mentioned, Minnesota is the one where this is likely to be the biggest problem.

I read that MPR release a bit ago and it is interesting to me that they go out of their way to mention on the first page that the race is still fluid despite the 10 point margin. I don't know quite how to take that statement. Is it validation that these recent polls there that have been so tight are more accurate?

What has been great though is that we've gotten three polls from Minnesota right here together. Sure, you could argue that it would be better to have them spaced out a bit, but when all the data is collected at the same time is when these differences between firms get talked about a bit more.

Robert said...

Scott and Josh,

All of that may be so. I think Obama has a much better ground game (getting out the vote) than McCain. McCain also will have trouble juggling independents and his base. Having said that, the pre-election polls will need to be close. The poorer, younger voters that may not be showing up in the polls are also more likely to stay at home if they think Obama doesn't have a chance. If the predictions are this close on election day, I think the election will go to Obama, but the trend lines are beginning to look very similar to those of Kerry back in 2004.

Unknown said...


Superficially, the trend lines look like Kerry-Bush. But the internals are very different. One way to see this is to look at electoral vote's comparison charts. The volatile total is eerily similar. But omit the very close states, and Kerry showed weakness that hasn't appeared for Obama.

So Obama, unlike Kerry on this date, seems to be sitting on about 250 electoral votes. What we're seeing now is evidence is that he could very well lose a close election, but is very unlikely to get blown out. Obama, on the other hand, could win a blow out. That gives McCain no margin for error; he's got a lot of Bush states to defend, while Obama doesn't have a lot of Kerry states that are in peril (realistically, New Hampshire is the biggest question mark.)

Also, note that on this calendar date in 2004 we were after the Democratic convention, but before the Republican one. That means Kerry should have been benefiting from a bounce, but it didn't materialize. If Obama gets no bounce out of the convention, then he's in grave danger of a modest lost. If he gets even an average bounce, then the dynamic is very different from 2004.

Pinyan said...

Useful blog that I'm using (among others) to try to track trends in the predictions. One thing to note however is that the commenter Allen you referenced is actually from, not (that guy is named Scott). Apparently there's only so many different web site names you can come up with for election, umm, projections.

Anonymous said...

Oops. Thanks for the heads up, Pinyan.

My apologies to Allen. A fix is on the way.