Princeton Election Consortium

Innovations in democracy since 2004

Outcome: Biden 306 EV (D+1.2% from toss-up), Senate 50 D (D+1.0%)
Nov 3 polls: Biden 342 EV (D+5.3%), Senate 50-55 D (D+3.9%), House control D+4.6%
Moneyball states: President AZ NE-2 NV, Senate MT ME AK, Legislatures KS TX NC

The paranoid style in progressive vote-counting

November 11th, 2008, 12:17pm by Sam Wang

During a campaign I am avidly attentive to polls and tactical details. But I don’t enjoy Election Night itself. If I could resist the social draw, I’d make an early night of it, then wait until the next day to see the results.

The basic problem is that returns are slow to come in and often misleading. This low quality can foster worries about vote fraud. Yet careful consideration suggests little support for the idea that significant-scale fraud occurred in any major race – including the Alaska Senate contest…

As I wrote a few days ago, pre-election polls had great predictive power this year, as they did in 2004. This is why I advised you to focus on one small state – New Hampshire. It’s small and relatively homogeneous, and could test whether the overall measured spread was correct. It was.

Erratic reporting of results can feed inaccurate thinking about the dynamics of races. Some people seem inclined to overestimate the likelihood of technology problems, suppression, and fraud in voting. This style of thinking may be primed by the 2000 election, a scarring event for progressives and liberals. It is true that Republican officials like to act in ways that reduce the other side’s vote. But it is a long way from known tactics to large-scale fraud. Existing known tactics are likely to make a difference only when races are within one or two percentage points.

This year, several mini-dramas have been resolved in ways that are consistent with polls.

President – Indiana, Missouri, North Carolina. Polls predicted that these would be toss-ups, as indeed they turned out to be. A fourth state, ND, appeared to be near-tied, but polling was sparse and dominated by partisan polling organizations.

President – Virginia. There was consternation in the evening’s count, as the race appeared far closer than the pre-election average of Obama +4%. But after late votes and absentee votes were counted, the final result of Obama +6% was a good match.

Senate – Alaska. This is an interesting case that is currently attracting comment: Why is Stevens leading, if polls showed Begich ahead? Why does the turnout appear so low? In both cases, I believe there is less to the story than meets the eye.

Polls were volatile, a fact that would escape attention if you were only watching who was ahead. Indeed, I had failed to notice this. Begich led by 1-2% on 10/14-19; 22% on 10/28-30, soon after Stevens’s conviction; and 7% on 11/2.

What should we make of the fact that Stevens leads in votes counted so far? First, early voting is showing the same pattern as everywhere else: a tilt toward Democratic candidates. Second, despite speculations of suspiciously low turnout, Election-Day votes (224,000) and uncounted ballots (90,000) add up to 314,000, similar to the 313,000 votes recorded in the 2004 Presidential race. So there’s no evidence for skulduggery. Stevens currently leads by only 3,000 votes. I predict that the final margin, which should be announced tomorrow, will end up in the range of the last two polls, a Begich win by 2% to 7%.

To return to my previous post: The main message was that polls do quite well. We should always keep the simplest hypothesis in mind: – polls reflect sentiments, and so does voting – after the votes are all counted. Unexpected events can happen, and it’s a good thing to watch for them. But we should not necessarily be on a hair trigger.

Tags: 2008 Election

16 Comments so far ↓

  • Ashbel

    Question: Do you know whether anyone has taken a serious analytical look at whether substantial irregularities occurred in the 2004 Ohio presidential vote count? During the campaign, a left-leaning slightly paranoid friend complained frequently that the Republicans somehow rigged the voting machines to turn Kerry votes over to Bush. I know the 2000 Florida vote has been dissected, but am unaware of a legitimate look at Ohio in 2004.

  • Sam Wang

    Ashbel – I don’t know of a serious analytical look. I recall a piece done by Robert F. Kennedy Jr. for Rolling Stone that I found to be quite non-credible. There are others as well, all lacking in rigor. I can easily imagine lively comments in response to this.

    Here’s a plot of 2004 polls in FL/OH/PA. As you can see, it does not support the idea that Kerry should have won Ohio.

