Pollster.com

Articles and Analysis

 

Can McCain Make an October Comeback?


With Obama taking a clear lead in the national polls with about a month left in the campaign, the question on most minds is whether McCain will be able to make an October comeback and win the election. The problem that McCain faces is that an increasing share of the electorate is committing to one of the candidates at this point, and recent history indicates that few are likely to change their minds.

This is an obvious point, but If McCain is going to get back in this race, he can do so in one of two ways: (1) he can win over undecideds or (2) he can change the minds of those who are currently planning on voting for Obama.

Let's look at the first point. I created the chart below using the super-cool new flash tool that Pollster.com rolled out last week. This chart shows the Pollster.com trend for undecided respondents. At the end of August, the undecided trend was around 10%. By October 1st, that number dropped to just above 5%.

Presently, Obama holds a 5.6% margin over McCain in the Pollster.com trend. Thus, even if you allocated every undecided voter to McCain, it still wouldn't be enough for him to overtake Obama (though it would certainly make for a very close race).

Of course, it is highly unlikely that all of the undecideds will go for McCain, so what about the second option--changing the minds of Obama voters? It turns out that in recent elections, it has been fairly difficult to change peoples' minds in October. The National Election Study conducts a panel survey of voters for each presidential election; they interview respondents face-to-face (no cell phone only problem here) in September and October and then re-interview them after the election. This allows us to get a sense of how common it is for citizens to change their minds in the last month of a campaign. I pulled out the respondents the NES interviewed in September of 2000 and 2004 and the results are in the table below. The columns are the vote preferences expressed by respondents during the September interviews and the rows are the candidates that they actually reported voting for when they were re-interviewed after the election.

nes_votechangers.PNG

In 2004, 94.5% of those who intended to vote for Kerry in September reported having stuck with their choice after the election, compared to 95.7% of those intending to vote for Bush who actually did so. The percentages of those sticking with their candidate are just slightly lower in 2000, but the overwhelming pattern here is that very few voters seem to change their minds in October. In 2004, 4.2% of Kerry supporters changed their minds in October and voted for Bush. Even if McCain manages to get that many defectors, it would only improve his standing by about 2% in the polls (UPDATE: This would also cost Obama 2% in the polls, thereby trimming 4% off Obama's margin). But that assumes that there won't be any defections away from his candidacy. What actually happened in the past two elections is that what few defectors there were largely canceled each other out.

What is also striking is that even people who aren't firmly committed to a candidate appear to end up voting for that candidate in November. The chart below divides respondents into those who said that they were "strong" supporters of their candidate before the election or "not strong" supporters. In the last two elections, at least 80% of the weaker supporters of a candidate stayed with that candidate on election day.

nes_voterchangers2.PNG

The patterns among individual voters are also evident when you look at the aggregate trends in recent presidential elections. You can see this if you look at the charts from Mark's "convention bump" post in August. In 1980, 2000, and 2004, there was virtually no movement in the polls during the final month of the campaign. In 1988, Bush added a little to his lead in October and in 1996 Dole gained some modest ground on Clinton, but in neither case did the October gains make a difference in the outcome. In 1992, Bush gained significantly on Clinton in October, but attracting supporters from the third party candidacy of Perot may have accounted for some of those gains. In any event, Bush still fell short.

Despite the fact that McCain is only down by 5-7% nationally, time is running out and a comeback seems like a tall order. In the new era of partisan polarization, major October shifts in the presidential polls are unlikely. There are few undecided voters left to persuade at this point and in recent elections we've seen that few voters change their minds once they have settled on a candidate.

 

Comments

I think that there a number of factors that are contributing to the increasing spread between the two candidates. The main one being the meltdown of the economy. I feel that many people may associate the Bush administration with the potential McCain administration because of party affiliation. I hear many people say that voting for McCain is like re-electing Bush to another term. Although this may be a bit extreme and unfair comparing an administration that has an approval rating around 20% and McCain's administration, some may make a correlation between the two. In addition, many felt after the debate between senator Obama and senator McCain that Obama was the victor. I think he gained trust from some voter that were indecisive between the two candidates. This may have contributed to Obama's increased lead in the polls over McCain.

