Princeton Election Consortium

A first draft of electoral history. Since 2004

David Brooks – now with Ro-mentum!

October 25th, 2012, 10:30am by Sam Wang


(Welcome, readers of FiveThirtyEight, Ezra Klein, and Votamatic!)

David "Bobo" BrooksIt was fun to learn of David Brooks’s addiction to polling data. He spends countless hours on them, looking at aggregators, examining individual polls, and sniffing poll internals. From all of this, what has he learned?

  1. Today, President Obama would be a bit more likely to win.
  2. There seems to be a whiff of momentum toward Mitt Romney.

(Emphasis mine.)

I am having a sad. All of that effort, and his two conclusions still have two major errors. Evidently he does not read the Princeton Election Consortium. Let us dissect this.

1. President Obama would be a bit more likely to win. This is false – he’s a lot more likely to win. Look at the Princeton Election Consortium’s EV histogram, which tabulates all 2.3 quadrillion possible combinations of states to give a clear snapshot of the race:

Today's electoral vote histogram

In a race today, President Obama would win with about 90% probability. The true probability is even higher, since the Meta-Analysis does not correct for individual pollster errors. We could – but the political blowback from unskewing polls is too large.

2. There seems to be a whiff of momentum toward Mitt Romney. Ah, yes…Ro-mentum! Bobo has taken the bait. He is probably looking at other aggregators, where for various reasons (q: do you want me to write about that sometime?) the real trends are harder to see. Let’s roll the instant replay.

As you can see, Ro-mentum ended around October 11th, the date of the VP Biden-Ryan debate and reversed around October 16th, Debate #2. Now the median EV expectation is at a plateau around Obama 293 EV, Romney 245 EV. Viewed through the all-important Electoral College, Obama has a Popular Vote Meta-Margin lead of 1.5%. This measure is precise to within <0.5%, far better than any single poll. If anything, the race is starting to look a bit static.

Of course, some change may well happen over the coming 12 days. Based on past races (see “The Presidential Predictor sharpens,” Sept. 29), here is how much movement we can expect.

The “1 SD band” indicates the 1-sigma range for future Meta-Margin change. The election is 12 days away, over which we could expect movement of up to 1.4% in either direction – about equal to the current Meta-Margin. If drift were random, Romney’s chances of catching up would be 1 in 6 – the roll of a die.

But which way will things actually go?

The largest unknown factor that might help Romney is undecided voters. In national surveys ending on Oct. 23 or after, there are still 4.0 +/- 0.7 % undecideds (n=7). Based on past elections (“How will the last holdouts break?” 11/3/2008), undecideds break about equally, with a tiny advantage for the challenger. Romney can expect a net benefit of +0.3 +/- 0.8%. That benefit is a main contributor to the 1 in 6 chance I give him above. However, the uncertainty (0.8%) is larger than the average benefit (0.3%), so it could also hurt him.

In President Obama’s favor, he (a) won debate #3, and (b) has had a Meta-Margin lead of +3.3+/- 1.3% this year. Movement is more likely toward this mean than away from it. The red strike zone in our history graph reflects this, and gives a re-elect probability of 89%.

Don’t be like David Brooks. Finally, a word on examining individual polls. In the comments section, I am seeing breathless statements like how the TIME poll shows Obama +5% in Ohio, and won’t that move the Meta-Margin, and so on. Hey…relax. In case you haven’t noticed, the Meta-Margin’s not moving that much. Try to avoid caressing those individual polls too much!

With that, I end with something in Brooks’s column that I did like:

If you do have to look at polls, you should do it no more than once every few days, to get a general sense of the state of the race.

That’s not bad advice. Instead, spend the time on GOTV in a swing district near you. Here they are.

Finally: in comments, I encourage you to avoid citing individual polls. Give it a try.

Tags: 2012 Election · President

295 Comments so far ↓

  • piktor

    Call it the ‘ Unstoppable Mittmentum ‘ !

    It was coined at TPM in 2008, by one of the savvy journalists there.

    We all know how all that unstoppability ended.

