Obama’s “Fuzzy Math” in Ohio Early Voting

Regular source for tons of great campaign information, Adrian Gray, has a column today in Politico demolishing the Obama campaign memo claiming early voting strength in Ohio.  I am no big fan of tracking early voting because the numbers are incredibly opaque so drawing substantive inferences is fraught with guesswork and assumptions easily challengeable by anyone with an opposing view.  The Nevada situation is a special case for the Battlegroundwatch.com blog which is why I blog that state so extensively and leave the other states to people like Gray and Larry Schweikart.

In today’s column, Gray outlines the egregious mischaracterizations and misrepresentations in the campaign memo as only a veteran campaign operative would know:

At this point in an election cycle, many campaign staffers are busy fighting the press on what they call “process stories.” The candidates and their staffs want to talk about their plans and policies while reporters covering them find their audiences demand a play-by-play of the horse race. The result is constant overstuffing of campaign metrics and polling that only serve to muddy the waters for most political observers. In a close race, such as we have today, there is often plenty of data for both sides to use to their favor. One poll says this, another says that.

Obama memo

This makes it especially surprising to see the piece put out by President Barack Obama’s field director this week on early voting in Ohio. When things are ugly for a campaign, these types of memos can start flying. It is troubling for the president’s supporters that they could not come up with at least a handful of positive data points in Ohio. I worked as director of strategy at the Republican National Committee during the difficult 2006 election cycle — I know firsthand how hard it it is to come up with positive data in a negative cycle.

The takedown

There are normally three signs you know a campaign metrics memo is purely spin.

1. Anecdotes: “We have seen groups as big as 100 voters going to vote in Athens, Ohio.” Only 604 democrats have voted in person in the entire county and no more than 40 in a single precinct (that would be Athens 3-5, for those scoring at home).

2. Unverifiable Data: “Precincts that Obama won in 2008 are voting early at a higher rate”: This is unverifiable and misleading because there is no such thing as an “Obama precinct.” Every ten years, the entire country rebalances its voting districts based on a constitutionally mandated census. In 2010, this process redrew the lines of reportable voting areas that were used in 2008. So this year, we have entirely new precincts, thereby making it impossible to validate their claim.

3. Cherry-picking random sub-poll data: “Time poll shows the President up 60-30” among early voters. That sub-sample was asked of 145 people and was one of many of similar ilk (with a huge variation in results). Their central data argument is that 43 more people told Time’s pollster over a two-day window they supported Obama. If that is their best claim to a lead in Ohio, it is a troubling picture for the president.

The reality of Ohio early voting

I have always been a believer in data telling me the full story. Truth is, nobody knows what will happen on Election Day. But here is what we do know: 220,000 fewer Democrats have voted early in Ohio compared with 2008. And 30,000 more Republicans have cast their ballots compared with four years ago. That is a 250,000-vote net increase for a state Obama won by 260,000 votes in 2008.


  1. PeterJ
    Posted October 26, 2012 at 4:34 pm | Permalink | Reply

    Not OH related, but RCP just moved CO to the Obama column in the no tossups simulation. This would be a problem because with CO, all Romney needed was 1 of OH/PA/MI, or a 2 state combo of either WI/MN + 1 of IA/NV/NH. All assuming Romney takes FL of course.

    • EpiphoneKnight
      Posted October 26, 2012 at 4:40 pm | Permalink | Reply

      Mitt will win Colorado. Pretty much for sure.

      • Brad
        Posted October 26, 2012 at 5:01 pm | Permalink

        Early voting has CO R+3, so yeah, not buying RCPs move.

    • fab4gal
      Posted October 26, 2012 at 4:40 pm | Permalink | Reply

      I don’t see how that’s possible since here in CO there is absolutely no Obama support, and I read a report just yesterday that already more Republicans than Democrats have voted early here. Rasmussen has Colorado 46% Obama and 50% Romney and from what I’ve read on this site, Rasmussen uses a model that uses too high of a democrat turnout.

      • MikeN
        Posted October 27, 2012 at 12:33 am | Permalink

        I won’t believe a Colorado or Nevada win until the votes are in. Democrats overperformed polls in both states in 2010. I think I’ll give the benefit of the doubt to the tally of who got whom to the polls to vote.

    • Tom
      Posted October 26, 2012 at 4:43 pm | Permalink | Reply

      Don’t worry about the rcp average. It won’t favor Romney. Think of it this way. When you have 10 polls, and all of them over sample dems, or over include minorities, or many other reasons aren’t correct .. you get a bad average. Romney is winning big in CO. Mitt-Mentum.