____________________

pdollar:

Interesting article. Quick questions though - if 4% of Obama's supporters went to McCain, wouldn't that move McCain 4% relative to Obama in the polls (and not the 2% stated above)? Here goes the reasoning: 4% of Obama supporters is about 2% of the (voting) population. Having them leave Obama for McCain would case a 2% decrease for Obama AND a 2% increase for McCain, thus shrinking the margin by 4%? Anyway, great article, I think the point stands, but I did want to clarify the above...

____________________

Brian Schaffner:

Thanks for the comment "pdollar." You are correct, of course, and I have updated the post to clarify the point.

____________________

Gary Kilbride:

Excellent points. McCain couldn't afford to have such a disastrous close to September. That's far more impactful than some of the commenters here want to concede. Bush in 2000 had a horrible September but Gore bailed him out with the sigh fest in debate #1. Otherwise that race was all but decided. I remember Ed Rollins emphasizing on CNN on the day of that debate that the race was over if Gore were to win the debate.

There is a foundational aspect in play. That's never emphasized enough. A candidate whose party holds the situational advantage is eligible to rebound. That's Obama in 2008. If he were trailing I'd give him a decent chance. But when the natural terrain is tilted against you and you lag at this date, best of luck. It's not unlike Kerry in 2004, with the garbage claims on progressive sites that he was a "closer." You even had Lawrence O'Donnell blabbing that repeatedly on cable. Yeah, Kerry was a closer in a heavily Democratic state. That had no bearing in a national election facing an incumbent with his party in power only one term. Kerry's best chance was for late undecideds to bail him out. But there weren't many undecideds throughout that fall, and the dynamic of a 50/50 country with basically 50/50 approval rating for both Bush and Kerry didn't lend themselves to 75% of undecideds preferring one of the 50/50 choices.

I think the race will naturally tighten to the 2-4 range, below 4 if I had to guess. But barring an Obama gaffe(s) or outside event, McCain is a decided underdog. This isn't 1980, where a hesitant country accepted a candidate based on a sole debate very late in the cycle, on October 28. McCain is a known quantity and it's getting late. The first debate was opportunity for a direct comparison and potential momentum reversal for McCain. I'm feeling much better about Obama's chances now that the debate hurdle has been cleared.

Frankly, I'm amazed we haven't seen Reverend Wright or other desperate negative tactics, something to change the focus and potentially the polling.

____________________

Chicago:

All the polls and analysis here are great, but they are ignoring the elephant in the room...the evidence that people lie to pollsters when a black is on the ballot. What's the best estimate of this 6%, 12%? There are old white people in Ohio and FL etc that have never voted for a black in their life AT ANY OFFICE LEVEL, and you think they will start with President. HELLO reality check.

____________________

boomshak:

Your analysis, of course, includes all of the utterly bogus polls (such as the ones from AP/Time/CBS) that show Obama ahead by 7 or more.

All other polls together only show a lead of 4.2.

Anyone can do an analysis that supports their own foregone conclusion by using erroneous and biased polling as their baseline.

____________________

jac13:

You failed to include another likely reason for the daylight between Obama and McCain: McC's twitchy, erratic, lurching responses during the first week of the economic crisis, and his failed grandstand stunt of "suspending" his campaign (which he didn't actually do) and trying to parachute into the bailout negotiations and "save the day." Meanwhile Obama took a more measured approach and proposed conditions for the bailout bill that ultimately ended up being belatedly endorsed by McCain and included in the bill. IMO McCain unwittingly helped Obama with one of his perceived weaknesses: ability to manage a crisis.

The game ain't over, for sure, but we're in the bottom of the 9th.

____________________

mysticlaker:

Boom Idiot. Go away.

____________________

DemocratAbroad:

McCain could have started a comeback if he had simply addressed the Senate before the bailout vote, making sure he had a catchy soundbite for the evening news.