  • Rick

    I see new CO and FL polls from a Democratic pollster that will move the MM. However, I’m not sure how I feel about including partisan polls from either side. In some ways I think RCP is right to exclude them. Surely no one thinks that these pollsters are above massaging the numbers or releasing only favorable polls. And yet they’re included in the pollster data base. I suppose that the argument would be that they cancel each other out, but crap canceling out crap doesn’t seem like a good way to go. For 2016, we need an accrediting agency for pollsters.

    • CobaltinSF

      You don’t think Rasmussen is a partisan poll? It comprises 25%-35% of RCP poll numbers. It appears to consistently skew Republican by 2-4%. I’ve come to conclude that RCP is the Fox News-lite of polling sites.

    • Justin S

      RCP includes Gravis as well who are even more partisan than Rass

    • Lou

      Mark Blumenthal from Pollster.com and Andy Kohut from Pew appeared on the The Diane Rehm Show to talk about polling and made the point that so-called “partisan” polls, in their judgment, generally very accurate and conducted by professional pollsters with high standards. That doesn’t mean that they don’t have “house effects,” but their point was that parttisan polls shouldn’t be automatically discounted. I don’t know whether they are right or not, but they are both respected pollsters and thought you might be interested in their point of view.

    • Steve

      Huff Post Pollster for example lists the new Gravis poll for Florida.

      http://elections.huffingtonpost.com/pollster/2012-florida-president-romney-vs-obama

      Pollster believes that Gravis has a republican house effect and adjusts for that in their model. I am OK with pollsters that have an observed house effect – even though they may be partisan. Models can adjust for those if they wish – and readers can choose the models they want to follow.

      I am looking at all the popular models now and love the MM because it is frequently updated and has a transparent algorithm. There is broad agreement among the models about swing state predictions with the exception of New Hampshire. We await more polls …

      As a Democratic partisan, I am hoping that Florida will flip from Romney to Obama as the polls leave behind the effects of the debates and begin to incorporate Florida absentee and early voting results – the latter starting this weekend and continuing for a total of 96 hours in the more populated Florida counties.

    • Boberto

      I’m ending my lurking days on this post…

      I really like that this site just includes it all, with the assumption that biases will cancel out. This assumes an underlying equality of both error and tendency toward thumb-on-the-scale behavior across the partisan divide (if it exists), which may or may not be true (there does seem to be a proliferation of low quality, high volume right-leaning pollsters – which RCP seems like it really wants to include, stated rules be darned).

      However, as a scientist myself, I’m pretty hesitant about throwing out data. If the MM was literally hugging the 50/50 line for months, and I felt that some clarity might come out of seeing the results of only what I maybe even subjectively felt to be higher quality data, yeah, I’d want to start sifting even if it was just dinking around in R to see what it’d do. Right now though, I’m reassured that the bias I’d assume may be in the system is counter to what I’d hope actually occurs (An Obama win is being indicated, despite my worry that there may be a higher degree of partisan polling on the right). That seems less scary to me than the always lurking problem of confirmation bias if we start trying to pick and choose.

    • ChrisD

      There’s a new PPP poll on Pollster that has NC tied. Its mid-date is 10/24, so it will knock off 2 0r 3 of the Romney-favored polls in the 5 pm update. NC will then be either tied or O+1.5, and the MM will rise again, EEBE.

      http://elections.huffingtonpost.com/pollster/2012-north-carolina-president-romney-vs-obama

  • Olav Grinde

    Question: Anyone have a link to the exact page where Pollster.com lists its latest state-level and national polls? And latest polls of Senate and House races (without having to click on every single state)? I can’t seem to find them.

  • Larry Pareigis

    Dr. Wang, long-time lurker, first-time poster. Discovered you through Andrew (the Human In-Trade) Sullivan’s blog and just wanted to say thanks for your cogent analysis & puckish wit this election season. It’s made your blog a several-times-a-day mandatory read. Cheers.

  • zenger

    “Steve, because now they want it back to O to create the illusion of a horse race, and can get more desperate ad dollars. This is the same thing they do every election.”

