What to expect on Election Day – Nation and Iowa
Not that I have a lot of extra time these days, but I thought I would throw together a quick prognostication about what I see going down on Tuesday.
Most of my thoughts will focus upon the Presidential election, as you can imagine this is the headline of the night.
I went to RealClearPolitics.com a couple weeks ago when I was nervous about polling, and I played with their national Electoral College map after spending a few minutes looking at state polling.
First side note – there is some value in looking at “polling averages” (as RCP does) to detect trends in polling, but not so much in the status of an actual race. The premise of a poll average is that the fishing net for polls is cast wider – but the weakness is that it lumps high-quality polls with…well, junk. As a result, what you get for a state-of-the-race reading is a mediocre return if you only look at that top-line number.
This qualification side note will make more sense later – as I believe a lot of the polling in state races has been assuming a very slanted field toward Democrats on Election Day. In other words – when a poll is taken, the poll publisher has to make a decision on how to “weight” the responses to the polls in order to reflect the actual outcome of the electorate on Election Day. You see, the purpose of an election poll is to forecast the outcome of an election. I know this sounds profound. To do this, the pollster must first forecast who will show up at the polls to vote.
Many of the state-level polls, and some of the national polls have been giving a lot of weight to a Democrat-heavy turnout (or a relatively light Republican turnout), I can only guess based upon the last Presidential election – 2008. As you might recall, this was a “wave” election for Democrats: historically high turnout for Democrats, low for Republicans, and Independents breaking solidly to voting for Democrats.
To leave the side note for a moment, I went to RCP to take these poll averages at face value and get a state-of-the-race and possibly see some trends. My understanding was this – the state-of-the-race using this kind of polling filter would be the most beneficial to re-electing the current President in my view. I guess you could say, it is a ‘conservative’ viewing of the polls. In other words, if Romney can “win” in these conditions, hopefully Election Day will prove better.
My methodology was: if a particular state was decisive by more than 1-2 points, I assigned the state a “leans” designation to that candidate. If it was genuinely too close to call, I left it open. Here you can see what I came up with on 10-22-2012:
Probably the most controversial state, of course is leaving Ohio open. There are a lot of polls out that show it decisively in the column of President Obama, however there are quality (Rasmussen) polls that show the opposite. The most recent polling suggests Obama +2, if you believe that sort of thing.
The point is this: you can see why there has been an insane level of focus upon Ohio in recent weeks – the outcome of this state is clearly impactful on the election just by looking at the map. Ohio’s status is very much in play and it has a lot of votes to offer either challenger…potentially a death blow to either side.
The next thing you might notice is this: the path to victory for Romney/Ryan is not as “narrow” as you might read in the media if they do happen to lose Ohio. There is no doubt the path is more difficult without Ohio – however you can see that if Romney’s polling holds in places like NC, VA, NH, and CO (all very tight swing states), then he would only need to win Iowa plus flip one other “leans Obama” state…like Wisconsin. This is only considered “narrow” in the media because many media types still do not consider these states in Romney’s column – and some don’t even consider FL for Romney despite consistent (admittedly close) polling in Romney’s favor.
Wisconsin polling is also very close. The GOP has demonstrated an excellent ground game there after re-electing Scott Walker twice in two years. And oh by the way, there is that thing about the bottom of the GOP ticket being a popular Congressman from Wisconsin. Further note, GOP early voting in Wisconsin is trending ahead of 2008, and particularly ahead of Democrat gains which is obviously good for us and for Romney/Ryan.
In short, this map is just a path to 270 Electoral Votes in a squeaker.
My next question: is there any evidence for more, and an early-night win for Romney/Ryan?
My next target after Wisconsin is Pennsylvania. This state has often been a tempting prize for Republicans that generally has been accurately been described as “fool’s gold” for the GOP in Presidential elections. It is a narrowly divided electorate that generally favors Democrats for President statewide, but it is very close and dangles 20 electoral votes as a reward. The TV markets, however are expensive as it includes the large metroplexes of Pittsburgh and Philadelphia on either side.
However the counterpoint for Republicans is the demonstrated weakness for the Democrats in the Democrat primary. There was a HUGE “no-vote-turnout” in the Democrat primary in Pennsylvania this spring – and if you were to map that turnout it coincides with some key areas for Democrats tomorrow. Namely: Pittsburgh and even eastern Ohio. You see, Pennsylvania is not just coal country – which has been hit hard by this President’s EPA and other administrative adversaries. Pennsylvania also includes a large section of natural gas deposits in shale rock. Some of the “fracking” debate you might have heard about in media is right through the heart of Pennsylvania in a deposit called the Marcellus Shale. Technology now exists to allow companies to safely harvest vast quantities of natural gas – energy – from these deposits deep underground. This has created jobs and cheap sources of energy in places like Pennsylvania – except where the federal government has stonewalled, delayed, and outright blocked some areas for harvesting this resource.
As you might imagine, this has created some bad feelings in Pennsylvania for this President’s administration – some of that hardship being from otherwise Democrat voters. So much that in their June primary, these Pennsylvania Democrat voters actively voted for “No Candidate” rather than fill a circle for Barack Obama. As a result, an early-night upset win (and sign of good things to come) might be a Romney/Ryan win in Pennsylvania. The last time PA went Republican in a Presidential election was 1988. This is just one example, but one of the most juicy.
I mentioned above that, for whatever reason a lot of the state polling has been giving a large benefit to Democrats for tomorrow’s election. I will dispense with sinister conspiracy theories – most of these polling outfits are looking to make their living from being correct in their polling, so it is in their best interest to accurately predict tomorrow’s result.