Instead, voters were left with his campaign panic, his wanting to fire an official whom the POTUS can't fire, his allying himself with petulant House Republicans one minute, then endorsing the bill the next, even taking credit for its creation, when in reality his role was minimal at best. McCain could have neutralized all of this confusion with a good five minute speech in the Senate, but he chose not to do so.

I think the race will narrow again as anxiety over the bailout fades in the coming weeks, but that depends a lot more on McCain than Obama. If McCain doesn't manage to contain his scratchy voiced, angry geriatric persona, he's going to alienate more and more undecided voters and insure an even more comfortable victory for Obama/Biden.

____________________

m913:

@Chicago

I have seen some comments about the so-called “Bradley effect”—that people have lied to the pollsters, saying they will vote for a black candidate and then, in the privacy of the voting booth, voting for someone else. I never understood it. Why lie to the pollster?

In any case, in this election cycle, I think that for every closet racist who has distorted the polling by lying about favoring Obama, there is probably another voter completely missed by the pollsters (cell phone, poor demographic weighting) who will vote enthusiastically for the first black POTUS.

Net effect—no Bradley effect.

____________________

djlewis:

When comparing to 2004, remember the Bin Laden effect.

In case you've forgotten, Osama Bin Laden "endorsed" Kerry a couple of weeks before the election. There was an immediate shift of several points toward -- guess who -- Bush. A recent report on conversations inside the CIA confirms what they never said publicly, that Bin Laden's move was an obvious ploy to help elect -- guess who. And, of course, lots of voters fell for it, and the "endorsement" arguably won the election for Bush.

So any comparisons between 2004 and 2008 regarding October shifts are really pretty meaningless... unless, of course, you expect another event like that. But such events and their effects are inherently unforecastable.

____________________

Adrian B:

For the objective statisticians:

The outliers of Obama's lead are (as of this morning) +4 and +11. All the other polls make it between +6 and +8.

You can argue about weighting and ID but when you have around 12 polls you can see the clear pattern and it almost eliminates a margin of error (and the trend lines on Pollster really help showing this up - they have Obama +6)

For the partisans - make up whatever you want to believe!

____________________

Wahoo:

@Boomshack. Polls have always had a Republican bias as does most of the mainstream media in the US. You cannot have it both ways. When they start to say Obama leads ... the McCain ticket needs to worry.

@Chicago ... I am afraid that you are right to a degree. I hope that that degree is small, but it may be larger than a democratic nation should ever admit.

RE: Bidden v Palin. I think that the bias of the moderator Ifill has been negated by a slick move by McCain. He has moved quickly to challenge the moderator to be unbiased. Which of course will work.

I think that the US will be in for a huge shock once that the real Palin actually comes out (if she ever let's that happen before the election). As a UK observer she reminds me of Mrs Thatcher. Watch out is all I will say. Thatcher was NOT the loved and worldly wise leader to most people in the UK, she was an extremist and a dangerous person. There is an air of embarrassment when her name is mentioned in the UK. Thankfully those extremist days of the 80's are over. Palin will try reignite that extremism. Something not many mention is that McCain is an old guy, he could die any moment ... is Palin really the President that you want?

____________________

KS Rose:

I'd like to mention the effect of observation in this study. As someone who's studied cultural anthropology it seems to me that there could be a significant effect of interviewing process itself. When you're talking to someone who's asking you questions like this face to face and who intends to talk to you again later, it is likely that you 'firm up' your position by making a commitment to the future interview as well as examining more deeply how you feel about a candidate vs someone that just answers a phone.

I think it's not necessarily true that you can extrapolate from these results to the general public.

____________________

ericsp28:

Let's compare the current map to the electoral map from 2004. All but 2 of the toss up states are states that Bush won. Recent polling in MN seems to indicate that Obama has a comfortable lead there, so I think it is safe to go ahead and move that back to Obama's column. We haven't seen any polls from NH in over a week, so that one is hard to measure. Obama had a very slight lead at that time and I suspect that the national trend will hold there as well, though I concede that McCain may have a slight demographic advantage there and could take the state if he wanted to spend the time and money on it. I will be generous for the sake of this analysis and give NH to McCain, even though I doubt that will actually happen.

