actually silvers model is prob based on simulations (the 10 days thing)
― flopson, Wednesday, 31 October 2012 20:13 (eleven years ago) link
-> on election day, God rolls a 10-sided die (7 sides obama, 3 sides romney)― flopson
If we're gonna go the AD&D route, we have to consider modifiers. Does God have to roll a Sandy response check and add a +1 incumbent mod?
― Frobisher the (Viceroy), Wednesday, 31 October 2012 20:13 (eleven years ago) link
yeah silver runs like 100,000 simulations for his numbers
― iatee, Wednesday, 31 October 2012 20:14 (eleven years ago) link
http://www.nydailynews.com/polopoly_fs/1.1178383!/img/httpImage/image.jpg_gen/derivatives/landscape_370/image.jpg
― Josh in Chicago, Wednesday, 31 October 2012 20:14 (eleven years ago) link
ftr i understand statistics but a romney win would shock me to the core
Probably because you've seen Mitt Romney speaking, and thinking, and laughing, and doing other Mitt Romney things. The numbers can never adequately contain his Mitt Romneyness.
― clemenza, Wednesday, 31 October 2012 20:15 (eleven years ago) link
xp LOL at Obama being drawn w/ classic evil cartoon eyes.
― Frobisher the (Viceroy), Wednesday, 31 October 2012 20:19 (eleven years ago) link
tbh i think sandy is going to give obes even more of a poll boost in key states before tuesday.
― sug ones (omar little), Wednesday, 31 October 2012 20:22 (eleven years ago) link
OBES
― the ones that I'm near most: fellow outcasts and ilxors (Alfred, Lord Sotosyn), Wednesday, 31 October 2012 20:25 (eleven years ago) link
http://www.zillow.com/blog/files/2011/12/Lagasse_Photo.jpg
― balls, Wednesday, 31 October 2012 20:26 (eleven years ago) link
http://cdn.dealbreaker.com/uploads/2011/05/drudge-siren.gifBREAKINGhttp://cdn.dealbreaker.com/uploads/2011/05/drudge-siren.gif
http://cdn.dealbreaker.com/uploads/2011/05/drudge-siren.gifNATE SILVER CHERRY-PICKING DATA, SAYS, "YES, I AM DELIBERATELY CHERRY-PICKING"http://cdn.dealbreaker.com/uploads/2011/05/drudge-siren.gif
http://cdn.dealbreaker.com/uploads/2011/05/drudge-siren.gifhttp://cdn.dealbreaker.com/uploads/2011/05/drudge-siren.gifhttp://cdn.dealbreaker.com/uploads/2011/05/drudge-siren.gif
― but the boo boyz are getting to (Z S), Wednesday, 31 October 2012 20:26 (eleven years ago) link
8th paragraph down
http://fivethirtyeight.blogs.nytimes.com/2012/10/31/oct-30-what-state-polls-suggest-about-the-national-popular-vote/?hp
of course i'm totally being a facetious smartass but still he might not want to use those words when the barbarians are at the gate
― but the boo boyz are getting to (Z S), Wednesday, 31 October 2012 20:27 (eleven years ago) link
Do I cherry-pick myself?Very well; I cherry-pick myself
― the ones that I'm near most: fellow outcasts and ilxors (Alfred, Lord Sotosyn), Wednesday, 31 October 2012 20:28 (eleven years ago) link
is it really that hard to understand what "70% chance of winning" means? do we as adults really need (flawed) baseball and d&d comparisons to make this more intuitive?
― congratulations (n/a), Wednesday, 31 October 2012 20:28 (eleven years ago) link
it's actually not intuitive at all! people don't naturally think in probabilities
― iatee, Wednesday, 31 October 2012 20:30 (eleven years ago) link
BREAKING
NATE SILVER SAYS IT CERTAINLY COULD TURN OUT TO BE THE CASE THAT THE STATE POLLS ARE SYSTEMATICALLY BIASED AGAINST ROMNEY
http://cdn.dealbreaker.com/uploads/2011/05/drudge-siren.gifWOO WOOO WOO WOO WOO WOO - WOOOooooooOOOooooooOOOOoooo - WOOWOOWOOWOOOWOOWOOhttp://cdn.dealbreaker.com/uploads/2011/05/drudge-siren.gif
― but the boo boyz are getting to (Z S), Wednesday, 31 October 2012 20:31 (eleven years ago) link
But perhaps national polls tell the right story of the race instead — meaning that the state polls systematically overrate Mr. Obama’s standing?It’s certainly possible. (It keeps me up late at night.)
It’s certainly possible. (It keeps me up late at night.)
NATE SILVER'S BAMBOOZLING OF DATA KEEPS HIM UP LATE AT NIGHT FILM AT 11
― johnathan lee riche$ (mayor jingleberries), Wednesday, 31 October 2012 20:31 (eleven years ago) link
I hope Shakey will join me in dancing on Drudge's grave when the inevitable aneurism finally claims him
― Gandalf’s Gobble Melt (DJP), Wednesday, 31 October 2012 20:32 (eleven years ago) link
NATE SILVER'S VOICE IS HIGH AND HIS ARMS ARE SKINNY
WOOWOOWOOOWOOOWOO-WoooooOOOOOOOooooooOOOOOoWOOWOOWOOWOOOWOO
― but the boo boyz are getting to (Z S), Wednesday, 31 October 2012 20:32 (eleven years ago) link
i'm assuming you all make the same sound effect when you see the drudge sirens
― but the boo boyz are getting to (Z S), Wednesday, 31 October 2012 20:33 (eleven years ago) link
can we really trust a gay man's statistics in a straight man's world?
