The New Hampshire Polls: What Happened?

If you’ve already decided that it was a conspiracy, in which Hillary got her friend Ray Buckley to rig the New Hampshire voting machines, you probably don’t care about this. (And you’ll probably want to avoid it, because it mentions something that seems like a pretty big hole in your argument: that exit polling showed the same Hillary win that the machines did.) But for the rest of us obsessives, this article in the Columbia Journalism Review is kind of interesting: The Polls: What the #$!% Happened?

No real answers, but some informed musing about what was and wasn’t a factor.

Update: Oh, and immediately after posting this, I came across the following Joshua Micah Marshall item at TPM: Enough.

There is also something perverse about the quick knee-jerk reaction to assume that any election that dramatically doesn’t go your way was stolen. It stems from the same fidelity to assumption and desire over fact that so many of us have excoriated in the present administration. There is a sullen childishness at work in this thinking that no robust political movement can ever be built on.

What he said.

24 Responses to “The New Hampshire Polls: What Happened?”

  1. J.A.Y.S.O.N. Says:

    What happened is she won by 3%. A win is a win but we need to stop looking for cosmic significance in there.

  2. hossman Says:

    Knee jerk assumptions of fraud/hacking are of course silly — but putting my tin foil hat on: there are some definite oddities in these charts comparing how candidates fared when you categorize their votes by how the votes were counted (by hand or by diebold machines that have proven to be “hackable”)…

    …there’s a meme going around that the republicans were hacking the diebold machines to eliminate any “front runner” in the democratic primaries, because the longer the democrats are attacking eachother, the less time they have to stand together and attack the republicans. if that’s the case Romney is the guy to talk to considering how well he did with diebold machines.

    Taking my tin foil hat off to consider the question “were the machines hacked?” the answer is: “how the fuck should i know?” … the only one that could possibly know/prove the answer to that question at this point is the person who did it (if it were done). BUT! … i’ve seen people claim that the discrepancies shows in the charts linked to above correspond directly with the polling numbers — if so i’d like to see that comparison in an easy to digest format. I’d also like to see some logical rational (non-hacking) explanation for the machine vs. hand discrepancies. Anyone?

    (i’m totally willing to accept some argument like “rich people like clinton more and rich towns use electronic voting.” … provided someone has stats showing that rich people actually tend to like clinton and that the towns using electronic voting tend to be more affluent … i know crap about new hampshire towns.)

  3. shcb Says:

    For what it’s worth I heard today that Hillary does well with more traditional Democrats, union members, older voters, and college educated women. She also does well with the poor. Obama does better with young voters and yuppies, seems there would be some overlap there with the educated women. So it seems Obama is taking a chunk out of the middle and Hillary has the ends of the economic spectrum.

    The conservative thought is that Edwards is helping Obama because Edwards is so leftwing that Obama doesn’t have to lean so far left and can take a bigger chunk out of the middle.

    I’m not making any points here, just throwing it out there.

  4. ymatt Says:

    From a strategy point of view, I think this is perfect. Obama takes Iowa and makes a big splash, vetting his “viability”. Hillary edges him out in NH though, making it clear that it isn’t a done deal, motivating the Obama supporters (young, independent, etc) to get off their asses and get to the voting booths.

  5. shcb Says:

    Michigan will be interesting since that is Obama’s turf but Hillary would traditionally have the edge with the demographics of that area.

  6. shcb Says:

    I meant Illinois

  7. knarlyknight Says:

    Princeton Universty demonstrates the simplicity of hijacking electronic voting machines and urges immediate action: (links to the Princeton site within).

    It’s quite hilarious, in a sad way, that exit polls were considered to be fully reliable right up until electronic voting machines became so prominant in the 2000 election. In New Hampshire, exit polls showed Obama won by a whopping 5%, which in the past would have been considered a certain victory, but the machines said otherwise. So Hillary squeaks past Obama to victoria. Who is doing it, for what reason – I have no idea and I do not care. Its your country and you are losing it to the owners of the voting machines. What a friggin joke.

    How could anyone be so naive (aka moronic) as to cast an election vote on someone else’s hardware/software?


  8. knarlyknight Says:

    * to victory (I have no idea why Hillary would sneak to victoria)

  9. J.A.Y.S.O.N. Says:

    There is a precedent out there for people reporting falsely who they voted for based on racial lines. No one wants to say they didn’t vote for the black guy even if they really don’t want to. Aren’t we supposed to be using Occam’s Razor here?

  10. enkidu Says:

    The item I love from the black box voting Princeton review is the fact that the memory cards that store the os, the apps and final results are all securely locked away… yet the company has a photo on their web site of the key… and the Princeton folks were able to make a perfectly useable key from the photo.

    I am sure their video is still up on youTube.

  11. hossman Says:

    > There is a precedent out there for people reporting
    > falsely who they voted for based on racial lines.
    > No one wants to say they didn’t vote for the
    > black guy even if they really don’t want to.
    > Aren’t we supposed to be using Occam’s Razor here?

    …so people who voted in areas that used electronic voting machines were self conscious about admitting that they “didn’t vote for the black” but people in areas that used paper ballots are all strong and willing to be honest about their actions?