I just happen to think that most of them are doing it badly.
This excerpt from a recent post by Dan McLaughlin sums up the overall argument as cleanly as I can imagine:
If Romney goes on to win independents nationally by 5+ points and carry independents by more than a few points in states like Ohio and Wisconsin, he will win.
Obama can only overcome that kind of deficit among independents by decisively winning the partisan turnout battle – indeed, the polls that show him winning nationally or on a state-by-state basis do so almost uniformly by projecting a decisive advantage in Democratic turnout – but when you look for evidence outside of the polling samples themselves of that Democratic turnout advantage, you won’t find it.
In nearly every swing state and national poll you find Romney winning self-identified Independent voters by significant, if not large margins. Yet in many of those same polls, Romney/Ryan still trails Obama or, at best ties Obama in the advertised top line number. As the quote above states, this can only hold true if the electorate at tomorrow’s polls proves to be overwhelmingly biased to the Democrats, as it was in 2008.
One chart in Mr. McLaughlin’s post I cited above is rather telling – the wave election in 2008 was biased to Democrats not really because of some huge Democrat get-out-the-vote effort, although you hear a lot of bluster about that in the media. Democrat turnout was at historically high levels in 2008, but note also that Republican turnout was the lowest since 1984. Note also that Independent voters broke heavily toward Democrats and away from Republicans in 2008.
Case in point: NBC reported a poll of Ohio this past Saturday, showing President Obama +6 there. Chuck Todd of NBC tweeted that if this poll “used the average of the 2004 turnout and 2008 turnout” in this same poll, the result would have changed to Obama +3.
On our OH poll, w/ Obama up 51-45, we ran results with a tighter party ID split (avg of 04 and 08 elections) and it came out 50-47 Obama
— Chuck Todd (@chucktodd) November 3, 2012
The obvious follow-up question would be: what if 2012 turnout is just like 2004? Or a better question: what if 2012 turnout is somewhere between 2004 and the 2010 midterm election – where Independents swung heavily toward Republicans…much the same as Romney is winning Independents in every poll today.
Analyzing that chart again from Mr. McLaughlin, and you might see the 2008 poll result looks like a statistical outlier because of the combination of low Republican turnout, higher Democrat turnout, and a marked swing of Independents to Democrats. Counterpoint, the 2010 is an outlier in the opposite direction. In my view, there is no polling evidence that the electorate will look like 2008 tomorrow…other than the election of 2008.
In fact, Romney/Ryan is drawing crowds of 17,000 and 20,000 supporters in Colorado – not necessarily a state known for its Republican ground game. In fact, in 2010’s midterm election (generally heavy Republican around the country) a solid candidate in Ken Buck narrowly lost his bid for the U.S. Senate in Colorado.
Also note this rally in Bucks County, PA with over 30,000 supporters…an area that went for Obama +9 in 2008 and Kerry +3 in 2004. This area is very much blue-collar Democrat.
My Fearless Prediction: Romney 315, Obama 223; Popular vote margin 52-47%.
I’d also say with recent polling there is an outside chance that even Michigan goes to Romney, in part thanks to Mitt Romney’s Michigan roots…but I lived in Michigan for several years and it always seemed to be just close enough to see, but always out of reach for Republicans. Color me skeptical on Romney in Michigan.
Interestingly enough, I just read this article from true election guru Michael Barone, and it seems we see the state-by-state breakdown and EC totals exactly the same (at least at my first glance). Color me pleased with myself!
I expect an early night victory for Rep. Steve King in IA-CD4. The early absentee voting in this already-Republican district is outpacing the normally-strong Democrat campaign for early voting. Because Republicans generally favor traditional Election Day voting, you would need to believe Christie Vilsack is capturing an inordinate share of Independent or even Republican voters to unseat Rep. King. This will not happen in western Iowa, which already favors Republicans by registration and culture.
IA-CD3 is next most likely to swing for the GOP, again with early voting breakdown only “narrowly” favoring Democrats over Republicans 43-34; and a strong incumbent in Rep. Tom Latham who I believe plays better in mostly-rural SW Iowa compared to Democrat Rep. Leonard Boswell. Boswell will still own Des Moines’ Polk County and a significant portion of Council Bluffs’ Pottawattamie County, but I expect a relatively moderate Latham will do well enough in these areas plus the balance of the district to win the night.
Next up is IA-CD1, where candidate Ben Lange is running a rematch with Rep. Bruce Braley after a very narrow loss in 2010. Mr. Lange recently claimed their polling shows their campaign up 1.5 points in this race. Early voting is 44-28 in favor of Democrats, but another 28% are Independents and I believe Lange’s advantage of running a second time around plus Romney at the top of the ticket might help pull Lange into Congress this year.
Least likely is IA-CD2. Lawyer (but solid candidate) John Archer is trying to unseat Rep. Loebsack in SE Iowa. Again early voting is 45-28 in favor of Democrats, but another 27% of Independents have also voted. I cast this as least-likely for the GOP because Loebsack has shown a habit of pulling close races out in his favor, although by chance I’ve seen a few of Loebsack’s TV ads and at best, they are lame ad-hominem attacks. Standard Democrat fare.
Fearless Iowa Congress prediction: GOP holds one, GOP pickup two, Dems hold one. (GOP +2)
Fearless U.S. Senate prediction: Tied 50-50
Cross-posted at Grundy County Republicans