Slashdot Mirror
Stanford Predicts The Presidential Election
Can Sar writes "Today is the official launch of Stanford Predicts, a non partisan group trying to predict the 2004 Presidential Election. This project is led by and based on research by Professor Samuel S. Chiu of the Department of Management Science and Engineering at Stanford University. Stanford Predicts is solely interested in predicting the likelihood of either candidate winning, for purely scientific purposes. While the formulas themselves were developed in previous years by Professor Chiu all data analysis is being done by undergraduate students. Stanford Predicts will be continuously updated with new predictions until election day. Please check out Stanford Predicts for more information."
American University Professor Allan J. Lichtman, author of Keys to the White House uses a 13 key system to predict the Presidential winner (popular vote), and right now, the keys system favors Bush (9 to 4). Gore won his analysis in 2000 and the popular vote, but not the Presidency. It's possible something similar could happen again.
I'm a Bush supporter and I cannot believe that'd he win by such a margin...
But, as expected, Kerry wins the West and East coast states while Bush wins in "flyover" states. I expect a GWB victory this November -- but I think it'll be more along the lines of 57% to 40% in terms of the popular vote with the third parties picking up the rest of the slack.
I'm not popular enough to be different.
Homer Simpson, The Simpsons
Wasn't it a public knowledge that the election will take place? It was all over the news.
Sincerely,
Pan Tarhei Hosé, PhD.
"Homo sum et cogito ergo odi profanum vulgus et libido."
I particularly hate both candidates but I'm rooting for Bush so I can watch as all the hippies and Kerry explode in anger and riot in the streets or something
This guy has the same sort of daily predictions. Funny his prediction for today is the opposite that Stanford predicts. I sent him an excel file demonstrating the use of error estimates and probabilities to get a better prediction, but haven't heard back. Though even with that the prediction would still be in Kerry's favor, so I'm not sure what all Stanford is and isn't taking into account. He apparently gets a lot of crap in his email from opponents.
Bush is more likely to win the Presidency, even if he doesn't win the popular vote.
I've been following along the election with www.electoral-vote.com, and while it's good to have predictors, I wonder how much impact these have on the outcome, do sites like this have a detremental effect on the election by creating a self fulfilling prophecy? Have there been studies on this effect, say a voter sees that bush has a 72%chance of winning, and decides that the country can't be wrong and goes along with it.
Anyway, I like electoral-vote's way of going about it better, shouwing actual state polls and the number of electorial votes each candidate has rather than a straight up prediction of the outcome in a percentage. Lets you see how close the race is as far as votes and what your effect based on your state can be, it's a lot more empowering when you see your state is very close and has a lot of electorial votes, and how close the candidates are.
WikiAfterDark.com It's a sex wiki, go now!
For everyone who says "I want to vote for a winner" there may be someone else who says "Well, I'll need to make sure I go out and vote for my guy so that doesn't happen!"
I couldn't find info on which polls they used. Of course, some pollsters do a better job than others, and some even engage in push polling?*
So, it seems to me that feeding a different subset of polls will garner different results, and that the equilibruim is very stable -- change the Ohio or Florida poll by two percentage points toward Kerry and I'd bet the odds go from 3:1 to 1:1 pretty damned quickly. Likewise, fudge the CO, NH, and MN results toward Bush 2 points, and it might go from 3:1 to 5:1.
Surely they could do a better job about releasing their data and their polling selection methodology...
* baiting an answer. For example: Would you vote for George Bush even though he lied about WMDs and his wife once killed a man? Clearly not a good idea if one seeks accurate polling, but it's done all the time nevertheless. Just ask wiki about Sen. McCain's black baby born out of wedlock.
Support a few technologists in Washington.
I'm no computer scientist, but -- wouldn't a Monte Carlo technique do this reasonably well?
What I'm listening to now on Pandora...
This site does the same thing only they're much more trasparent about their methods. They have also started to use poll data from multiple sources to minimize bias.
I predict that I may not be the only one that is completely fed up with people wasting too much time and energy on predictions, especially innocuous and chaotic issues such as the election a week before the fact. If you want to predict Earthquakes and weather patterns, cool, but no matter what you predict, things are still going to be chaos because it doesn't take a PhD to figure out one side is going to be unhappy about losing. Thank you professor Obvious.
Why don't you academics stop jerking off over week-early predictions and do something productive like research a cure for cancer, or at least see if you can predict the number of people who could possibly care about this big waste of resources and ask yourself if your time and talent are not better served elsewhere?