  • Ashbel

    Another question: Given the fact that the 2008 polls (at least averaged) were very close to the actual results, does that suggest that Obama’s get-out-the-vote effort was overrated? It seems to me that if Obama had a superior GOTV battle plan, then he should have outperformed the polls. Of course, if both sides have equal GOTV efforts, it’s a wash. If I recall, Bush did somewhat outperform his national polling in 2004, so maybe all that has happened is that the Ds caught up with the Rs this year.

  • Paul

    Re Ashbel & Sam’s follow-up:

    I, too, am skeptical that any widespread election fraud is actually occurring.

    The one notable exception is the haphazard removal of names from voter rolls, which is real and well-documented, and may have turned the 2000 Florida result. But even if it did flip the result, it could only do so because the race was so close in the first place. As Sam points out, these questions of fraud and disenfranchisement are currently only turning elections that are very, very close.

    However, the mere fact that stories of election fraud are so widely believed indicates a serious problem. It is not enough for elections to be accurate; people must *believe* that they are accurate. Lack of trust per se is a serious problem.

    To that end, election process reform (e.g. dumping electronic voting machines) is a good idea regardless of whether elections are accurate. Good quantitative reporting would also go a long way toward assuaging mistrust of elections. Newspapers ought to have quantitative editors review every article, just like copy editors. It’s a sad comment on the state of journalism that one guy with a blog (even a very smart one) can beat the pants off of every major news source with his election analysis.

  • Mike L

    Sam, I concur on the need to obviate paranoia.

    All the same, I have had considerable experience with Australian elections which are run by a national commission (A.E.C.) in a country which adopted the American Federal/state model, including the Senate as a states house.

    Americans consider themselves to be the prime role model for the encouragement of democracy around the globe, yet we continue to allow each state to go its own route for in conducting elections for national offices.

    The latest example is the current trashing of the MN Sec. of State by those who think Mr. Franken has lost and should have conceded regardless of the state’s recount provision. The loser’s party is certain to feel aggrieved and irate.

  • Sam Wang

    Ashbel – I’ve wondered about that. Voting did not deviate from polls in 2004 either, at the peak of the vaunted GOTV effort on the Republican side.

    One possibility is that because such efforts start in advance, the effects are already pre-captured in polls. A second is that GOTV efforts on both sides cancel one another. Finally, the net effect might be 1% or less, which could matter in a super-close races like in Indiana and North Carolina. I’ve heard secondhand that the last is the view of the Obama campaign. However, they have to say that, don’t they.

    Paul – I agree completely that faith in the voting process is currently quite low considering that counting seems to be fairly accurate. In this year’s exit polls, only 49% this year were confident that their votes would be counted accurately.

    Mike L – The counterpart to paranoia on the Democratic side is the view on the Republican side that the process can be gamed. Either case is a corruption of our faith in democratic processes. It would be great if voting were reformed to be more uniform with ultrahigh levels of fairness and reliability.

  • Michael

    I agree completely that faith in the voting process is currently quite low considering that counting seems to be fairly accurate. In this year’s exit polls, only 49% this year were confident that their votes would be counted accurately.

    I think we can attribute this largely to the general innumeracy of the electorate. Many people can’t (or at least don’t) properly contextualize reports of a relative handful of spoiled or disqualified ballots, improper registrations, or what have you. The sensationalist claims of both parties and the media don’t help matters either. And finally, the intensive post-election coverage of those races that are exceedingly close fosters the impression that elections can often turn on a few fraudulent votes or a few disenfranchised voters.

  • Mark

    Regarding GOTV, my impression was that a large part of the Obama campaign effort was directed at early voting. Since these voters had already been to the polls prior to election day, their preferences would have been reflected in polls taken immediately prior to the election. Also, they would have passed likely voter screens.

    A successful GOTV effort primarily directed at early voting could have a significant effect on final margin and still be consitent with pre-election polling.

    To find cases where election day GOTV had an impact, we would need to look at individual states where the effort was greatest. Indiana and Pennsylvania might be examples. I suspect the difference made by election day GOTV efforts is within the MOE of the polls, so this effect would be difficult to measure. (<3%)

  • Richard

    I think it’s a mistake to automatically dismiss widespread voter fraud in Ohio and Florida in both 2000 and 2004. There were stories in Florida in 2000 of destroyed Democratic absentee votes, there were the large-scale disenfranchisement of black voters, and there were the undated military votes that probably shouldn’t have counted. Secy of State Blackwell did his damndest in both elections to suppress Democratic and black voting in 2000 in Ohio, with the result that “likely voter” scenarios in 2004 in Ohio reflected continued voter suppression. And as the above example shows, the 2004 Florida polls showed a close race, which it turned out not to be.

