Slashdot Mirror
The Votemaster Is...Andrew Tanenbaum
A reader writes: " www.electoral-vote.com, a site of daily updated maps of the US electoral college based on a number of polls is probably a site that the policially inclined check daily. Well, it has been revealed that the person behind the site, AKA the votemaster, is none other than Andrew Tanenbaum, noted author of numerous CS books." He's also known for a little discussion with someone named Linus Torvalds.
Even before the 'slashdot effect', the site has been unavaiable. All morning.
The effect is real. It varies by election and by area. In some places (the Northeast) the effect can be as large as 6% of people who will apparently lie to you on the phone and say they are undecided or voting for the Democrat. In the Midwest it's less pronounced and the effect barely exists in the South, though i've never been very close to a poll done there.
It has something to do with either the Republicans not wanting to answer the phone or alternatively not wanting to be judged by the pollster, i'm sure.
Zogby talks a little about this in one of the FAQs on his website.
HBI's Law: Frequency of calling others Nazis is directly correlated with the likelihood of the accuser being Communist.
He's a US citizen living abroad:
"My name is Andrew Tanenbaum. I am one of the 7 million U.S. citizens living abroad. I am a professor of computer science at the Vrije Universiteit in Amsterdam, The Netherlands. Most of you have never heard of me but in an itsy-bitsy, teeny-weeny corner of the universe I have done enough stuff that Google has somehow managed to dig up 10,000 pages referring to me."
There are several similar sites using slightly different formulas. Another good one is here.
"with their freedom lost all virtue lose" - Milton
He's a US Citizen, just not living in the United States.
Nice long post. I can sum it up easy. While he is clearly a Kerry supporter, the polls are non-biased and based on existing polls.
If you followed the site for some time you would see that. There is even a movie on the site to show you how much the polls have been swinging back and forwards.
If anything his site shows how pointless polls are, or that the undeceided voter is completly clueless and changes their mind every 5 minutes.
The only poll that really matters is tomorrows.
On another note "600,000 Iraqis". Can you quote a source for that? The only figure I can find is for displaced and not killed under sanctions. Also you should note that Saddam was grossly inflating the deaths (especially children deaths) in order to try and stop sanctions.
Heartland? Like California, Texas, Florida and New York???
Quibble aside, the gist of your comment is correct. Americans have an instinctive tendency to go our own way, right or wrong. And most of the "up-for-grabs" electoral votes are in the midwest, like Michigan, Wisconsin, and Ohio.
Stop by my site where I write about ERP systems & more
Except he's a US citizen living abroad, which is not the same as not a US citizen.
A friend of mine recently moved to Canada for work and told me that lots of US expats she knows there are voting for the first time in years (often for the first time since they left). If you're living abroad you vote in the last state where you were a resident and you only get to vote for president (maybe senate, too, but I think just prez). Many of those people last lived, and are very likely to vote for Kerry (in Canada, the far right is mostly to the left of the US Dems).
It's going to be an interesting election night...
(sarcasm appreciated except for the nit)
I think he meant the majority of Bush's support comes from the Midwest, where states like Arkansas and Missouri and Iowa, while contested heavily, will need to be carried by either candidate if they hope to win decisively. Bush's support in the deep south and western US (save the west coast) is not usually contested as these represent the Republican base.
Slykens, he ALWAYS uses the poll with the most recent median date. He's currently using a Zogby poll with a median date of 10/31 in Florida, making it more recent than the Quinnipiac poll. He has explained this many times over the past few weeks: He only averages polls that have the same median date and polling length.
There's a rival site that attempts to use "meta analysis" techniques to crack that stubborn +/-3% margin of error. I'm not all that well versed in statistics, so I can't comment with any degree of reliability, but it might be worth a look.
Predicted median with undecideds: Kerry 280 EV, Bush 258 EV
Median outcome, decided voters only: Kerry 252 EV, Bush 286 EV
The author of the site, Sam Wang, has published some of his methodology in the form of a matlab/octave script.
Which makes it really strange, since A&M is the most conservative public university in the country. It's also where W's dad has his presidential library.
Don Negro
Perl 6 will give you the big knob. -- Larry Wall
The only poll that really matters is tomorrows.
Absolutely.
I live in a non-swing state, so I've been volunteering for some get-out-the-vote-in-swing-states phone banking over the past couple days, and I no longer believe the polls. People in swing states are getting so many phone calls that many of them no longer answer the phone, they put messages on their machine saying if it's a political call please go away, they hang up right away, etc. They are extremely popular right now, and most of them seem to wish it would all go away.