    Worked in several newsrooms. This does not happen. There is no convocation of a monolithic media to determine narrative. Ad Department has zero say.

    Love this site. Fascinated by the statistical analysis. The media analysis, however, doesn’t measure up. Probably because it can’t be measured.

    • Reason

      Cousin works for a large media outlet. Does happen. She explained producers, for whatever reason, tell “journalists” to ignore one thing for another to give the story “legs”. Her words, not mine. I give what she says more credence, no offense.

    • Jack

      I’ve also worked in a newsroom, and I think the manipulation of the story happens more on television (particularly cable news) than print or Internet. The need to keep viewers tuned in requires stories pushing tension and anticipation. Newspapers tend to give more in-depth coverage. Internet feeds ones beliefs: users typically go to websites that reinforce their established views.

    • Mike

      It may not have happened in the newsrooms you were in, but it has happened, even historically (William Randolph Hearst, anyone?), and does happen today.

  • Tim in CA

    This post is hysterical. It’s always refreshing to see Mr. Brooks called out when he strays into the ridiculous…

  • Chris in NC

    Whenever Brooks makes comments like the ones Prof. Wang cited, it means he’s on a subject he doesn’t know a lot about. But he just had to get in on the subject because it’s popular now.

    I love it when people like him boil all of the detailed work that experts like Prof. Wang do to “ah, looks like a close one, with the President a bit more likely to win”. That’s the type of thing you say before looking at any evidence at all. I would say this is one of Brooks’ mailed in columns.

  • Ronk

    Sam,
    How does the concept of regression to the mean work in these analyses as the polls accumulate? What would it predict relative to what we are seeing?

    • Sam Wang

      The idea is that those people polled in June-September did not disappear – and at the time, they had an opinion. Therefore surveys from that period tell us something about the likely range of future opinion. In other words, the race will tend to move in the direction of where it’s been in the past. For a re-election race this is especially true.

  • Ross C

    The 3pm MM up to 1.82, with O getting 297 EVs. Ro-mentum!

  • Stuart

    Huffpost now has O leading in Florida as well as CO, NH and VA.

    http://elections.huffingtonpost.com/2012/romney-vs-obama-electoral-map

    Now that really is good news for Ro-mentum.

  • Ms. Jay Sheckley

    Dr Wang has asked us to _try_ not talking abut individual polls. And what an easy day to do that. Beautiful numbers.
    As of October 25, 3:00PM EDT:
    Obama: 297
    Romney: 241
    Meta-margin: Obama +1.82%
    It’s been proven that individual polls are nowhere near as accurate as PEC.
    Cmon people!

    • Olav Grinde

      Good numbers, indeed.
      I’m hoping we’ll soon see New Hampshire and Florida colored light blue as well.

    • E L

      C’mon, Ms. Sheckley: Dr. Wang is asking us to act like responsible, rational adults. What fun is there in that? I’m too old.

  • ChrisD

    Pollster just put up the new MI poll, a tie, from FMWB (Foster McCollum White Baydoun ). MI will drop, I think, from O+7.5 to O+6 in the 5pm update. Read halfway down the page in the link below to see what Nate Silver thinks of this polling outfit:

    http://fivethirtyeight.blogs.nytimes.com/2012/08/20/aug-20-when-the-polling-gets-weird/

    • Froggy

      BFD. A drop from O+7.5 to O+6 in one state (actually I have it as a drop from O+7 to O+6) isn’t even worth 0.01% on the MM.

    • ChrisD

      Gee, Froggy. I’m not claiming it’s a BFD. Just noting its appearance and offering info for anyone who cares to know who the pollster is.

    • Froggy

      Chris, my apologies for the dismissive and nasty tone of my earlier message. It was uncalled for.

    • ChrisD

      Thanks, Froggy. I realize my tendency to be an obsessive poll-watcher can seem annoying, but I try to be objective about what they report, be it good or bad for Obama.

  • Steven J. Wangsness

    Thanks, I think, to the folks above who mentioned Electoral Scoreboard and Pollyvote. Along with Desart and Holbrook, whom I read about yesterday on Votamatic, I now have an ever-growing stable of sites to visit to feed my election OCD.