In other words,
We, and the rest of the world, are screwd.
This is my problem with these sorts of things. While the polls are always statistically sound i have a 800-lb gorilla-sized sneaking suspicion that the polls being conducted do not accurately represent the electorate, in which case the statistical rigor gives way to a sort of bias in these results.
I've thought for a long time (since last spring) that Bush will lose by a not unsizable margin and people may actually be surprised on election day by the way the polls had failed to capture the public's true intent.
This is all purely anecdotal of course but i just think that since all of these polls are via land-lines (at who knows what time of day), they no longer capture a validly random sample. After all a shrinking percentage of people i know (all of whom vote) even have a land-line, and far fewer actually talk to any pollsters or their ilk - the urge to hang-up on these sorts of callers is just too overwhelming...
Though it may very well be me who is surprised on election day this is what has been brewing in my head lately...
We'll see, although i would bet that there'll be partying in the streets around the world on Nov. 2nd/3rd should Bush lose.
-tid242
With a few exceptions, secrecy is deeply incompatible with democracy and with science. --Carl Sagan
Either Bush or Kerry will get into office. They will spend our federal tax money however they want, generally kissing the ass of big business. We will continue putting hundreds of thousands of people in jail for drug use. We will continue pushing a litigious society with no hope for tort reform. Illegal industry groups (MPAA, RIAA) will be given even more power. And no matter what, Bush's friends will become much richer, and Kerry's friends will become much richer.
Feel free to supply your own!
These type of things are bad for two reasons:
1) They meddle with the elections. If a Kerry supporter sees this, it may discourage him to actually go and vote, because "awww... this scientific poll says that Kerry already lost, I don't even need to vote"
2) How can a poll really claim to be scientific when it gives 100% chance to any canditate for a state? Sure, its pretty safe to assume that Kerry will win CA, or Bush TX, but you really never know what could happen in a situation like this. Saying 100% percent means that there is no way that this could happen any other way under any conditions.
I've been following elections for over 20 years in over six countries and various states. In that time I've learned two things (a) generally polls are eerily accurate (ignore them at your own peril) and (b) every so often there are certain elections which one can tell, almost from the get go, that previous historical norms won't apply and hence polls will mispredict the outcome.
The 2004 presidential election is one of those elections in which participation rate of voters will be way out of the norm, on the coat-tails of the 2000 stalemate and the strong anti-Bush feeling from the democrats.
Using historical data, Bush is slightly ahead as reported by the Stanford poll or electoral-vote.com. If we correct the data assuming a slightly higher participation for the democrats, the polls give an edge to Kerry of 284 electoral votes vs 254 for Bush.
We all get along together like tornadoes and trailer parks.
Besides the UMN site already mentioned above, I highly recommend everyone regularly visit RealClear Politics (whose rolling averages have become a de facto barometer for journalists), The Horserace Blog (Jay Cost crunches the numbers in a way that puts the mainstream press' attempts to shame, and explains every step of his analyses), and Daly Thoughts (the best single state-by-state analysis of poll trends).
They use the results of a number of polls. Since these polls are more or less independent of each other, it's mathematically acceptable to aggregate them (provided you do the simple stuff like weighting the polls according to sample size, etc.).
You say that the polls themselves are all biased in the same direction, reflecting the viewpoint of likely voters who answer their landline. While I can't invalidate that completely, the fact that multiple polls find similar results tends to weaken the idea. The question is open whether people who don't answer their landline lean toward one side enough to change the results. Also, polls of kids, who usually tend to track their parents' viewpoints, agree with the telephone polls.
It's possible that your friends think the same way you do, so to you it feels like everybody hates the President, when in fact most people like him.
sigs, as if you care.
This guy's clearly an example of an earlier story on slashdot.
(I kid, I kid.)
Ben Hocking
Need a professional organizer?
that partisan supporters on both sides will wage a vicious court battle, which will be cut off halfway through by the supremes, who will say, "Didn't you hear us four years ago? Any contestation of Bush's victory could do harm to Bush's presidency. Therefore, Bush is the victor." See p152, recursion adj.
"A witty saying proves nothing." ~Voltaire
"d'Oh!" ~Homer
The polling methodology is flawed. All of these predictions are based on data collected from these polls. You might as well just watch the Redskins this Sunday or check out Tradesports.com.
Where's the polling data to back these numbers up? Just clicking on the link to Wisconsin shows Bush with a 92% likelihood of winning, even though the headline states, "Kerry and Bush Remain Tied Among Likely Voters in Wisconsin". I wanna see sources, not magic numbers.