On the few occasions that you do get a a live person, pretty frequently they say "this is my fifth call today, and someone just left the front door, would you please take us off your list". I apologize, and thank them, but because many of the groups aren't allowed to coordinate (or don't when they could), getting off one group's list doesn't help much.
The pollsters are calling all the same people, and probably having just as hard a time. They have to make a lot of corrections for systematic error, and I would suspect that the popularity of the swing state voters makes their correction factors less useful than in a more typical year.
Every once in a while you get someone who didn't know where to go to vote, or who needs help getting to the poll (which we help with). They make it worthwhile.
Here's some interesting information from a previous Votemaster comment:
The attackers have tried repeatedly to break in, but the server is a rock-solid Linux system which has stood up to everything they threw at it and hasn't crashed since I got it in May.
The full Google cache of the page is here
SourceHosting.net, LLC
Ready. Set. Code.
http://www.sourcehosting.net/
This Brazos county is one of the rare reversal of which the college community is predominately Republican and the nearby residential areas are Democratic.
Enjoy!
I don't think polls are "pointless", but many people are very clueless about statistics (including, apparently, almost everyone in the media).
The talking heads on the news regularly talk about how a poll has "swung" one way or the other. For instance, this morning a poll came out that showed Bush up by 2% in the popular vote, 48% to 46%. The day before they were tied, I believe at 46% to 46%. Everyone involved talked about this as a real effect even though the margin of error (MoE) in the poll was 3%! Statistical variation completely explains those two results, it is quite possible that voter sentiment didn't change a bit!
Even beyond that, again by the nature of statistics polls are not as reliable as they are portrayed. The above mentioned MoE is only good for a 95% confidence level. In other words, there is a 5% chance that the reported numbers lay outside the MoE! So, it is best to view poll numbers with a very large grain of salt...
Galileo: "The Earth revolves around the Sun!"
Score: -1 100% Flamebait
Check the trade balance. It's the US that needs to trade with the rest of the world to survive - not the other way around.
it's in my head
Mirrors:
www.electoral-vote2.com
What's funny
www.electoral-vote3.com
Is that
www.electoral-vote4.com
This comment can't be posted
www.electoral-vote5.com
because of all the repetition
www.electoral-vote6.com
In the comment
www.electoral-vote7.com
Due to listing all the mirrors
www.electoral-vote8.com
Forgive me if I think this
www.electoral-vote9.com
Is really stupid
--Remove chicken to e-mail
And the widespread corruption in the UN Oil-For-Food program that directly benefitted French, German, and Russian companies had absolutely nothing to do with their opposition to the war.
Actually, Kerry wouldn't take office until January, so he'd have at least a couple of months to come through with all that stuff.
I don't care if it's 90,000 hectares. That lake was not my doing.
You do not get to vote for president. None of us do. Your state does. You vote for your state's electors, since that is the election system your state has set up. It is the state's choice to cast all of its votes for the state's popular vote winner (although one state currently has a ballot measure which would split up the electoral votes in some situations). We have a federalist system. If you do not and cannot understand the governmental system we use in this country and why we use it, then it's probably best that your vote "doesn't count."
Texas? WTF? In 2000, Gore won in only one county there. Brazos county, home of Texas A&M University, and a damn disproportionate conglomeration of college students.
Not true. Gore won several TX counties near the Mexican border.
Also, a better site than Tanenbaum's for predicting the winner is here. Sam Wang of Princeton University uses a statistical method for averaging all recent polls rather than rely on just the latest for his predictions.
Personally I'm predicting a blowout for Kerry. This is based for starters on Wang's data. 2nd, last night on MSNBC's Hardball, Chris Matthews said that the exit polls from early voting in Iowa had Kerry 11 points up. 30% of Iowa has already voted. There has also been a huge early turnout in Democratic areas in FL, NV, GA, and NC. 3rd, a recent Zogby poll of 18-29 year-olds with cell phones gave Kerry 55%, Bush 40%. Every other poll I've seen is based exclusively on land lines, so if the 18-29 year-olds vote this year (and granted, they usually do not), the polls could be way off. Finally, Karl Rove's strategy is based on getting some 4 million more Evangelical Christians to the polls than went in 2000. Problem is that the size of this group may be a myth. A devout Christian friend of mine invited me to a party Friday night with some of his church buddies. Not a group I normally hang out with, but I like being exposed to new ideas. Turns out this small sample favored Kerry over Bush by 50-40. A few were still very undecided (yes, even today there are still undecided voters in Ohio!). All of this leads me to believe that Kerry will clobber Bush.
FreeSpeech.org