# The weirdness of polls... What's up? **URL:** \ **Category:** Politics & Elections\ **Created:** [October 17, 2012, 5:39pm UTC](https://boards.straightdope.com/t/the-weirdness-of-polls-whats-up/638033 "2012-10-17T17:39:08Z")\ **Posts on this page:** 20\ **Page:** 3
**Author:** ![Leaper](https://avatars.discourse-cdn.com/v4/letter/l/4bbf92/32.png) [@Leaper](https://boards.straightdope.com/u/Leaper)\ **Post date:** [October 18, 2012, 6:42pm UTC](https://boards.straightdope.com/t/the-weirdness-of-polls-whats-up/638033/41 "2012-10-18T18:42:29Z")
I’ve read that argument before, and it’s certainly reassuring for someone with my political views. But it still doesn’t feel right; doesn’t that sort of dynamic make national numbers (more) irrelevant? That seems to fly in the face of the conventional wisdom I know of, expressed repeatedly in this forum, that state numbers and national numbers are pretty tightly intertwined. ---
**Author:** ![adaher](https://avatars.discourse-cdn.com/v4/letter/a/dec6dc/32.png) [@adaher](https://boards.straightdope.com/u/adaher)\ **Post date:** [October 18, 2012, 6:57pm UTC](https://boards.straightdope.com/t/the-weirdness-of-polls-whats-up/638033/42 "2012-10-18T18:57:31Z")
> [@Leaper](#): > > Okay, fine, if Romney is up seven, then he should be leading in all the states that matter, right? Maybe I or someone can find their latest state-by-state numbers? They say Romney is up five in swing states, but they don’t break it down by state. ---
**Author:** ![aldiboronti](https://avatars.discourse-cdn.com/v4/letter/a/9fc348/32.png) [@aldiboronti](https://boards.straightdope.com/u/aldiboronti)\ **Post date:** [October 18, 2012, 6:58pm UTC](https://boards.straightdope.com/t/the-weirdness-of-polls-whats-up/638033/43 "2012-10-18T18:58:52Z")
[Romney takes first lead in electoral college.](http://www.realclearpolitics.com/epolls/2012/president/2012_elections_electoral_college_map.html) Another straw in the wind. ---
**Author:** ![HurricaneDitka](https://avatars.discourse-cdn.com/v4/letter/h/96bed5/32.png) [@HurricaneDitka](https://boards.straightdope.com/u/HurricaneDitka)\ **Post date:** [October 18, 2012, 6:59pm UTC](https://boards.straightdope.com/t/the-weirdness-of-polls-whats-up/638033/44 "2012-10-18T18:59:58Z")
> [@adaher](#): > > They say Romney is up five in swing states, but they don’t break it down by state. Yeah, it matters a lot if you’re counting Michigan and Pennsylvania as swing states or Missouri and North Carolina. ---
**Author:** ![RTFirefly](https://avatars.discourse-cdn.com/v4/letter/r/c77e96/32.png) [@RTFirefly](https://boards.straightdope.com/u/RTFirefly)\ **Post date:** [October 18, 2012, 7:44pm UTC](https://boards.straightdope.com/t/the-weirdness-of-polls-whats-up/638033/45 "2012-10-18T19:44:41Z")
> [@Gangster\_Octopus](#): > > The crosstabs show that all of the differenc is from the south. > > East: O 52, R 48 > Midwest: O 52, R 48 > South: O 39, R 61 > West: O 53, R 47 > > I don’t know what states make up which region, but it is a pretty stark contrast. Chances are good that they’re using [Census regions](http://www.eia.gov/emeu/mecs/mecs2002/census.html). ---
**Author:** ![Linear\_Crack](https://avatars.discourse-cdn.com/v4/letter/l/8e7dd6/32.png) [@Linear\_Crack](https://boards.straightdope.com/u/Linear_Crack)\ **Post date:** [October 18, 2012, 8:28pm UTC](https://boards.straightdope.com/t/the-weirdness-of-polls-whats-up/638033/46 "2012-10-18T20:28:28Z")
> [@Little\_Nemo](#): > > …If Obama was as obviously terrible as you’re saying, wouldn’t people have known it all along and given Romney a solid lead throughout his campaign? And if Obama has had some means of concealing his poor performance all this time, why would it suddenly stop working?.. Maybe there is some entity out there covering for his ongoing failures. -LC ---