    Thank God this’ll be over in 12 days (unless it’s decisive-provisional-ballots time in Ohio, please God no…).

  • Rick in Miami

    Sigh … new Washington Post “The Fix” by Chris Cillizza concludes that Ro-mentum is real, or at least suggests it is:

    http://www.washingtonpost.com/blogs/the-fix/wp/2012/10/25/is-mitt-romneys-momentum-real-or-fake/?hpid=z1

    His evidence? Romney is doing better in the polls now, compared to before debate 1. In summary, two points -> trend -> momentum. Maybe he should look at a fuller time series.

    • Steve in Colorado

      My read of that was that we wasn’t disagreeing with either conclusion- Romney definitely had some momentum after the 1st debate, but polls have flattened for him recently.
      He seems a bit of a puss, but his conclusion seemed to be:
      “Depends how you define momentum”

    • Reason

      Cilliza has always been a right wing hack who pretends he is objective. Ignore what he writes and says. We here in the DC area have for years now.

    • Rick in Miami

      It seemed to me to lean pretty heavily towards Ro-mentum. I didn’t see that article noting anywhere that the polls have flattened recently. Every comparison seems to be between two points in time: immediately pre-debate #1 and now. And it concludes, “What we don’t [know] is whether Romney has peaked or not. But it’s tough to argue that there hasn’t been real movement toward him in the past three weeks. “

    • Rick in Miami

      Gotcha, Reason. Given that “The Fix” is often a boilerplate of bullet items, I hadn’t picked up on a right-wing slant before and thought this was another case of “conventional wisdom” in the media.

    • Pankaj

      Chris Cilliza is one of the worst beltway pundits. He hardly gets anything right. I put him in same category as Chuck Todd, Andrea Mitchell, Mark Halperin.

    • Joey Bagadonitz

      Yeah, the Fix is garbage. From what I can tell Cilizza is definitely not on anyone’s short list of particularly influential writers here in town.

    • Joey Bagadonitz

      *Here in town meaning, DC.

    • Obama 2012

      over at 538 the story is similar to what it is here… at 538 Romney peaked on October 12th… — and Obama was still the favorite at 56% in the Now Cast) — since then Obama has gained 18% points to his current 74% level.

      If you look at the MM here you see that it dropped to it’s lowest point at about the same point and since then it has been climbing up for Obama.

      It’s quite obvious that if anyone has momentum in this race it’s President Obama.

    • paddler

      Chris Cillizza is a perfect snap shot of what the DC pundit class is thinking about a given subject at a given point in time. Whatever he’s saying is the group think of the moment one way or the other. Don’t hate the player hate the game.

  • Reason

    What I am most impressed with, is that Dr Wang started the Ro-mentum schtick yesterday and now the MM is higher. I feel I should ask him what are the winning lottery numbers tomorrow. I also sent this site to family and friends who are political junkies like I am. It rules.

    • Obama 2012

      there have been a lot of very good state polls for Obama today, I’m expecting the MM to rise again before the day is through.

  • The Red Pen

    Unskewed Polls has Romney winning 359 EVs including Oregon.

  • Richard Vance

    Ladies and gents, The talking heads and the print journalists all have deadlines and hope to get listened to/read. Thus they talk endlessly about insignificant things to fill air time/blank space. That’s their jobs just understand that and go on with your day..

    I really enjoy refreshing the PEC tab to see what joys may appear. Thanks Sam.

  • Les Honig

    But dig this counter article…new ABC poll showing Romney extending his lead to 3 points…50-47…will that feed the R momentum? And is Sam accounting for it yet? Bummer!

    • Obama 2012

      This site only uses state polls — I wouldn’t get too worried about these LV national trackers for multiple reasons:

      (1) their LV models are probably bogus. most of these polls show huge LV swings for Romney from the RV #s, I just don’t think there’s any convincing argument for this.

      (2) they don’t have a good track record.

      (3) the #s they are spitting out don’t make any sense in relationship to all of the state polling.

