Friday, October 26, 2012

2012 Election: Polls & Predictions… A narrow and divided victory for Obama?

It’s getting close to the end now and I am starting to get some gut feelings about this one.  Still, overall, I wouldn’t be surprised by either outcome in the presidential race.  However, I would be slightly less surprised if Obama won. 

November 6 is going to be an interesting night.

Game Changers

A little over a week out… Is it too late for any game changers? 

Probably, but a few things do come to mind.  First, Florida looks like it is tightening up again.  If it tilts blue, done.  Early night a week from Tuesday and we all get a good night’s sleep.

Also, as we get a little closer, there is always the possibility that the press starts paying more attention to the fact that it is really going to be tough for Romney to win the Electoral Vote unless the polls are as far off as Dewey / Truman in 1948.  If the media starts talking about Obama’s Electoral Vote Firewall instead of Romney’s questionable momentum (basically manufactured by GOP talking heads, not recent polls), then this could break more in Obama’s direction.

Why?  Because 1% to 2% of these idiotic, undecided swing voters are going to vote for whoever is in the lead because they want to vote for the winner.  Usually this segment is small enough that it does not effect the outcome of the election, but in super tight races?  Who knows?  Usually when super tight races head into election day, they are too close to call and these folks stay home.  But if Obama is looking solid heading into election day, these folks might show up and vote for him, increasing his possible margin of victory.  

Hell, they might even win Obama the popular vote, but more on that later.

Another possibility, especially if the press starts giving more time to Obama’s Electoral Vote advantages…  Romney starts making some last minute, desperate Hail Mary swings through the silly zone.  When this guy goes big, he starts getting strange. 

At this point, though, can any further goofy headlines effect Romney?  Who knows?  But if he is feeling desperate, we may see some interesting moments a la John McCain’s “See, I’m Not To Old To Be President” marathon bus tour on the eve of the election.

Most likely… I predict we might see things firming up a little bit more for Obama in the state by state races, but I think we are pretty much heading into the popular vote within the margin of error, therefore tied.

So yes.  I am going ahead and posting my predictions for November 6. 

I reserve the right to change my mind later if something crazy happens.

Numbers & Predictions: All Hail Nate Silver

2012-10-26. 02.First of all… Let’s jam through the who’s, what’s, and what’s everyone saying…

Five Thirty Eight: Nate Silver's Political Calculus: Oct. 25: The State of the States -

Thursday was a busy day for the polls, with some bright spots for each candidate. But it made clear that Barack Obama maintains a narrow lead in the polling averages in states that would get him to 270 electoral votes. Mr. Obama also remains roughly tied in the polls in two other states, Colorado and Virginia, that could serve as second lines of defense for him if he were to lose a state like Ohio.

The day featured the release of 10 national polls, but there was little in the way of a consistent pattern in them.

Nate Silver's Newest Prediction: 73% Obama!:

Although the race is still close in several swing states, statistical guru Nate Silver’s newest prediction is 73.1% Obama to 26.9% Romney

Nate Silver’s track record on election predictions is pretty solid, so much so that the GOP is going after him, apparently.

Republicans Desperate to Spin Romney as the Front-Runner Are Becoming 'Nate Silver Truthers' | Alternet:

…in recent days, the Romney-Ryan campaign has claimed that it's moving ahead. As Jonathan Chait noted, “This is a bluff. Romney is carefully attempting to project an atmosphere of momentum, in the hopes of winning positive media coverage and, thus, creating a self-fulfilling prophecy.” Despite zero evidence that Romney has made any gains since receiving a healthy bounce from the first debate, reporters appear to be buying it, with a raft of lazy stories about Mitt Romney's supposed “momentum.”

A significant problem for conservatives bent on spinning this alternate reality is New York Times ' polling guru Nate Silver and his 538 forecast model, which called 49 out of 50 states accurately in 2008 and is considered the industry's gold standard (the model also pretty much nailed the 2010 mid-terms).

Yeah, don’t even get me started on the “raft of lazy stories about Mitt Romney's supposed ‘momentum.’”  I called that one back before the first debate.

Democracy In Distress: How Mitt Romney will win the first debate:

…what happens next really depends on media spin.  Not the partisan talking heads, but the producers, writers, editors, reporters and directors out there.

The media wants a story to tell.  If the election is pretty much settled a month out, that leaves four weeks of dead air time…  Which they will fill by trying to create the feeling that the race is much closer than it really is.  The problem?  People will start believing it, and everyone loves a come from behind underdog, right?

Sorry, got distracted there for a second.  The press is dead to me and I mourn them from time to time…

Numbers & Predictions: Nate Silver vs. My Gut

So, Nate’s current prediction... 

2012-10-26. 03.

This map brings the Electoral Vote in with Obama winning 303 to 235.  Even if he only calls 49 out of 50, it is a happy night for Team Obama. 

Unless, of course, Nate blows the Ohio call for Obama. 