**Author:** ![Richard\_Parker](https://avatars.discourse-cdn.com/v4/letter/r/35a633/32.png) [@Richard\_Parker](https://boards.straightdope.com/u/Richard_Parker)\ **Post date:** [October 18, 2012, 8:30pm UTC](https://boards.straightdope.com/t/the-weirdness-of-polls-whats-up/638033/47 "2012-10-18T20:30:03Z")
Apparently I am not a likely voter according to Gallup. You give the “voting” answer for 6 of the following 7 questions to be a likely voter (with some modifications for 18-22 year-olds and those who vote regularly by mail): Thought given to election (quite a lot, some) Know where people in neighborhood go to vote (yes) Voted in election precinct before (yes) How often vote (always, nearly always) Plan to vote in 2012 election (yes) Likelihood of voting on a 10-point scale (7-10) Voted in last presidential election (yes) Since I recently moved (and therefore haven’t voted in my precinct) and will be voting by absentee ballot this year only and therefore haven’t bothered to locate my neighborhood voting location (since I will be poll watching elsewhere as an attorney), I don’t count as a likely voter. I get that I’m likely a distinct minority here, but sort of humorous since I’m about the most likely voter possible. ---
**Author:** ![Jas09](https://avatars.discourse-cdn.com/v4/letter/j/d07c76/32.png) [@Jas09](https://boards.straightdope.com/u/Jas09)\ **Post date:** [October 18, 2012, 8:30pm UTC](https://boards.straightdope.com/t/the-weirdness-of-polls-whats-up/638033/48 "2012-10-18T20:30:56Z")
> [@Linear\_Crack](#): > > Maybe there is some entity out there covering for his ongoing failures. The DJIA and NASDAQ? BLS? ---
**Author:** ![Linear\_Crack](https://avatars.discourse-cdn.com/v4/letter/l/8e7dd6/32.png) [@Linear\_Crack](https://boards.straightdope.com/u/Linear_Crack)\ **Post date:** [October 18, 2012, 9:05pm UTC](https://boards.straightdope.com/t/the-weirdness-of-polls-whats-up/638033/49 "2012-10-18T21:05:29Z")
> [@Jas09](#): > > The DJIA and NASDAQ? BLS? I was thinking more along the lines of the MSM, DNC, and BLT. -LC ---
**Author:** ![Jas09](https://avatars.discourse-cdn.com/v4/letter/j/d07c76/32.png) [@Jas09](https://boards.straightdope.com/u/Jas09)\ **Post date:** [October 18, 2012, 9:08pm UTC](https://boards.straightdope.com/t/the-weirdness-of-polls-whats-up/638033/50 "2012-10-18T21:08:18Z")
It’s the DNC’s job to cover up for him. BLT’s are delicious. But the market’s are all-knowing, I’ve been led to believe. And the invisible hand sure seems to love itself some Obama. 🙂 ---
**Author:** ![Sam\_Stone](https://avatars.discourse-cdn.com/v4/letter/s/ecccb3/32.png) [@Sam\_Stone](https://boards.straightdope.com/u/Sam_Stone)\ **Post date:** [October 18, 2012, 9:58pm UTC](https://boards.straightdope.com/t/the-weirdness-of-polls-whats-up/638033/51 "2012-10-18T21:58:03Z")
> [@adaher](#): > > They say Romney is up five in swing states, but they don’t break it down by state. Here are the state-by-state battleground averages from RCP: **Romney Leading:** Georgia: Romney +12.3 Indiana: Romney +12.5 Montana: Romney +9.3 Missouri: Romney +7.7 North Carolina Romney +5.6 Arizona: Romney +5.3 Florida: Romney +2.5 Colorado: Romney +.7 **Obama Leading:** Connecticut: Obama +11.8 New Mexico: Obama +10.5 New Jersey: Obama +8.7 Oregon: Obama +8.5 Minnesota: Obama +7.3 Pennsylvania: Obama +5 Michigan: Obama +4.2 Nevada: Obama: +3 Ohio: Obama +2.4 Iowa: Obama: +2.3 Wisconsin: Obama +2 New Hampshire: Obama +.8 Virginia: Obama: +.8 So New Hampshire, Virginia, Wisconsin, Iowa, and Nevada look to be within reach of Romney. Florida and Colorado look to be in reach of Obama. Right now, RCP has Romney leading in the electoral college 206 to 201, with 131 electoral votes ranked as ‘toss-ups’. If you assign the toss-ups to the candidates in terms of who has the absolute lead now no matter how small, you get 38 electoral votes for Romney, and 93 for Obama. So Obama is still winning, and if the election were held today you have to call him as the favorite. However, all the trends are moving towards Romney - even after the last debate. The only three polls that have released results that incorporate polling done after the last debate all show gains for Romney. Rasmussen went from +1 to +2, Gallup from +6 to +7, and IBD from -2 to 0. So far, it doesn’t look like the debate stopped Romney’s momentum at all. What remains to be seen is if he can carry enough of it forward to cross the line ahead of Obama. ---