    (I’m not saying you’re wrong, I’m just saying Occam’s Razor doesn’t seem to apply to your argument … if you want to look for the simplest rational explanation you have to find an explanation that can show a correlation between peoples attitudes and whether or not they live in an geographic area using electronic voting machines. maybe it’s geo-economic, maybe it’s geo-racial, … i dunno, but it can’t just be “people as a whole tend to lie”

  12. shcb Says:

    I think it is that people don’t think it is anyone’s business who you voted for. And I agree. If someone asks me I say Independent or I voted for the Communist party, depending on how bad a mood I’m in. The media can wait until the results are in.

  13. knarlyknight Says:

    There will always be a small margin of error for grouchy, anti-social liars like that; the point is that for more than 100 years exit polls have been reliable enough to be considered accurate, then coincidentally with introducing “black box” ballot “counters”, the election results no longer correspond to the exit polls. I’d bet there is a a very simply explanation for that, like what the Princeton folks demonstrated (following a long line of similar demonstrations.)

  14. shcb Says:

    so, are there any proposed solutions? better systems etc?

  15. hossman Says:

    “so, are there any proposed solutions? better systems etc?”

    …frankly, there’s nothing special about the software they’re working on (and have used successfully just recently in a primary). Any system that has the following would be vast improvement to what most states are using:

    1) source code that is publicly available for review by anyone (it doesn’t have to be Open Source — it can be copyrighted with all rights reserved, but it has to be readable by anyone)
    2) anonymity for voters
    2) a paper trail for every vote … so after someone votes they can see, on a piece of paper, the choice they made and that piece of paper is kept in a locked box and can be used to hand verify/recount the digitally computed totals if people question the results.

  16. shcb Says:

    the last time I voted here in Colorado I think that is the system we used, I’m pretty sure there was a cash register type roll of paper under glass so you could read what you voted for as you voted. That sounds reasonable to me. I’m not computer literate enough to know what you’re talking about with the source code but I take it that anyone with that knowlege could check the logic? One of the problems I have with mail in voting is the anonymity you are talking about.

  17. TeacherVet Says:

    Knarly: “…for more than 100 years exit polls have been reliable enough to be considered accurate…”

    Exit polls were “invented” in 1964 by a woman named Ruth Clark. They were not used to project outcomes in U.S. presidential elections until 1984 (Reagan’s second term run).

    They were not used to project victory for Teddy, Cal, Herbert, FDR, Harry, Ike, JFK, LBJ, Dick, Jerry, Jimmuh, or Ron’s first election, and proved their fallibility starting with their initial use in 1984.

  18. knarlyknight Says:

    Well done TV, it appears to me that I am mainly mistaken in my knowledge of exit polls, however I do maintain that their accuracy has been much better than pre-election polls.

    For the purposes of the discussion relating to pre and post election polling, I am wrong.

    Does that mean that there has been no tampering with computer tabulations of election results – despite the demonstrations by Princeton et al about how simply it can be done?

  19. hossman Says:

    shcb: yes, by publicly available source code i mean there needs to be a way for anyone to read teh code and be confident that it doesn’t have any bugs (ie: it counts wrong) or bias (ie: it counts wrong only for certain candidates). Mail in ballots *could* be anonymous, it they didnt’ have the requirement that you sign the envelope. The theory behind signing the envelope is that it serves the same legal purpose as when you sign in on arrival at the polling, you are swearing that you are who you say you are and are entitled to vote … unless your polling place only gets 5 votes, it’s pretty anonymous — with mail in ballots you have to trust that all of the envelopes will be opened and separated from the ballots before the ballots are inspected/counted … if you don’t have that much trust, then you probably can’t trust mail in voting at all because even without signing the ballot you have no way to be sure that the ballot you recieved in the mail isn’t encode with identifying info when it was addressed to you (when voting in person you can pick any ballot you wnat from the stack)

  20. hossman Says:

    For people who are interested in the aftermath and followup to the various discrepencies in NH, two great links…

    The most important soundbite about the entire issue from the arstechnica link…

    “when you control for town size and a few other factors, vote-counting method (Diebold or hand) still correlates with the outcome (Clinton or Obama) to a non-trivial degree. The remaining question is whether there’s some still unknown demographic variable that accounts for the correlation between a district’s vote counting method and who came out ahead there, or whether monkey business was involved. I personally am leaning toward demographics as the final explanation, for various reasons that, in the end, are so vague as to not be worth going into here.”

    Funny … Didn’t i say that a few days ago? (but less eloquently?)

  21. TeacherVet Says:

    Knarly, Princeton’s evidence that something can possibly be done with apparent relative ease isn’t the same as providing conclusive evidence that it actually occurred.

  22. knarlyknight Says:

    Well, duh.

  23. TeacherVet Says:

    My point exactly.

  24. knarlyknight Says:

    This recount sham rigamorol is hilarious! Never seen officials looking so guilty, they look like little kids who are hiding a bunch of cookies fom the cookie jar while ymom asks them what they have behind their backs. Nary a straight answer from any of them, and don’t miss the security tape at the end:

Leave a Reply

You must be logged in to post a comment.