      • jeff
        Posted October 26, 2012 at 5:59 pm | Permalink

        Rcp has become a fraud this election cycle. They would include a poll conducted by a 5 year old kid if it in any way helped Obams.

    • Porchlight
      Posted October 26, 2012 at 5:51 pm | Permalink | Reply

      I really don’t understand RCP sometimes. They moved NC from leans Romney to tossup a couple of days ago. Yeah, sure.

    • Posted October 26, 2012 at 8:06 pm | Permalink | Reply

      RCP is arbitrary.

      I think <5% is Tossup, so one dud poll can shift the states in and out of Tossups and Leans, like PPPs did to NC.

    • Posted October 26, 2012 at 10:35 pm | Permalink | Reply

      2.25M people voted in Colorado last time. 600,000 have already voted early this time. According to the Colorado Sec of State, Repubs have returned 19,000 more ballots than Dems. See, http://coloradopeakpolitics.com/
      Given Romney’s lead among Colorado independents, he is ahead by about 5%, with a quarter of the total vote counted. Furthermore, his lead has grown in absolute and relative terms for each of the first four days. Don’t count on a late Dem surge, their numbers weaken in every state as you get closer to election day. Colorado is over. Most people have not figured this out and are still watching polls, but one group of skiled political operatives (Obama’s campaign) knows it all too well and that is why they are acting so crazy.

  2. Henry Reader
    Posted October 26, 2012 at 4:36 pm | Permalink | Reply

    CNN releases hit piece just in time for the weekend. O 50% R 46%. By double digits white men support Romney. White women support Romney by 6. I have to believe non-whites and women were oversampled to give drones false hope.

  3. Tom
    Posted October 26, 2012 at 4:42 pm | Permalink | Reply

    There are so many pieces of evidence pointing to obama losing and Romney winning. Since the left wing media is reporting it, we don’t get it given to us clearly, we just need to piece the pieces together like a detective on a case and then believe the truth. Romney is winning. Mitt-Mentum.

    • fab4gal
      Posted October 26, 2012 at 4:44 pm | Permalink | Reply

      Yep. These two sentences alone prove Mitt will win OH: “220,000 fewer Democrats have voted early in Ohio compared with 2008. And 30,000 more Republicans have cast their ballots compared with four years ago.”
      Game over for Obama.

  4. Michael
    Posted October 26, 2012 at 4:46 pm | Permalink | Reply

    Keith, the math is:

    Romney winning NON-early voters by 7 points, 51-44.

    All Obamas lead comes from early voters, where he’s up 59-38.

    For the math to yield a 50-46 lead, CNN is saying that at least 37% of the electorate (and as much as 42%) has already voted.

    Two things we know to be true:

    Per GMU, the early vote is at 18.1% of the total 2008 electorate (which means its likely even less as a % of 2012 since turnout will be up)

    Second, the Democrats have about a 6.4 point lead in early voting vs a 20.3 point lead in 2008.

    • Brad
      Posted October 26, 2012 at 5:05 pm | Permalink | Reply

      And beyond the -14 for Dem early vote, Indies are pulling for Romney this cycle by roughly the same amount they went for O, +8.

  5. Adam
    Posted October 26, 2012 at 4:52 pm | Permalink | Reply

    Great writeup Keith. The good is news they’re side has to spin, spin, spin to keep up the optimism. However, we have to be a little cautious, if just because many Obama votes won’t be logged as Democrats this year, just due to the primary cycle. Some of Obama’s early-voting lead is undoubtedly buried in the unaffiliated early vote.

  6. bman77
    Posted October 26, 2012 at 4:52 pm | Permalink | Reply

    Didn’t McCain actually win folks that voted on election day in Ohio back in 2008? Add that to this improvement with early voting and things are hopefully looking good.

    • fab4gal
      Posted October 26, 2012 at 4:54 pm | Permalink | Reply

      Yes, McCain won on election day by 100,000 votes. The only reason Obama won Ohio in 2008 was the huge number of early votes for him.

  7. Medicine Man
    Posted October 26, 2012 at 4:56 pm | Permalink | Reply

    Get use to these shock polls. Will be a slew a week from today along with Monday night.

  8. Posted October 26, 2012 at 4:57 pm | Permalink | Reply

    by my fuzzy math…2008 Obama won the whole state by 260,000 total votes. In 2008, by the above numbers obama started election day 220,000 votes ahead of McCain and finished the day 260,000 votes up?