Oh.  Oops.  Sorry, that was the conventional wisdom I’ve been hearing from press outlets hungry for a story. 

If only one of Nate’s Obama states go for Romney instead, pulling off the 49 out of 50 prediction from 2008, then Obama still wins “easily.”

Colorado, Virginia, and Ohio pretty much ALL have to shift over to Romney for the U.S. to elect its first Mormon president with a final electoral count of  275 Romney to 263 Obama.  (P.S.  I have no problem with a Mormon President and, I fear, that might be the coolest thing about a Romney Administration…)

Of course, there are other paths to 270 for Romney, but this would be the most likely.

There are also a couple paths that lead to the House of Representatives deciding this one.  Even more unlikely than a Romney presidency, but more likely than in most elections.

Nate’s calling this for Obama with a pretty big margin in the Electoral Vote and a bigger margin in the Popular Vote than I am comfortable with right now. 

I think it’s going to be closer.  How much closer?  2000 close? 

I hope not.  But my Electoral College prediction looks frightenly similar to the 2000 numbers.

My Prediction for the Electoral Vote: Obama Wins, 272 to 266

Right now, my personal prediction looks a lot tighter than Nate’s. 

2012-10-26.  03.

This is the tightest it can be with Obama still coming out on top.  Right now, too many things have to break Romney’s way for him to win, in too many states that are leaning blue.  However, unlike Nate Silver, I do not see Obama sweeping all of the “tied” states right now.  This is my worst case scenario for an Obama victory, but I do not think Obama will break 300, though as of now I think he will get 270.

And I am worried enough about shenanigans in Ohio that I am tossing the state to Romney in my prediction.

Obama-Romney Race May Hinge on 2 Ohio Counties | RealClearPolitics:

If the race for president can be boiled down to two key counties in one key state, then those jurisdictions are Hamilton and Cuyahoga, here in the Buckeye State.

And, as Dan Rather put it on election night in 2000, "This race is tight like a too-small bathing suit on a too-long ride home from the beach."

My Prediction for the Popular Vote: Mitt Romney Wins

As for the popular vote, unless some of the game changers I mentioned above occur, I am predicting that Romney will win the popular vote.  Yeah, I know Nate went with Obama.  But I don’t.

2012-10-26. 05.

I don’t like or trust CNN polling this year.  I think they are swinging towards Romney where possible and I believe that they are holding back on making some pretty solid calls on the electoral map just to keep viewers interested and watching, but I do feel like Romney has a slightly insurmountable lead in the COMPLETELY MEANINGLESS national polls. 

However, CNN is not the only one showing Romney holding steady in the national polls. 

2012-10-26. 08.

The polls are pretty much all over the place, but with Obama only leading in three of eleven and tied in one other, I think it is looking pretty bad for Obama in the popular vote.  But, as I said earlier, it doesn’t matter.  Even if we all slept through our civics classes back in junior high, everyone should be real clear on this after the 2000 election - the national popular vote decides nothing.

This Day In History

So, how does this year compare to years past…

2000 was all over the place, a lot like this year. Below is a link to 10 different polls from this date in 2000. 

Bush led in 8 out of 10 polls by margins ranging from 2% to 7%.  Gore led in two polls, by 2% and 3%.

However, the most interesting numbers in these polls come from the swing between Likely Voters and Registered Voters in Newsweek’s poll.

Bush Gains Back Lead in CNN Poll; Gore Maintains Zogby Poll Lead:

Thursday, Oct. 26, 2000

NewsWeek – likely voters (Oct. 18-20)

  • Bush 48%
    Gore 41%

NewsWeek – all registered voters (Oct. 18-20)

  • Gore 45%
    Bush 42%

If this year’s election is still up for grabs, then I predict it will be decided by which side gets their base out and by which side can motivate swing voters leaning in their direction to actually show up and vote. 

This could be a real problem for both sides.  Obama’s base is somewhat dissatisfied with him not living up to the superhuman expectations built up in 2008, and Romney’s party actually thought about going with winners like Newt Gingrich and Rick Santorum before giving into the inevitable and nominating the only candidate that had with a snowball’s chance in hell of beating Obama.

When it comes down to it, I think the advantage here goes to Romney.  I think the far right’s hatred of Obama will overcome their ambivalence about Romney.  And I don’t think the far left has a motivational force of that caliber working on its side this time around, though the “rape is a gift from God” stories couldn’t have hit at a worse time for the GOP.  But it is probably not enough to balance out the Obama hatred.

However, will that be enough to beat Obama?  Probably not.  In 2004, the far left had this hatred for the GOP candidate working on their side.  They also had their own version of Romney in John Kerry.  It was a close election, it came down to Ohio, there might even have been a few shenanigans in Ohio, but not enough to make a difference in the outcome, most likely.  It was Bush’s night with or without the shenanigans.

Electoral Maps 1972-2008:

The 2004 US Presidential Election

The electoral map shown below depicts the results of the 2004 U.S. presidential election in which George W. Bush defeated John Kerry.  Bush carried 31 states and 50.7% of the popular vote.