      (4) yes there are a few with Romney leads, but there are also a few with Obama leads so there’s no clear pro-Romney signal from the daily trackers.

    • jd

      I saw the same article, and just started laughing. The article is written by Gary Langer, who also owns the polling company Langer Research, who conducted the poll.Talk about self serving BS. Anyways this site as been an outstanding educational tool for my children and myself.

    • Richard Vance

      National Polls are useless. We have that old Electoral Vote system. Here in Alabama Romney will likely take 88-90% of the vote but all votes above the one that gives him the state’s entire EV are useless, yet those are counted in the national polls. The only thing that counts now is actual turnout in swing states in early vote, election day, and absentees. Obama can be ahead by 30% in the polls but if those who prefer him don’t show up in force its hello President Romney. We must focus on the ground game, I suggest that for close elections this PEC method can be improved by estimating the resources input to and efficiency of the ground game…

      Do people track that? If we now have American Crossroads spending $12 million on Ohio turnout and no Democrat counter, the Ohio poll and thus the PEC estimate is pretty much meaningless.

  • KentK

    It’s reached the stage that I go into a decline when the MM drops by 0.02% – as it just did!

  • DrOrbit

    Sam only does state polling. That abc poll is bogus.

  • Matt

    I think this election is good for punditry, or at least I am optimistic that it will prove good for punditry as a profession. Jim Fallows articulated the case very well in a recent blog post. We have a very real disagreement between the old guard pundits, like David Brooks, who base their arguments on intuition, conversations with insiders and other non quantitative measurements. On the otherside we have the new class of pundits, like Dr Wang, who are basing their arguments entirely on actual measurements of who is winning and a statistical analysis of those measurements. This isn’t the academy awards, after all. We do have an actual and very thorough measurement of who is winning. We also have a looming massive and completely through experiment of that measurement.

    I not only hope that this divide remains through the election, I actually hope it becomes more accute. I want one side to be proven spatacularly wrong by the other in the hope that those proven wrong are discredited.

    At least one can hope, right?

    • Reason

      Except we saw what happened with Bush and now it looks like a lot want to bring in a guy who is even worse. And the Tea Party.

  • DrOrbit

    Why is the MM now at 1.7 now. It was just 1.8.

    • ChrisD

      It dropped just a smidgen, from 1.82 to 1.78. VA dropped from O+1.5 to O+1.0 and MI from O+7 to O+6, while PA and WI (I think) ticked up. So offsetting impacts.

  • Reason

    What happened in VA and is Sam counting that poll that has MI even?

    • Obama 2012

      I think when a poll is that ridiculous it’s better to just not count it as it’s throwing off the #s … I suppose then you get into judgment calls… but come on, even in Michigan? from the same pollster that said Romney was up by 15% in Florida? it’s 100% BS.

    • ChrisD

      Just click on VA in the Power of Your Vote table above and calculate the median of the last week’s polls. It was O+1.5 at 3pm and now it’s O+1 with the addition of JZ Analytics O+2 poll. Look, as new state polls appear and the one-week windows shift, those states’ medians are going to bounce around quite a bit, and the MM will, too. It’s better just to look at Dr. Wang’s chart of the MM over time, e.g., daily. It’s been slowly rising into the O+1.5/O+1.8 range for a few days now.

  • Reason

    Oh and did he remove NC? Is that going to R?

    • Obama 2012

      PPP just came out with an _even_ poll in NC! no way anyone should be calling that state for Romney.

    • ChrisD

      Hmm. I have no idea why NC disappeared. With the new PPP poll, it’s either tied or O+3, depending on whether the Rasmussen R+6 poll is within the one-week window or not.

    • Froggy

      Chris, NC should be tied. The Ras R+6 stays, since the model uses a minimum of three polls, no matter how stale they are. Although the PPP NC poll was up on Pollster in time for the 5pm update, it apparently hasn’t made it yet to the datastream that the model uses. I’ve seen this happen before a few times. It gives us something to look forward to tomorrow.