    What we do know is that the kind of updated “likely voter” scenarios that worked in 2008, with increased turnout amongst younger people and African Americans did not work in 2004, and the results reflected that. But what if indeed they should have worked, and those people did go to the polls, and their votes didn’t count?

    Insofar as the popular vote goes in 2004, I’d be curious to see if anyone actually conducted a state by state analysis, with emphasis on those with Republican Secretaries of State, to see if the pre-election polling actually reflected the final tallies, which would lead to the , what was it, 1.4 million lead that Bush eventually wound up with.

    I am aware that as of now, this is a contrarian position, but I’m really not convinced.

  • Matt McIrvin

    Mark Blumenthal did a good series on after the RFK Jr. article came out, taking apart his claims. Blumenthal concluded that there were serious irregularities in Ohio but they probably did not swing the vote from Kerry to Bush.

  • Sam Wang

    Richard – You are imagining a nefarious sequence of events on the weakest of evidence, and in the face of empirical evidence that contradicts your position. I couldn’t ask for a better example of the paranoid style.

    Take a look at a neuroscience op-ed I’ve written that relates to the topic of how beliefs form. Then stop and think about all the evidence. Ask yourself whether it fits better with your idea, or with the converse idea. Be as critical as you are in the direction you currently lean.

  • scotsw

    Well, some good points Sam, but your post seems to live in an academic wonderland completely unhinged from real-world news reports.

    The long lines and shortage of voting machines in heavily democratic precincts in Ohio ’04 were not fabrications of the imaginations of conspiracy theorists. The fact that the Republican Secretary of State was also the co-chair of the Bush campaign in that state is not a paranoid fantasy.

    Your cocksure confidence in the sanctity of numbers ought to take into account the actual documentation of the abuses which took place in Ohio in 2004.

    A statistical study of the reported 2004 Ohio results showed that 6% of precincts reported “impossible” vote totals.

    The testimonials about voter suppression, the stories of disenfranchisement, the irregularities reported at recount — these are legion.

    The outrage with Republican electioneering tactics isn’t paranoia. It’s based on a long catalogue of abuses which seem to be applied consistently over time and space to different elections.

    In other times and places, power was pursued by men of naked ambition with arms and violence. It is not so hard to believe that some men would contrive to grab executive power in the world’s most powerful nation by chicanery.

  • Michael

    …and there were the undated military votes that probably shouldn’t have counted…

    Except that those ballots were not counted. You’re conflating two distinct issues. The same judge who ordered the re-examination of all of the rejected absentee ballots specifically upheld the rejection of those ballots that were not dated.

  • scotsw

    Apologies — the link to the study is broken… There’s plenty of other documentation in the link above that one.

  • scotsw

    I should also mention that my initial concerns about chicanery in Alaska appear to be unfounded — I agree with Sam that the early votes tell most of the story: It appears that turnout was normal, and many Democrats simply voted in the early vote.

    In my long post above, I’m objecting mainly to Sam’s credulousness regarding Ohio ’04, where countless irregularities were simply swept under the rug. Simply because the outcome was withing the realm of plausibility, that does not prove there was no electioneering. The evidence to the contrary, in fact, is mountainous, and needs to be dealt with.

  • Richard

    Sam — I read your article earler. I”ve also interviewed several people on the subject, including Mark Crispin Miller. I can turn it all around, and argue instead that enough facts present themselves that to argue there was not sufficient chicanery to change the results of 2000 and 2004 is itself a belief on your part that can’t be changed. Electoral fraud is rampant around the world, and the fact is that the Bush and Cheney people appear to have no moral qualms about anything, if they can get away with it.

    A few months ago, I interviewed the noted Bosnian writer Aleksander Hemon. He said he felt Americans were naive to think that there weren’t bad people out there. He said a quick examination of what happened in his native Sarajevo puts the lie to that element of American naivete.

    I am not, by the way, arguing that chicanery existed and that the 2004 election was stolen (the 2000 election was clearly stolen by the Supreme Court, if nowhere else). But citing your article about beliefs tells me nothing I wasn’t already aware of.

Leave a Comment