2012-10-26. 09.

Here is a look at the 2004 polls from this point in that election.  Today we probably have about a 1% spread between Romney and Obama. Kerry was down by around 2%. All within the margin of error for everything.

2012-10-26. 06.

One thing is for certain, this is no 2008, when Obama comfortably had the popular vote wrapped up by this point.

2012-10-26. 07.

Are We Locked In?  Is This A Done Deal?

Well, in the past, including these elections we are looking at here, there have been some changes in the polls heading into the last week.  Usually whomever leads through October wins, which is what the GOP is counting on right now.

However, this is not always the case.  Of course there was Dewey / Truman in 1948.  The GOP candidate was leading Truman by fairly large margins from the spring on.  The final Gallup poll had Truman losing with 44.5%, and he was behind by about 5% at the end of October.  He ended up winning with 49.9% of the popular vote.

In 1952, in the Gallup poll, Stevenson shot up by about 10% in the last couple weeks of the election, with Eisenhower at 51% in the final survey.  It wasn’t enough and the General won with 55.4% of the popular vote.

1960…  Nixon closed around a 4% gap to a 2% gap in the last month or so, and this momentum continued to election day with Kennedy barely squeaking out a 50.1% victory in the popular vote.  In fact, this election was so close, that in a different day and age, it might have gone like 2000.

In 1968 we had a three party election with Wallace absorbing 13.53% of the popular vote and winning five states (46 electoral votes).  Through October into November, Humphrey closed an 8% gap to about 1% going into election day.  But Nixon’s lead held, of course.

Perhaps the modern election that most resembles the 2012 race is the 1976 contest between Ford and Carter. 

1976 Gallup tracking poll: Ford vs. Carter(very interesting reading!!):

Ford made up additional ground following the third debate in late October, again pulling even. In the final pre-election poll, Gallup's numbers indicated a statistical dead heat among likely voters, with Ford at 49% and Carter 48% (the unallocated numbers had Ford at 47% and Carter at 46%). The actual outcome was 50% for Carter and 48% for Ford. The election was so close that it was not certain that Carter would win until the morning after Election Day.

Then of course, we have the 2000 election… At one point in October, Gore was actually down by 13% in the Gallup survey.  By this week in 2000, Gore had closed the gap to 5% according to Gallup, and he continued to surge (if Gallup’s numbers were right, which is unlikely looking at the chaos in the polls that year). 

Of course, Gore ended up winning the popular vote 48.38% to 47.87%.  Of course, 543,895 popular votes count for nothing compared to Bush’s five extra electoral votes (271-266).

So is this over?  Absolutely not, when it comes to the popular vote.  However, Obama is looking pretty solid in the Electoral College unless the polls are off by 1948 margins, which is pretty unlikely considering the refinements in the polling process over the last 64 years.

Could This Be Another Bush / Gore Style Nightmare?

Short answer, yes.  It is possible. 

Could it break like 1960, where Nixon chose not to contest the counts in several close precincts?  Maybe. 

My gut tells me that it all depends on who is up and who is down.  I suspect the man that will say anything to be president would push it as far as it went in 2000, clinging to any chance at all to be president (for all you tea partiers that think I am talking about Obama, sorry).  Obama, I suspect, is smart enough to see how damaging that process can be to the country, and might not take it that far.

Even in 2004, there were enough questions in Ohio that some thought that Kerry should have called for some recounts.  He, however, chose to accept what was probably inevitable and to go out as a classy winner instead of a sore loser.  This was not like 2000 where it was very likely that more voters did vote for the candidate that lost than the candidate that won in the disputed state.  This was less unlikely to be the case in Ohio in 2004.

Unfortunately for Obama, the elections this year most resembles, poll wise, are 1960, 1976, and 2000.  Like 2000, the polls are a bit all over the place.  Like 1976, we have an fairly unpopular incumbent and an untested outsider hovering within a percentage point of each other.  Like 1960, we have a candidate losing the first debate only to slowly crawl his way back in the polls through the later debates and up into the election. 

The final results for the 1960?  “In the national popular vote, Kennedy beat Nixon by just one tenth of one percentage point (0.1%)—the closest popular-vote margin of the 20th century. In the Electoral College, Kennedy's victory was larger, as he took 303 electoral votes to Nixon's 219 (269 were needed to win).” (Wikipedia) 

303 is also the number Nate Silver is calling for Obama as of the evening of the 26th, according to his map, at least.

So, parallels?  1960, Romney as Kennedy, Obama as Nixon because of the debates…. Romney wins.  1976…  Incumbent loses to challenger.  Romney wins.  2000…  Late surging Gore pulls off a popular vote victory and loses the election.  Romney wins. 

But, Obama fans, take faith in the 2004 election….  In many ways, this year’s election bears more resemblance to that one than any of these others when looking at more than just the polls. 

And our, oh sweet Jesus, Bush wins in the end. 

Related Posts

No comments: