Statistics

Predict who will win the US Presidential Race

When you have a chance, please log on to

and guess who will win the election this year.

This poll closes on 11:50 pm 14 September 2008. No guessing will take place after that.

I am testing the ability of people to guess elections at a point where the amount of information known about each candidate is roughly the same. The site is completely anonymous.

Please do not try and stuff the ballot box by voting more than once. I will not release any results until after the election is over. (I will also remove duplicate records.)

Please tell everybody you know, of every political background, liberal or conservative. Email them the link above. If you can, link to this page on other blogs so that we get as large a sample as possible.

Please, pretty please answer the 6 questions honestly.

Once the election is over, the analysis will appear at this web site.

Thank you very much!

Categories: Statistics

21 replies »

  1. Wow. Just a minute or two after I posted this, we got over 50 responses!

    Please email this page or the link to friends on both sides of the aisle.

    Thanks everybody!

  2. Briggs

    I am interested in the result – but will resist the temptation to vote as I am from the “Land Down Under” and ineligible to vote for either.

    Wish Palin was on our scene – it really is boring down here with low calibre pollies on all sides of the spectrum trying to run our lives.

  3. Andrew,

    You make a good point, but that option was purposely left out because of the realities of the American political system.

    There is an overwhelming probability that the winner will be one of McCain or Obama (there is, of course, a chance, however miniscule , a third-party candidate could win, but it is so small as to be negligible).

    Since this is the case, since it will be one man or the other, you realistically have only one choice, and which of these two would you rather see?

    Thanks for the comment.

  4. Briggs,

    I did not read the entire post before casting my vote; I probably should have but it’s too late. I am not American -although lived here for 6+ years now- so you decide what to do with my vote.

    Anyway, I understand it would take longer for voters but what about giving the opportunity to them to express their ideas on the “issues”. I always wonder how Americans cast their vote: based on candidates stands on issues, “chemistry”/gut feelings, the party represented, etc.? I, for once, consider myself neither a conservative nor a liberal (still cannot decide) but I came from a Country (Italy) where people are more ideologized and tend to give preference to the party no matter what the candidate says/does.

    Why don’t you include also a list of questions such as:

    When human life begins? (Conception/9wk/3mo/6mo/birth)
    Do you support Abortion? (Y/N)
    Do you support Death Penalty? (Y/N)
    Do you support Embryonic Stem Cells Research? (Y/N)
    Do you support the No Child Left Behind solution for education? (Y/N)
    Do you support ANWR Drilling? (Y/N)
    Do you support the Kyoto protocol? (Y/N)
    Do you believe in Global Warming? (Y/N)
    Do you believe in Human Caused Global Warming? (Y/N)
    Do you support the Guns Assault Weapons Ban? (Y/N)
    Do you support the Homeland Security Patriot Act? (Y/N)
    Do you support the Citizenship path for Illegal Immigrants? (Y/N)
    Do you support Iran Sanctions and Military Action? (Y/N)
    Do you support the Iraq War? (Y/N)
    Do you support Minimum Wage Increase? (Y/N)
    Do you support Same-Sex Marriage? (Y/N)
    Do you support Same-Sex Civil Union? (Y/N)
    Do you support Federally administered Universal Healthcare? (Y/N)
    Do you support privately administered Healthcare? (Y/N)
    Do you support increase in federal regulations? (Y/N)

    And finally:
    Does the candidate XYZ believe in what he says? (Y/N)
    Will the candidate XYZ do what he says? (Y/N)
    Is the candidate XYZ fit for the job? (Y/N)

    This is just a list that comes to mind but surely not the best.

  5. Joy, Thanks!

    Marco, those are all excellent questions, but the answer to them is less in doubt. That is, we generally know how people who say “Y” on your first question will vote.

    The aim of this study is to see how good people are at guessing two months away from the election. This is different than an ordinary poll which asks “Who will you vote for?”

  6. I’m waiting to see how Palin survives the initial flurry of stories before I vote. Thanks for setting the deadline to the 14th!

    I also need to get out of Dupage county and talk to people in other counties. Otherwise, I have no way of guessing who might win. 🙂

  7. I see your point Briggs and I got the spirit of the survey, however mine was a suggestion to complete the picture of US voters.

    About the first question I am still unsure on how people saying yes to the second AND the third question will vote. And I seem to understand that your goal is not just to find out how many voters will guess correctly the outcome but also to “correlate” it to their personal choice.

    I just wanted to go one step further and not only find out if they can guess correctly and for whom they vote but also “why”.

    But I also understand that then it would be much more work and not in the spirit of the quick original survey.

    In any case I am looking forward to knowing the outcome of this nice experiment – although you could be accused of partisanship because probably your readers/voters do not represent a “statistical sample” of the US population :).

    Marco

  8. Canadian election writ to be dropped tomorrow,(Sunday), how about a pole for that? It would be a little more interesting since Canada is so much more politically diverse in represented parties, we have a smidge right of centre, a smidge left of centre, Left of reality, regionalists, and enviromentalists with conservative social polices represented. I think if you added that voters add their geographic location, you could see how Canada, is just about 3 different countries, and why it stands a good chance of having minority Governments for quite some time.
    In this one I went McCain, simply for the fact I think Himself and Palin simply appeal to a wider audience.

  9. Marco,

    Understood. But I also didn’t want to frighten people away. Start asking them questions on controversial topics and they—rightly—become suspicious.

    You’re also right that my sample is not inclusive (the word “random”, while often used with respect to surveys, is meaningless). I still would like to get more liberal-minded voters.

    Know any?

  10. I can’t complete your survey because the only reasonable answer to “Who will win the election?” is “I don’t know.” I’m not a fortune teller!

  11. I just popped over from Pharyngula myself, and as a devoutly agnostic left-leaning physicist, I hope I don’t skew your results too much. 😉

    Seriously, it’s good that the survey is so simplistic. Adding more questions such as Marco suggests would almost certainly destroy its value because the respondents would self select for 2 reasons:

    1. Laziness – Too many questions! I’m not wasting the time to go through all these!

    2. Question Bias – gives away the ideological leaning of the questioner, which will turn off people from opposing philosophical camps. There are several such examples that jump out at me in the list Marco posted:

    “Do you support Abortion?” – I think the intent of this is to distinguish between what we loosely term “pro-choice” and “pro-life”, but the question is terrible. I am strongly pro-choice, and would never say I “support Abortion.” I would like there to be as few abortions as possible. That’s why I also strongly support measures to reduce unwanted pregnancy in the first place (contraception, sex ed.). But I don’t believe we can hold a gun to women’s heads and tell them how to use their own bodies. Would Marco agree with a rephrasing of this question as “Do you believe female bodies belong to the state?” or “Do you favor putting women in jail for making reproductive choices?” These are functionally the same question, but I don’t think the political bent of most folks here would like these phrasings.

    “Do you support increase in federal regulations?” – about what? Are all regulations equal? Is demanding that black people be allowed to exercise their right to vote an “increase in federal regulations”? Should it be illegal to dump toxic waste in the local river? Clearly this question is phrased according to a ridiculously dogmatic view of the role of government bordering on anarchy.

    If I had come here and seen these questions in a survey, I would immediately have dismissed the whole exercise in right-wing narrative reinforcement and blown it off.

  12. Rheinhard,

    You’re a sweetheart, thanks for taking part.

    Too, those kinds of political questions would not help with the actual goal of this study.

    Whoever wins will not change how this study is analyzed.

    There’s a lot of anger out there. Once somebody finds out that you don’t intend to vote the same way they do, tempers flare, particularly on the internet. There’s a few who came in from Pharyngula who were incensed that the guessing study even existed.

  13. SamSun,

    I’m not sure whether to thank you or not! I sure did get a lot of people from that blog, but despite my many pleas to them, they are intent on being facetious. Many have cast multiple votes, one right after the other, and many are claiming they want Obama to win and are Conservative.

    One commenter seemed to think this trick would throw me for a loop and really get his point across about the evilness of the current president. How trying to screw up my study is going to influence Bush is anybody’s guess. A lot of guys over there are letting their passion overwhelm their reasoning abilities.

    Not all who came from there had ill intent, for example, Rheinhard.

    But it looks like I might have to throw out all data from people who came from this site, because it’s getting difficult to tell the genuine from the juvenile.

    Anyway, thanks for attempting to help. I appreciate it.

  14. Well, I don’t think wanting Obama to win and thinking of oneself as “conservative” are necessarily mutually exclusive. There is a lot of policy today that goes by the name “conservative” that would have made Barry Goldwater nauseous.

  15. Rheinhard,

    True enough. But if those conservative Obama supporters all come from the same place, and all are Females of the same age, and all post within a minute of one another, you begin to suspect something fishy. I have hundreds of these today.

    Plus, some comedian thought he’d try and vote for Batman or whatever many, many times (my fault, as pointed out elsewhere, for allowing the html form method to be post instead of get).

    But I will still use this data—today’s I mean—as an interesting case study of mania.

  16. Briggs, deliberate sabotage from liberals?! Hahah…, sorry, but I find this hilarious. Throwing out data? They could be “independent” voters with conservative views. Darn, the convenient sampling method is not so convenient.

  17. JH,

    Well, like I told Rheinhard, I’ll still use the data, but analyze it with a different goal in mind.

    Sure, I’d love to have a real study with a good heterogeneous sample. But I am one guy, paying for everything out of his own pocket.

Leave a Reply

Your email address will not be published. Required fields are marked *