I also think it is safe to say that IA and NM will flip to blue this election cycle. McCain hasn't consistently held any lead in either state going back as far as they have archived polls on this site.

That leaves NV, CO, MO, IN, OH, VA, NC, and FL as toss ups on the current pollster.com map. In the scenario I've outlined above, the ONLY state McCain can afford to lose and still have a chance to win the election is NV. If Obama flips only CO it will end with an EC tie and go to the House where, presumably, Obama would have the advantage (although I don't necessarily agree that that is a foregone conclusion). If McCain loses any other one state in that list then he cannot win the election. Also, if he loses any pair of those states at all, he cannot win the election.

Now keep in mind that recent polling seems to indicate that Obama has a comfortable lead in OH, FL and VA, a slight lead in NC and CO.

Undoubtedly, there is still a lot of time for things to change, and a game changing event could happen at any time between now and election day. But at this point in the election cycle, I can't imagine a more advantageous position for Obama to be in.

____________________

parker:

In addition to undecideds and winning over Obama supporters, what about a third category over potnetial voters - those who weren't going to vote? Couldn't a larger turnout work in McCain's favor in some areas?

____________________

drdr:

Shorter boomshak: If you throw out the polls I don't like, the results are more favorable to my candidate.

I believe the pollster.com approach of averaging all polls to minimize the effects of outliers is more intellectually honest.

____________________

Liza:

Dear trolls:
#1 - There is no inherent advantage in portraying a candidate further ahead than he actually is. In fact it does the campaign a disservice by reducing attention to states where they may be more of a shortfall than a particular poll represents.
#2 - This is not a sports event. It is not a supreme course case. It is a representative democracy. If candidate A wins, it simply means there are more people (electorates) who voted for him than for candidate B. It does not mean he is stronger, more moral, or even right. Let's keep this in perspective.

____________________

Chicago and m913, there is a recent study posted on my site on the disappearing Bradley effect. Basically, the size of the Bradley effect was 2-3 points up until the mid-1990s. However, since then it has disappeared. So I don't think there is a pony in here for McCain. Take a look.

Sam Wang
Princeton Election Consortium

____________________

NMMatt:

Palin as Maggie Thatcher. Wow. Thanks for the laugh. I haven't heard anything that ridiculous since, well, maybe when Dan Quayle compared himself to JFK.

____________________

TrojanWave:

Sam beat me to the punch, but you should also have a look at Nate's analysis on 538 as well as his metrics for allocating undecided voters. Although there is no overall Bradley effect in post-2000 elections, based on the Democratic primary results it may be the case that Obama underperforms his polling numbers slightly in Appalachia and overperforms them in states bordering Illinois and states with a large black population. These effects may not be due to actual lying to pollsters but because people who are still undecided within a couple days of the election tend to break for candidates who are more demographically similar to them. So it's possible that Obama may run a couple points worse than his polling in OH, PA, and MI, but he is equally likely to run a couple points better in WI, MN, and VA.

____________________

Stav:

There seem also to be interesting anecdotal hints that the Bradley effect may now be, as Allen Toussaint might say, a river in reverse.

Specifically, I've read repeatedly in news accounts comments by voters in conservative areas to the effect that they're thinking of voting for Obama but afraid to tell anybody about it. (There was one such example in yesterday's Times, reporting on a Florida rally held by Wes Clark.)

Of course there's no way to measure this thing from scattered anecdotal reports, but it does make a certain amount of common sense. A trailing indicator, perhaps, of the prevailing attitudes about race in America: many people have moved with the times but are afraid their friends and neighbors haven't moved with them. When they see other Obama buttons in a crowd of their peers, they are pleasantly surprised.

____________________

PeterPrinciple:

"Frankly, I'm amazed we haven't seen Reverend Wright or other desperate negative tactics, something to change the focus and potentially the polling."