    • ChrisD

      Thanks again, Froggy. I’m still learning about Dr. Wang’s methods. I’m surprised the PPP NC poll wasn’t in the 8pm update. And Nevada’s O+3 looks incorrect to me. There are only four polls in the last week, from what I see: O+4,O+3,O+2,O+2. Their median is O+2.5.

      http://elections.huffingtonpost.com/pollster/2012-nevada-president-romney-vs-obama

    • Froggy

      Chris, the model uses average polling dates (dropping fractions). The most recent NV poll is the Marist 10/23-24, which the model counts as 10/23. This means that the Mellman poll (10/15-17, O+8) is still being counted, since its average polling date is 10/16, and 10/16 is within a week of 10/23.

  • Martin

    Anybody know about Pollyvote.com? Are they legit? Do they have a good track record?

  • Les Honig

    Guys…how do you think this will affect the race and Sam’s analysis…new Marist Poll showing Colorado tied and O with a 3 point edge in Nevada. Story makes a big deal that in Sept he held a 5 point lead in CO…Will this feel into Chucky T’s
    “Romentum” mime?

    Here’s the story link which has the link to the stats:

    http://firstread.nbcnews.com/_news/2012/10/25/14701395-polls-obama-romney-tied-in-co-incumbent-has-narrow-nev-edge?lite#__utma=238145375.171711838.1344390121.1351154286.1351204409.5&__utmb=238145375.1.10.1351204409&__utmc=238145375&__utmx=-&__utmz=238145375.1351204409.5.4.utmcsr=nbcnews.com|utmccn=(referral)|utmcmd=referral|utmcct=/&__utmv=238145375.|8=Earned%20By=msnbc%7Cpolitics%7Cpolitics=1^12=Landing%20Content=Mixed=1^13=Landing%20Hostname=www.msnbc.msn.com=1^30=Visit%20Type%20to%20Content=Internal%20to%20Mixed=1&__utmk=89932248

    • Ohio Voter

      I am not concerned about the single Marist poll, no. Romney of course, gained, this past month but the current trend is unequivocally the President’s.

      One poll does not a Romentum make. For that single tied poll, there were two more polls that show an Obama lead.

  • A New Jersey Farmer

    Hello All:

    Lovely day when the mm goes your way, yes? From what I hear and read about the Obama ground game, they are running rings around Mitt in NV, OH, and VA. And this after the GOP was supposed to have caught up to the Democrats since 2008. Looks like the move back to Obama is happening right on schedule. Every day that the mm moves higher is a day that Romney has to work extra hard just to stay in place.

  • Matt McIrvin

    Steve Erickson says his fellow liberals who don’t believe in Ro-mentum are just in denial:

    http://prospect.org/article/news-flash-was-always-close-election

    Sigh.

    • Ms. Jay Sheckley

      Thanks, Matt, for the heads up. Steve is a brilliant novelist we got to know because our bookstore very much admires his work. Politically, all he lacks is this website. Working on it.

  • No Tribe

    I was thinking of sending my kid to Princeton, but maybe not now, if they are churning out kids that come up with predictions like this site… well, you’re not as bad as the votamatic crew with the EV skew, but that first graph makes me chuckle.

    OK though, I’ll stick around and see you guys through the election returns next week. It’ll be interesting to see how you deal with it.

    • Sam Wang

      I do not think you will be returning, but I hope you do.

      At a good university, teachers will present factual information that sometimes challenges prior assumptions. The analysis here is based entirely on data. I hope you find a suitable place for your child’s education.

    • Richard Vance

      Dear No Tribe, I’m sorry you can’t understand what Sam is doing. The numbers do the talking. I hope you can send your child to Princeton it will be good for both of you.

  • Jim Hearst

    Dr Wang, if there is one thing we have learned from this comment thread, it is that the temptation to discuss individual polls is simply too great for mere mortals to resist.

  • NC Obama Guy

    RCP is so pissing me off. They are so manipulative. Was there not a PPP poll in VA O+5. They only show PPP polls in states where it helps their Romentum narrative. Don’t tell me they don’t use PPP polls because they do in states where it makes their case.