**Author:** ![Leaper](https://avatars.discourse-cdn.com/v4/letter/l/4bbf92/32.png) [@Leaper](https://boards.straightdope.com/u/Leaper)\ **Post date:** [October 18, 2012, 10:01pm UTC](https://boards.straightdope.com/t/the-weirdness-of-polls-whats-up/638033/52 "2012-10-18T22:01:53Z")
> [@Sam\_Stone](#): > > However, all the trends are moving towards Romney - even after the last debate. The only three polls that have released results that incorporate polling done after the last debate all show gains for Romney. (The following is not snark, but a serious question[ing].) So, just two days after the debate, these polls are adequately gauging its effect on the election at large? What was Romney’s position two days after the first? ---
**Author:** ![Sam\_Stone](https://avatars.discourse-cdn.com/v4/letter/s/ecccb3/32.png) [@Sam\_Stone](https://boards.straightdope.com/u/Sam_Stone)\ **Post date:** [October 18, 2012, 10:08pm UTC](https://boards.straightdope.com/t/the-weirdness-of-polls-whats-up/638033/53 "2012-10-18T22:08:46Z")
The point is that the three tracking polls are the only ones that incorporate polling data taken after the debate. I’m not making any other claim - if you think it takes several days for the debate results to percolate through the population, that’s fine. But all the other polls definitely only include pre-debate polling, so if you believe the debate changed opinions one way or the other, those polls should be discounted. ---
**Author:** ![Nzinga\_Seated](https://avatars.discourse-cdn.com/v4/letter/n/b5ac83/32.png) [@Nzinga\_Seated](https://boards.straightdope.com/u/Nzinga_Seated)\ **Post date:** [October 18, 2012, 10:50pm UTC](https://boards.straightdope.com/t/the-weirdness-of-polls-whats-up/638033/54 "2012-10-18T22:50:56Z")
Well, the jokes on me. I thought I wasn’t gonna get caught up in the polling. Thought I could point and laugh. Then I heard the latest Gallup poll. Now I’m feeling some of the butthurt. ---
**Author:** ![Leaper](https://avatars.discourse-cdn.com/v4/letter/l/4bbf92/32.png) [@Leaper](https://boards.straightdope.com/u/Leaper)\ **Post date:** [October 18, 2012, 11:33pm UTC](https://boards.straightdope.com/t/the-weirdness-of-polls-whats-up/638033/55 "2012-10-18T23:33:08Z")
It may or may not help, but Nate Silver [has a post up discussing Gallup’s poll in particular.](http://fivethirtyeight.blogs.nytimes.com/2012/10/18/gallup-vs-the-world/) His summation at the end comes down to: don’t ignore Gallup’s result entirely. Take it in the whole universe of current poll results, compare and contrast, and then decide if Gallup is right and everyone else is wrong, or the reverse. It seems evident which side he comes down on. ---
**Author:** ![Omg\_a\_Black\_Conservative](https://avatars.discourse-cdn.com/v4/letter/o/7ba0ec/32.png) [@Omg\_a\_Black\_Conservative](https://boards.straightdope.com/u/Omg_a_Black_Conservative)\ **Post date:** [October 19, 2012, 12:13am UTC](https://boards.straightdope.com/t/the-weirdness-of-polls-whats-up/638033/56 "2012-10-19T00:13:07Z")
You know, I remember a while ago when Gallup and Rasmussen were showing a closer race while the rest of the pollsters were showing Obama cruising to victory that Nate essentially played the “well, are you going to believe Gallup/Rasmussen or every other pollster?” card. ETA\> I do believe Gallup is using a D+3 model. Rasmussen is using a D+5 model (used to be a D+3). IBD/TIPP is using a D+7 model. I’d say Gallup is closer to the likely turnout, with Romney ahead right now maybe 3 - 4 percentage points. ---
**Author:** ![Little\_Nemo](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/little_nemo/32/3120_2.png) [@Little\_Nemo](https://boards.straightdope.com/u/Little_Nemo)\ **Post date:** [October 19, 2012, 12:36am UTC](https://boards.straightdope.com/t/the-weirdness-of-polls-whats-up/638033/57 "2012-10-19T00:36:54Z")
> [@Leaper](#): > > I’ve read that argument before, and it’s certainly reassuring for someone with my political views. But it still doesn’t feel right; doesn’t that sort of dynamic make national numbers (more) irrelevant? That seems to fly in the face of the conventional wisdom I know of, expressed repeatedly in this forum, that state numbers and national numbers are pretty tightly intertwined. It’s like the World Series. You win the championship by winning four games. How many home runs you score in those games doesn’t directly matter - you can score one run and win a game or score ten runs and lose a game. And the series won’t be determined by how many runs you scored during it. But how many runs you score is related to how many games you win. So a team that scores more runs in the series will usually beat a team that scores fewer runs. Due to the Electoral College, you have a similar effect in Presidential elections. The guy who wins is the one who has the majority of electoral votes, not the guy who has the most popular votes. But generally the guy who gets the most popular votes is also the guy who gets the most electoral votes. ---