We're getting near the point where McCain is going to have to decided how he wants to go out -- in a dignified way, making a principled case against Obama that might resonate down the road (although not to McCain's own benefit), or kicking, clawing and screaming, like a demented old man raging against the dying of the light.

Given McCain's personality, and his behavior in the campaign to date, I think it's pretty obvious which way he'll go. Which means we're likely to be hearing a LOT more in the next few weeks about Rev. Wright, the Weatherman, Obama's half-brother in Kenya, etc.

This will almost certainly fail -- this being one of those rare elections (like '80 and '92) when voters are demanding the candidates talk about their needs, and won't tolerate the carnie sideshow stuff. It will also diminish McCain to the point where even some of his biggest media boosters may wish he would just disappear from public life.

But, like the William Holden character says of Howard Beale (a.k.a. the Mad Prophet of the Airwaves) in Network: "If that's the way he wants to go out, that's the way he's going out."

Politics as late-life crisis.

____________________

Tri:

I'm not sure the polls have truly reflected the uniqueness of this race. Which is, well, race. The exit polls in 04 indicated the Kerry was going to win, almost comfortably. Obviously, he didn't. It's been theorized that it was because some of those polled was too embarassed to say they voted for Bush, so they lied. It's possible that many people say Obama, when in fact, they have no intention of voting for him because of his race. And are too embarassed to say so. So for me, a black man becoming the president? I'll believe it when I see it (and I strongly hope I do come Nov).

____________________

Tri:

Oops, didn't read all the comments before posting. Looks like what I posted earlier has already been brought up.

____________________

An Rev.Wright/Ayers strategy will also depend on how the MSM will deal with that. They might be fed up with it.

The basic problem is that this is a guilt by association tactic, while Palin with Trooper-gate and McCain with Keating5 are personally involved and knee deep in sh!t... However, to see this as a problem, this requires rational thinking from a lot of people, so ... everybody's guess.

Besides, they already have ads out on tying Obama to Blagovich and one of the Daley brothers (who is squeeky clean). He's still sliding.

About the bradley effect, yes it might be gone. OTOH, that might also be a deceptive liberal pipe dream, drinking the koolaid, so it also might be even BIGGER than ever. Then again, there might indeed be a Cell-phone effect. We'll find out ... in Nov.

What worries me are two things:
1, An electronic fraud with the voting machines
and
2, some desperate rednecks with high powered ballistic devices and scopes, bend on "Saving this Christian Nation and the White Race, as Deemed By Our Lord and Saviour, etc." Especially since mercenary companies and other defense contractor will lose a lot of cash, come a Dem win, AND have the means to pull it off, including Intelligence community contacts to provide some sorta getaway, while serving up some patsy to take the fall.

I guess it's up to god and the Secret Service now.

____________________

gcazorzi:

what if: 94,5% of Kerrys voters had declared their preference for him before, but only 70% of those that declared their preference for Kerry actually voted, while 85% of Bush supporters actually voted for him? The factor of fidelity might be secondary to the motivation to go to the poll on the election day.
Mc Cain has a third option not mentioned in the article which is: who will manage to get as much voters to the polls on election day, this might be crucial. This is the target of the Mc Cain campaign, based on fear: get all your potential voters to vote, to avoid the disaster of, let's say, a black, liberal, untruthful democrat winning the elections. Obama has to answer by looking as much presidential as possible as he did in the first debate and avoid as much as possible a dirty campaign.

____________________



Post a comment




Please be patient while your comment posts - sometimes it takes a minute or two. To check your comment, please wait 60 seconds and click your browser's refresh button. Note that comments with three or more hyperlinks will be held for approval.

MAP - US, AL, AK, AZ, AR, CA, CO, CT, DE, FL, GA, HI, ID, IL, IN, IA, KS, KY, LA, ME, MD, MA, MI, MN, MS, MO, MT, NE, NV, NH, NJ, NM, NY, NC, ND, OH, OK, OR, PA, RI, SC, SD, TN, TX, UT, VT, VA, WA, WV, WI, WY, PR