― iatee, Wednesday, 31 October 2012 20:33 (eleven years ago) link
at this point the only drama nate can generate is drama over the validity of his figures. it's in the interest of his page views to say things like "i cherry pick data come at me mouth breathers"
― Mordy, Wednesday, 31 October 2012 20:36 (eleven years ago) link
do we as adults really need (flawed) ... d&d comparisons
fyi my dice comparison was not flawed maybe *you* need to brush up on your probabilities
― flopson, Wednesday, 31 October 2012 20:37 (eleven years ago) link
he's talking about cherry-picking the examples for that one blog post, not the data for his model
― ciderpress, Wednesday, 31 October 2012 20:37 (eleven years ago) link
"(flawed)" was only intended to modify the baseball part of the sentence, it was poorly phrased and i apologize
― congratulations (n/a), Wednesday, 31 October 2012 20:43 (eleven years ago) link
This should be renamed to the Nate Silver thread.
― Jeff, Wednesday, 31 October 2012 20:51 (eleven years ago) link
he casts a shadow like pitchfork on ilm threads
― push iatee (some dude), Wednesday, 31 October 2012 20:52 (eleven years ago) link
http://fivethirtyeight.blogs.nytimes.com/2010/12/15/a-bayesian-take-on-julian-assange/ seems legit
― zvookster, Wednesday, 31 October 2012 20:53 (eleven years ago) link
haha someone linked to that the other day. wtf nate.
― max, Wednesday, 31 October 2012 20:54 (eleven years ago) link
OBAMANIA: THE GATHERING
― Raymond Cummings, Wednesday, 31 October 2012 21:01 (eleven years ago) link
i.e. man this thread got weird
only on the internet: "to make this more understandable to the common man, imagine rolling a d10..."
― Doctor Casino, Wednesday, 31 October 2012 21:08 (eleven years ago) link
On intuition vs probability, my favorite illustration is the Monty Hall Problem. You can make people do the math themselves, and they still won't believe it.
― something of an astrological coup (tipsy mothra), Wednesday, 31 October 2012 21:08 (eleven years ago) link
Obama hasn't tapped enough mana to be able to bring out his Wall of Dem GOTV, and his Biden artifact hasn't done much to stop Romney's endless supply of Rethuglican Imps. He should have used a different deck, IMO...
― Frobisher the (Viceroy), Wednesday, 31 October 2012 21:17 (eleven years ago) link
No surprise that Obama is using a rainbow deck, nor Romney sticking to a straight red one. His Rovian Dragon was surprisingly ineffective though...
― Frobisher the (Viceroy), Wednesday, 31 October 2012 21:19 (eleven years ago) link
Now, if we want to examine this from a GURPS lens....
ok I'll stop now.
GURPS ironically well suited to simulating political fights.
― Mordy, Wednesday, 31 October 2012 21:27 (eleven years ago) link
I'm not especially math-geeky, but Silver's probabilities make great sense to me in that he's not really stating the probability of a win or a loss for each candidate so much as he is evaluating the probability that the state-by-state polls are exactly correct, or if the actual voting outcome will fall to one side or the other of the polls' margins of error, then he is synthesizing all these probabilities in terms of electoral college outcomes, and finally correlating these electoral college outcomes with the win for either candidate.
So long as the polling margins of error allow a way for Romney to amass 270+ electoral college votes, there is a non-zero chance for Romney to win. At this time, that non-zero chance is roughly 22.6%, according to Silver's models. iow, that would be the quantum state of the election, if the election were Schrodinger's Cat.
― Aimless, Wednesday, 31 October 2012 21:30 (eleven years ago) link
did the numbers move apart dramatically in '08 (or '10 I guess)? I would assume that the percentages snowball as you get closer without any movement towards the underdog.
― Matt Armstrong, Wednesday, 31 October 2012 21:32 (eleven years ago) link
Obama basically never lost a 4+ (at least) point lead after September.
― the ones that I'm near most: fellow outcasts and ilxors (Alfred, Lord Sotosyn), Wednesday, 31 October 2012 21:36 (eleven years ago) link
I mean in nate's model did it go from like 80 to 95% in the last week or so?
― Matt Armstrong, Wednesday, 31 October 2012 21:36 (eleven years ago) link
As I understand it, because the announced margins of error of the polls do not change as the election nears, if the polls theoretically kept producing unchanged results week after week, Nate's "win percentages" probably would stay static, too.
― Aimless, Wednesday, 31 October 2012 21:42 (eleven years ago) link
fwiw both votamatic and the princeton guy have obama at 90%+ to win, kinda curious where nates 25% is coming from, theres def not that much chance the polls will move to romney win in the next week, a lot of it has to be mistrusting the polls or some other mysterious thing
― lag∞n, Wednesday, 31 October 2012 21:52 (eleven years ago) link
he hides behind flowers
― beef richards (Mr. Que), Wednesday, 31 October 2012 21:53 (eleven years ago) link
nate's model includes non-poll economic measurements that 'predict the future' but that is slowly removed from the model as the election approaches. so even if the polls stay the same his numbers would be changing.
xp
― iatee, Wednesday, 31 October 2012 22:02 (eleven years ago) link
but that is = but that are
ya but theres a week to go so that stuff shouldnt factor anymore
― lag∞n, Wednesday, 31 October 2012 22:10 (eleven years ago) link
I mean the economy is not going to shift between now and the election
― lag∞n, Wednesday, 31 October 2012 22:11 (eleven years ago) link
is that last job report supposed to come out friday? will nate silver have time to cook those numbers too?
― johnathan lee riche$ (mayor jingleberries), Wednesday, 31 October 2012 22:12 (eleven years ago) link
just put the numbers in a pot nate
― lag∞n, Wednesday, 31 October 2012 22:13 (eleven years ago) link
btw we are on a first name basis w this guy now