    • Froggy

      RCP doesn’t list anyone’s polls when they’re commissioned by a candidate or an interest group. As far as I can see they’re applying this rule equally across the board. I agree that it’s weird when it’s applied to PPP polls, though. Same pollster, apparently the same methodology, but some of the polls count and others don’t.

  • ChrisD

    I’m guessing the MM will drop in the 8pm update. With the flurry of polls showing up on Pollster.com early this evening, it looks like VA goes from O+1.0 to R+0.5 and NV from O+3 to O+2.5. A lot of the new state polls today seem (to me) to be Dem-leaning.

    Meanwhile, I still don’t know what’s going on with NC. Its median should be based on only 2 or 3 polls: tie, O+3, R+6 — by PPP, Grove Insight (D), and Rasmussen. That’s a small and disparate group of pollsters.

    • NC Obama Guy

      Doesn’t VA go to tie or O+.5 because the old Rasmussen falls off?

    • NC Obama Guy

      I added up OH. There are 10 polls. High is O+5 Low is Tie. Remove the high and low. The average of the others (8 middle values is 1.75). Why is OH at 1%?

    • Froggy

      NC Obama Guy, we’re going for the median here.

      For OH, there are 11 polls within a week of the average polling dated of the latest poll, from the 10/17 Ras to the 10/23 Ras.

      Line them up from highest to lowest:
      O+5, O+5, O+3, O+3, O+2, O+1, O+1, Tie, Tie, Tie, Tie

      Remove the five highest and the five lowest, and you’re left with O+1.

    • ChrisD

      NC Obama Guy: You keep throwing out successive highs and lows until you get down to either a single poll that’s the median or two polls, which you then average.

  • NC Obama Guy

    Someone reassure me please. It seem national polls have stopped swinging Os way. I know we only worry about states here but I am very concerned about the media narrative…

    • MarkS

      Chill dude! Go find some interviews with “undecided” voters. (If you can’t find any, the Saturday Night Live parody is surely available, and is shockingly accurate.) There is absolutely no telling what these people “decide” on. Media narrative is as irrelevant as everything else.

      And, DON’T “add up” Ohio. You’re doing it wrong. Let Sam do it for you. He knows what he’s doing.

    • Dean

      I’m sorry to break this to you, NC, but the national trackers show a massive Romney surge today. He gained 0.35 since yesterday. News reports are coming in now stating that the lines of Romney voters in Ohio are so long that they’re stretching all the way to West Virginia, which is Ohio’s route to the sea ;)

    • NC Obama Guy

      Ha Ha I get it…

  • Reason

    You mean the ABC poll? That is all likely voters only. Also, it has less D in their sample. I want to know what happened with Va as it was +1 and the new PPP is o+5?

  • Ms. Jay Sheckley

    Agree with those who’d rather see North Carolina [pink, last seen at R+3] listed in Power of Your Vote than, say, Michigan at O+6.
    Actually, I’d rather see more states listed. Plenty of room in the margin.

    Please at least list all states which aren’t the darkest hues.

    Who’s with me? [As if we could vote on this!]

  • Reason

    Can someone please answer my questions about Va and NH?

    • ChrisD

      VA has had 8 polls in the last week: O+5,O+3,O+2,O+1,R+2,R+2,R+2,R+3. Median (O+1,R+2) = R+0.5.

      What about NH?

  • Sam Champion

    I vote for adding NC.

    I just heard on Hardball that Purple Strategies, a polling firm, will be releasing a “bunch” of swing state polls tomorrow — all with Obama up.

  • Reason

    Chris, thanks. That is frustrating. I saw NH went from O to R. So I wondered what polls affected that. And I still cannot believe we actually count the House of Ras.

    • ChrisD

      NH didn’t change from 5pm to 8pm. It’s still R+1. I’ve started downloading the Power table into Excel so I can see which states’ medians change with each update. (I’m a 68-y.o. retired actuary with lots of time on my hands.)

  • Reason

    Sam, did they say they are good for O? Purple strategies?

