MOBILE USERS: m.isthmus.com
Connect with Isthmus:         Newsletters 
Wednesday, December 24, 2014 |  Madison, WI: 35.0° F  Overcast
Collapse Photo Bar

Nate Silver's Flawed Model

Races for the Senate, U.S. House, etc. and other issues of national importance.

Re: Nate Silver's Flawed Model

Postby Detritus » Mon Nov 05, 2012 12:06 am

pjbogart wrote:These people do polls for a living. Bludgeon and pjbogart don't. They have no interest in looking like shitheads when the final tally is wildly different than their poll.

Does pjbogart normally refer to himself in the third person?
Detritus
Forum God/Goddess
 
Posts: 2542
Joined: Wed May 13, 2009 9:42 pm

Re: Nate Silver's Flawed Model

Postby kurt_w » Mon Nov 05, 2012 9:58 am

Bludgeon's argument boils down to this:

    1. The polls don't show enough people who call themselves "Republican"

    2. The polls show Mitt Romney with a small lead among people who call themselves "Independent"

    3. If you add more "Republicans" but keep the number of "Independents" the same, Romney would be leading.
This is what I mean when I say that Bludgeon is suffering from a severe case of self-delusion. The flaw in that three-point argument is obvious and straightforward. If it weren't about an election that Bludgeon is strongly emotionally invested in, I'm sure that he/she (?) would have no problem spotting the error in that line of reasoning.

Bludgeon is double-counting Republicans. Post-2008, a bunch of people who would formerly have called themselves "Republican" in polls now call themselves "Independent", even though they're still inclined to vote for Republican candidates.

If you count those people's votes as independents ("independents favor Romney!") and also try to add them back to the Republican total, you're inflating the Republican vote by double-counting.

I have pointed this out before and Bludgeon had no response, so I'm not optimistic about it sinking in, but might as well try.
kurt_w
Forum God/Goddess
 
Posts: 5408
Joined: Tue Jul 22, 2003 3:11 pm

Re: Nate Silver's Flawed Model

Postby Huckleby » Mon Nov 05, 2012 12:29 pm

The problem with polls is they don't do a good job of predicting turnout.

Nate Silver's latest projection of a 85% chance of victory simply looks at the polls at this late stage of the game, and says that 85% of past elections like this swung to the person leading in the polls.

I believe the polls. I don't buy any of the oversampling chicanery.

I think the votes are tight enough in the swing states that differences in passion can easily override Obama's narrow, consistent polling advantages. Or put another way, this is a far more high stakes and emotional election than in past instances. So I take Nate Silver's predictions with a giant block of salt.

I don't think Republicans are whistling past the graveyard, this thing looks like a tossup with only a slight edge to Obama.
Huckleby
Forum God/Goddess
 
Posts: 7231
Joined: Thu Oct 20, 2005 5:12 pm
Location: parents' basement

Re: Nate Silver's Flawed Model

Postby lukpac » Mon Nov 05, 2012 12:34 pm

A few relevant bits:

https://twitter.com/neiltyson/status/265183385847209985

Neil deGrasse Tyson wrote:It's just Math: If you win hardly any too-close-to-call polls in a state (eg. Romney/Ohio) then you'll likely lose the State.


https://twitter.com/neiltyson/status/265184140184412160

Neil deGrasse Tyson wrote:It's just Math: If two candidates are actually dead-even in votes, each will win about half of all "too-close-to-call" polls.


http://fivethirtyeight.blogs.nytimes.co ... ly-biased/

Although the fact that Mr. Obama held the lead in so many polls is partly coincidental — there weren’t any polls of North Carolina on Friday, for instance, which is Mr. Romney’s strongest battleground state — they nevertheless represent powerful evidence against the idea that the race is a “tossup.” A tossup race isn’t likely to produce 19 leads for one candidate and one for the other — any more than a fair coin is likely to come up heads 19 times and tails just once in 20 tosses. (The probability of a fair coin doing so is about 1 chance in 50,000.)

Instead, Mr. Romney will have to hope that the coin isn’t fair, and instead has been weighted to Mr. Obama’s advantage. In other words, he’ll have to hope that the polls have been biased in Mr. Obama’s favor. (I recognize that ‘bias’ is a loaded term in political contexts. I’ll explain what I mean by it in a moment.)
[...]
My argument, rather, is this: we’ve about reached the point where if Mr. Romney wins, it can only be because the polls have been biased against him. Almost all of the chance that Mr. Romney has in the FiveThirtyEight forecast, about 16 percent to win the Electoral College, reflects this possibility.

Yes, of course: most of the arguments that the polls are necessarily biased against Mr. Romney reflect little more than wishful thinking.