With the election being likely another 50/50 split, the real deciding factor is going to be how much voter fraud is going to occur, how much electoral fraud (Diebold is looking forward to delivering Ohio's votes to the President!), the margin of error with the voting machines, margin of error with the humans checking the voting machines, and the likelihood of another Florida.
Actually, if we can determine the probability of another Florida, we already know the outcome of the election (5 Bush, 4 Gor...er, Kerry) and we can all sleep in on Nov 2!
Here come da fudge!
Stanford predicts Bush will win the swing states, including New Mexico, and Kerry will win Colorado. Anyone want to bet?
--
make install -not war
HULK: 9,151 BUSH: 8,910 KERRY: 8,391
This is despite at least one "Kerry-Bot" which tried to stuff the ballot.
Hulk SMASH Celiac Disease
Both support the war in iraq.
Both have spending plans that are in the red and both say they'll cut their deficit spending in half within four years.
Both support the patriot act.
Both support curtailing the 2nd amendment.
Both have increased the size and scope of the federal government.
The differences are Kerry wants to tax and spend while Bush wants to borrow and spend.
Kerry - Pro choice, Bush - Pro Life
So in conclusion I'd say yes they are both asshats.
Service guarantees Citizenship! Questions Guarantee GITMO.... Amerika Uber Alles!
After all, we don't hear much these days about Bush's promise in 2002 to "[balance] the budget by 2005 without endangering the war against terrorism and homeland security efforts and without raising taxes."
These days, he likes to blame his inability to keep this promise on the 2000 stock market correction and the 2001 terrorist attacks, neither of which he could reasonably have foreseen in 2002. I'd like to see Bush have four more years of opportunity to be a man and take responsibility for his stewardship over the federal budget.
I feel that it would be impossible for Bush to win the popular vote yet lose the election, due to the lock he has on a number of small states with much greater elector/population ratios. Can anybody make up a scenario where this is remotely possible? I don't think this is going to be an inverse of the previous election, but a repeat.
I agree with you. Many Americans would truly be confused. For Stanford to try and predict the election based on a mathematical formula, is like when the Super Bowl outcome is prdicted by allowing the computer to play the two teams in a game of Madden 2005. I really do not think that it will work, too many variables.
Comment removed based on user account deletion
Picture of George Bush waving a peace sign and the slogan "I'll bet you'll vote this time, Hippie!!"
...which won't be out until Wednesday or Thursday for people w/o subscriptions. Relevent to this topic are the two best stories:
"Countdown to The Recount 2004" ("How to make your vote recount", "When will your next president be appointed", etc)
"Republicans Urge Minorities To Get Out And Vote On November 3rd" (ouch!)
You cannot apply a technological solution to a sociological problem. (Edwards' Law)
Jeb Bush.
92% chance that he'll manage to give Florida to his brother.
All you have to do is wait a couple of weeks, and you'll know who has won. It's the voters who decide the outcome in (at least in most cases - we might hope we don't see a repeat of the last election fiasco), not the pundits spinning their wheels.
You know, Rush Limbaugh made a good point about all of this polling data. And that is, none of it's from calling cell phones. And with more and more people using their cell phones as their main line, who really knows who THEY might vote for.
I know it shouldn't matter. But cellphone users are general younger people and those always on the move.
Life is not for the lazy.
I have often thought of the polls and I consider them to be self fullfilling. Assuming there is one that everyone takes as a "standard", then I would argue that the votes would tend to lay out how the polls say they should. This is due to certain people going whatever way they were going to already, and a number of on-the-fence people wanting to pick a winner.
I am curious to see if anyone else agrees or disagrees with this notion.
This is based off the poll data - all polls this year have incorrectable sampling errors that inflate bush's numbers
If you cannot keep politics out of your moderation remove yourself from the Mod Lottery.. NOW!
Wasn't it a public knowledge that Pentagon was struck by a American Airlines 757 plane, but this is not what really happenend http://www.sodaro.com/Plone/ps/
Reading the methodology on the site, it seems like they figured out a probability for each state, then ran a simulation where each state independently has its given probability of going each way.
This seems utterly ridiculous! If swing state #1 goes one direction, then it is much more likely that swing state #2 is going in the same direction. Because of this, their model will have results centered artificially close to the expected value (swings in their methodology cancel each other much more than in the real election), and the probabilities of each candidate winning are closer to 50% than the site indicates.