    • Sam Champion

      Steve McMahon of Purple Strategies simply said that Obama was up in all the swing state polls they would releasing implying that this was good for Obama.

  • Reason

    I am trying to grasp how you worked that Chris. Is there an example I can use? Because I added the average and divided and it came up O+ 0.5. Thanks.

    • Reason

      Sam, thanks.

    • ChrisD

      Just picture 1 unit to the left of zero (in Obamaland) and 2 units to the right of zero (Romneyville). Where’s the midpoint? A half-unit to the right of zero, in Romneyville. Hence R+O.5.

    • ChrisD

      Make that R+0.5. (Got my O and 0 mixed up.)

  • Reason

    Chris, I get that, but the 4 polls for O and the 4 for R are not even or even higher for R.

    • ChrisD

      Doesn’t matter. Take the two *middle* polls, O+1 and R+2 and average them. For example, the median of O+10,O+10,O+1,R+2,R+2,R+2 is also R+0.5.

  • Reason

    And Sam, those Purple polls get released tomorrow?

    • Sam Champion

      Yes, according to Steve McMahon, a principal there, the Purple Strategy polls will be released tomorrow.

    • Olav Grinde

      The narrative is changing.
      Reputable pollsters are releasing new polls from swing states.
      And the Meta-Margin is back up again.

      A Blue Friday, indeed!

    • Brian

      Funny, this was not how he was talking on air about an hour ago.

  • Show Your Work

    Interesting story today about Paul Ryan spending millions on tv ads to defend his House seat in Wisconsin. Maybe he’s a little better at reading the tea leaves than the top of the ticket is.

    At this point the pretend “we’re ahead” narrative may be more about trying to ward off down-ticket affects.

  • A New Jersey Farmer

    We still won’t see the effects if the third debate until at least Monday. That Obama is seeing better results now is good news.

  • Ralph Reinhold

    It is a big mistake not to include 3rd party candidates when the election is this close. Only a handful of pulls show them. In some states, Johnson is polling 2%, mostly from the ‘undecided’ which seems to really be torn between ‘non-of-the-above’ and ‘you didn’t list my choice’ this year. In other states, Johnson pulls 5 or even 9%, mostly coming from Romney. What about these states where there is 2% margin. Is Johnson pulling from undecided, Obama or Romney? I think Sam is generous when he says the effect of the undecided could improve Romney as much as 0.3%. I think it is more like zero. Either they won’t vote or they’ll vote third party with about 1% breaking for either of the two main candidates.

  • Soothsayer123

    From 297 to 291 EV in a matter of hours? What happened?

  • Obama 2012

    bit surprised the MM didn’t go up today; I saw a whole lot of really good Obama state polls today. I must have missed some bad ones.

    • ChrisD

      I think the MM went from 1.40 to 1.82 and back down to 1.50 with some intermediate steps as new polls were added thoughout the day.

  • NC Obama Guy

    Thanks for explaining. I’m a physician not a math guy. Oh well…

  • Sam Champion

    The Holy Trinity – OH, NV and WI:

    Since the Denver debate, Mr. Obama has held the lead in 16 Ohio polls against 6 for Mr. Romney. In Nevada, Mr. Obama has had the lead in 11 polls, to Mr. Romney’s 1. Mr. Obama has led in all polls of Wisconsin since the Denver debate, and he has had five poll leads in Iowa to one for Mr. Romney.

    • Reason

      Yeah. Now Watch. Ras will move OH to a Romney +1 or +2.

    • Mike

      “Yeah. Now Watch. Ras will move OH to a Romney +1 or +2.”

      Next week. It’s pre-ordained. I expected a tie this week, and we got it. Next week, it will be a sliver of a lead. Bank on it!

    • ChrisD

      Ras conducted OH polls on 10/17 and 10/23 that are up on Pollster. They shouldn’t have another one until 10/30. Gravis’s last poll was 10/18-19. They might show up. Or ARG, Pulse Opinion Research/Let Freedom Ring (R), WeAskAmerica, and Wenzel Strategies (R-Citizens United). They’re all Repub pollsters who did OH polls earlier this month.