Nevertheless, these arguments are potentially more intellectually coherent than the ones that propose that the leader in the race is “too close to call.” It isn’t. If the state polls are right, then Mr. Obama will win the Electoral College. If you can’t acknowledge that after a day when Mr. Obama leads 19 out of 20 swing-state polls, then you should abandon the pretense that your goal is to inform rather than entertain the public.

But the state polls may not be right. They could be biased. Based on the historical reliability of polls, we put the chance that they will be biased enough to elect Mr. Romney at 16 percent.
lukpac
Forum God/Goddess
 
Posts: 3362
Joined: Mon Feb 07, 2005 1:51 pm
Location: Madison

Re: Nate Silver's Flawed Model

Postby rrnate » Mon Nov 05, 2012 12:38 pm

Yup, Obama has this one locked up guys. It's as easy as looking at how many folks in the Republican leadership are broadcasting excuses for why they're going to lose - Rove, Barber, etc.

Romney's out there on his own now and he'll be blamed squarely by the folks in charge when he loses.
rrnate
Forum God/Goddess
 
Posts: 3668
Joined: Sat Feb 02, 2002 6:33 pm
Location: Madison's Corporate Underbelly

Re: Nate Silver's Flawed Model

Postby rabble » Mon Nov 05, 2012 1:47 pm

Huckleby wrote: Or put another way, this is a far more high stakes and emotional election than in past instances.

Funny. I could swear somebody else said that before.

Oh, yeah that's right. Last three or four elections.
rabble
Forum God/Goddess
 
Posts: 6550
Joined: Tue Jun 10, 2008 7:50 pm

Re: Nate Silver's Flawed Model

Postby Huckleby » Mon Nov 05, 2012 5:09 pm

For Republicans, the enactment of Obamacare is their Armageddon.

Never been an election with so much at stake in my memory.
Huckleby
Forum God/Goddess
 
Posts: 7231
Joined: Thu Oct 20, 2005 5:12 pm
Location: parents' basement

Re: Nate Silver's Flawed Model

Postby Bludgeon » Mon Nov 05, 2012 6:57 pm

kurt_w wrote:This is what I mean when I say that Bludgeon is suffering from a severe case of self-delusion. The flaw in that three-point argument is obvious and straightforward. If it weren't about an election that Bludgeon is strongly emotionally invested in, I'm sure that he/she (?) would have no problem spotting the error in that line of reasoning.

Bludgeon is double-counting Republicans. Post-2008, a bunch of people who would formerly have called themselves "Republican" in polls now call themselves "Independent", even though they're still inclined to vote for Republican candidates.

If you count those people's votes as independents ("independents favor Romney!") and also try to add them back to the Republican total, you're inflating the Republican vote by double-counting.

I have pointed this out before and Bludgeon had no response, so I'm not optimistic about it sinking in, but might as well try.

Strongly emotionally invested...? No. Ideologically invested, sure. Hanging out on Netflix, watching "Newlyweds" starring Ed Burns, nice movie, and doing what I have liked to do for the last ten years: hang out on TDPF and shoot politics with you's bastards.

Either candidate might win. There are about the same number of independents as there were in 2008. What's increased is the number of Republicans; they're missing in a number of polls. 2008 was the Democrats' high water mark for the partisan split over the last 32 years. To say 2012's electorate will feature more Republicans than 2008 is a given by any standard, so dispute it if you will?

I don't know who's going to win but I do know Republicans and Democrats are likely to be near tied, advantage debatable.

Image
Bludgeon
Forum God/Goddess
 
Posts: 1291
Joined: Tue Apr 06, 2010 2:27 am

Re: Nate Silver's Flawed Model

Postby lukpac » Mon Nov 05, 2012 7:37 pm

Bludgeon wrote:I don't know who's going to win but I do know Republicans and Democrats are likely to be near tied, advantage debatable.


See above. Does Romney have a chance? Of course. Is it as likely a chance as Obama? No.
lukpac
Forum God/Goddess
 
Posts: 3362
Joined: Mon Feb 07, 2005 1:51 pm
Location: Madison

Re: Nate Silver's Flawed Model

Postby Mad Howler » Mon Nov 05, 2012 8:44 pm

rrnate wrote:Romney's out there on his own now and he'll be blamed squarely by the folks in charge when he loses.