**Author:** ![GrandWino](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/grandwino/32/67_2.png) [@GrandWino](https://boards.straightdope.com/u/GrandWino)\ **Post date:** [October 19, 2012, 12:51am UTC](https://boards.straightdope.com/t/the-weirdness-of-polls-whats-up/638033/58 "2012-10-19T00:51:06Z")
> [@Leaper](#): > > His summation at the end comes down to: don’t ignore Gallup’s result entirely. Take it in the whole universe of current poll results, compare and contrast, and then decide if Gallup is right and everyone else is wrong, or the reverse. It seems evident which side he comes down on. > [@Omg\_a\_Black\_Conservative](#): > > You know, I remember a while ago when Gallup and Rasmussen were showing a closer race while the rest of the pollsters were showing Obama cruising to victory that Nate essentially played the “well, are you going to believe Gallup/Rasmussen or every other pollster?” card. > > ETA\> I do believe Gallup is using a D+3 model. Rasmussen is using a D+5 model (used to be a D+3). IBD/TIPP is using a D+7 model. I’d say Gallup is closer to the likely turnout, with Romney ahead right now maybe 3 - 4 percentage points. Silver’s post has some good information about Gallup’s history of being an incorrect outlier over the last couple of election cycles. When it’s way off from the rest of the pack, it tends to be wrong. We’ll see though. ---
**Author:** ![Omg\_a\_Black\_Conservative](https://avatars.discourse-cdn.com/v4/letter/o/7ba0ec/32.png) [@Omg\_a\_Black\_Conservative](https://boards.straightdope.com/u/Omg_a_Black_Conservative)\ **Post date:** [October 19, 2012, 1:02am UTC](https://boards.straightdope.com/t/the-weirdness-of-polls-whats-up/638033/59 "2012-10-19T01:02:19Z")
I’m far from unbiased, but I’ll just say I don’t think they’re as far off as some people are making it due to the fact that many polls this year have had ridiculous partisan splits and other weird anomalies. For example, I just got done looking at the internals of the [IBD/TIPP tracking poll](http://news.investors.com/special-report/508415-ibdtipp-poll.aspx) and they don’t look good for Obama. Romney is +11 among Independents (48 - 37), -4 among women (44 - 48) and +3 among men (47 - 44) while Obama is only retaining 84% of his vote from 2008 and picking up 4% of McCain voters from 2008 compared to Romney retaining 92% of the McCain vote from 2008 and picking up 10% of Obama voters from 2008. Yet the same poll has Obama ahead 46.2% to 45.7%. Many polls are like this, where the internals are terrible for Obama yet the topline has him ahead. It leads to me believe that many pollsters are using some ridiculous partisan split-- where Democratic turnout is either really high, Republican turnout is really low, or a combination of both-- in which case the topline is probably two or three percentage points more “pro-Romney” than they are indicating. So Gallup is probably overstating Romney’s performance by 2 - 3 percentage points while other pollsters are understating Romney’s performance by 2 - 3 points. ---
**Author:** ![Leaper](https://avatars.discourse-cdn.com/v4/letter/l/4bbf92/32.png) [@Leaper](https://boards.straightdope.com/u/Leaper)\ **Post date:** [October 19, 2012, 1:47am UTC](https://boards.straightdope.com/t/the-weirdness-of-polls-whats-up/638033/60 "2012-10-19T01:47:51Z")
You’ve brought this up before. Do you think IBD/TIPP is ignorant of this? Or that they’re trying to help Obama (as in another accusation of poll manipulation)? But I’ve sent an e-mail to Nate Silver asking about this, because I’m curious to see what his response is. In the meantime, does anyone else have the knowledge to address this point? [Previous page](https://boards.straightdope.com/t/the-weirdness-of-polls-whats-up/638033.md?page=2) [Next page](https://boards.straightdope.com/t/the-weirdness-of-polls-whats-up/638033.md?page=4)