The folks in charge didn't seem to want to release a better product this season, perhaps they perceive fairer weather on the horizon. When you look at the cast of characters they trotted to the table for this race does anyone think they were trying? Whatever happens with this cycle I can not forgive this lack of effort from what seems to be fair weather players.
Mad Howler
Forum God/Goddess
 
Posts: 1464
Joined: Wed Apr 04, 2012 8:36 pm

Re: Nate Silver's Flawed Model

Postby rabble » Mon Nov 05, 2012 8:59 pm

Mad Howler wrote:The folks in charge didn't seem to want to release a better product this season, perhaps they perceive fairer weather on the horizon. When you look at the cast of characters they trotted to the table for this race does anyone think they were trying? Whatever happens with this cycle I can not forgive this lack of effort from what seems to be fair weather players.

They still haven't figured out how to control the Tea Party they created without much forethought or planning ahead. It was a tool that had a great short term effect. Then it didn't go away like it was supposed to and is the reason they can't run the people they really wanted. So they're making noise and paying lip service while they wait it out.
rabble
Forum God/Goddess
 
Posts: 6550
Joined: Tue Jun 10, 2008 7:50 pm

Re: Nate Silver's Flawed Model

Postby Mad Howler » Mon Nov 05, 2012 9:17 pm

rabble wrote:
Mad Howler wrote:The folks in charge didn't seem to want to release a better product this season, perhaps they perceive fairer weather on the horizon. When you look at the cast of characters they trotted to the table for this race does anyone think they were trying? Whatever happens with this cycle I can not forgive this lack of effort from what seems to be fair weather players.

They still haven't figured out how to control the Tea Party they created without much forethought or planning ahead. It was a tool that had a great short term effect. Then it didn't go away like it was supposed to and is the reason they can't run the people they really wanted. So they're making noise and paying lip service while they wait it out.


I get what you are saying, but I will not excuse the mess that has been conjured over the last 40 or so years. Maybe it's just me, but I expect excellence from either side of our political dichotomy. Those inclined to mess with our public affairs should lead or get the fuck out of the way. Perhaps this may seem a tad idealistic, but there seems to be an extreme disortion in our political "economy" and we all better be ready for that bubble to break.
Mad Howler
Forum God/Goddess
 
Posts: 1464
Joined: Wed Apr 04, 2012 8:36 pm

Re: Nate Silver's Flawed Model

Postby pjbogart » Mon Nov 05, 2012 10:49 pm

Yikes! Nate Silver's election prediction has become a runaway train. Serious jumps today give Obama a 92.2% chance of winning with electoral votes of 315.3 - 222.7. Those numbers essentially would require Obama to win every single swing state except for one, or alternatively lose Florida but win North Carolina (which I no longer consider a swing State).

That's pretty bold and I'm sure he realizes that everyone is watching to see how close he comes, so he doesn't have much interest in jumping the shark.

Considering that there are already reports of voter suppression in Florida, at least minor hurricane cleanup in North Carolina and Virginia and a Republican Party determined to disenfranchise as many Ohio voters as they can get away with, predicting a blowout tomorrow seems like wishful thinking.

I sure hope he's right, though.
pjbogart
Forum God/Goddess
 
Posts: 6189
Joined: Thu Sep 04, 2003 4:57 pm

Re: Nate Silver's Flawed Model

Postby pjbogart » Mon Nov 05, 2012 11:29 pm

Stop the presses! Don't lay ink on those inauguration tickets just yet. Dick Morris says the final tally willy be Romney 325-Obama 213. Now those are some numbers that Bludgeon can live with!

• By focusing on the negative, Obama sacrificed first his personal popularity and then his dignity and presidentiality. No longer was he the hope and the change. He became nothing more than a nasty partisan, throwing epithets at his rival. A president does not let himself be quoted as saying that his opponent is a “bullsh--ter” or that voting is the best “revenge.” Even his dress was wrong. Instead of appearing in a dark suit, he dressed in an open-neck white shirt, trying to be everyman but succeeding only in not looking like a president.


Prediction: Romney 325, Obama 213

Well, there you have it. A prediction even bolder than Silver's. We'll see who comes closer to the final outcome.
pjbogart
Forum God/Goddess
 
Posts: 6189
Joined: Thu Sep 04, 2003 4:57 pm

Re: Nate Silver's Flawed Model

Postby Mad Howler » Mon Nov 05, 2012 11:41 pm

Interesting retweet by Silver:

“@nytjim: Political insomniacs: If you haven't played with "512 Paths to the White House," tonight's the night. http://t.co/dfpH5b6u
Mad Howler
Forum God/Goddess
 
Posts: 1464
Joined: Wed Apr 04, 2012 8:36 pm

PreviousNext

Return to National Politics & Government

Who is online

Users browsing this forum: Bing [Bot] and 2 guests

moviesmusiceats
Select a Movie
Select a Theater


commentsViewedForum
  ISTHMUS FLICKR
Created with flickr badge.

Promotions Contact us Privacy Policy Jobs Newsletters RSS
Collapse Photo Bar