    So am I areading this right that if the election started right now…Obama’s election day lead would only be 10,000 votes? Or am i way off in reading this?

  9. Matt
    Posted October 26, 2012 at 4:57 pm | Permalink | Reply

    How the heck is Obama leading with independents in the new CNN poll? Did they over sample those who already voted? If not, this doesn’t look good to me.

    • Posted October 26, 2012 at 5:08 pm | Permalink | Reply

      Yes they did, apparently.

      40% early voters vs <20% actually voted.

    • Medicine Man
      Posted October 26, 2012 at 5:11 pm | Permalink | Reply

      It’s what I’m saying. Either 1) BO has new messaging that appeals to Indies and is making the difference and we will need it to be verified by a good reputable pollster 2). This data stinks and is more of marketing piece because the criticism of the polls coming out of Ohio are two fold ( over sampling of Dems and Romney is winning Indies). This poll takes care both criticisms AND Obama hits 50 also…perfect!

      • Medicine Man
        Posted October 26, 2012 at 5:14 pm | Permalink

        Funny is that it came out a day after the GOP pressed regarding the state of the race….

  10. zang
    Posted October 26, 2012 at 5:10 pm | Permalink | Reply

    The problem with RCP, as everyone knows, is it as a garbage in, garbage out, situation. The idea is that somehow, everything will equal out. In the past, there were Republican pollsters (like Strategic Vision) who would issue polls painting a rosy picture for Republicans. These polls would counteract PPP or polls from some obscure community college showing Obama up by 15 points in Kentucky. There is no such counter force this cycle.

    • PeterJ
      Posted October 26, 2012 at 5:17 pm | Permalink | Reply

      Part of justification from 538’s supporters for that model is in fact that the RCP averages are “crude”, although of course Nate’s subjective weightings are D skewed. But the good points of the simple averaging that RCP uses are in fact no subjective weightings (other than an occasional subjective removal of an outlier), and the fact that they use only the most recent polls from each firm instead of the laughable case of 538 where Nate weighted an earlier PPP poll higher than a later one because he didn’t like the later one’s result (moving toward Romney). But it would not require a whole lot more fiddling for RCP to weight the average by sample size and MOE, while still relying only on the most recent polls of each organization.

  11. Ranger375
    Posted October 26, 2012 at 5:12 pm | Permalink | Reply

    “Bandwagon” tactics.

  12. fab4gal
    Posted October 26, 2012 at 5:13 pm | Permalink | Reply

    I am trying to understand the psychology behind this “Everything is fine” attitude from the Obama campaign. Since enthusiasm is already really low for Dems this election year, isn’t telling them everything is fine just more reason for them not to bother voting? I would think a “Oh no we’re losing, you have to vote NOW to help Obama win!” tactic would be more effective….but then I think, no one likes to back a loser, so that might make less Dems vote. Anyone have any ideas?

    • Tom
      Posted October 26, 2012 at 5:21 pm | Permalink | Reply

      A campaign can never admit they are losing. It disheartens everyone, particularly now-a-days with all the early voting. So it’s like the Animal House scene from the losing campaign where he says “remain calm, all is well”

    • Brad
      Posted October 26, 2012 at 5:22 pm | Permalink | Reply

      The “eveything is fine” play is all he has left. If the “inevitable meme” is shattered he has nothing to run on as his record is in shambles and his rhetoric is out of step with Indies. He has to depress R turn out to win.

    • PeterJ
      Posted October 26, 2012 at 5:23 pm | Permalink | Reply

      Assuming they really do believe everything is fine instead of just putting on a show for public display, I think they believe that Mitt cannot take CO, FL and OH, or CO, FL and a combo of MN/WI + 1 other smaller battleground state. Or that he has a possibility of actually taking PA or MI. And to some degree history justifies those assumptions because many of those states are for repubs the analogy one commenter made some while back of the Peanuts theme of Lucy holding the football for Charlie Brown. WI, IA, NH and OR in particular if memory serves, have a history of often coming tantalizingly close but not delivering for us.

      • Posted October 26, 2012 at 5:34 pm | Permalink

        I believe they think their Ohio firewall will hold. That there is no way Romney picks off NH, CO, IA, WI, or NV to offset his loss of Ohio.

        And they based on 2008 believe that their ground game is so far supperior to anything Romney even remotely believes he has. They think that on election day they will swing into action and get hoards of democrats at or exceeding 08 levels to the polls, period end of story for them. Democrats were bashed for a decade that the Rove Ground Game so outclassed them…Obama came along and his ground game out foxed Republicans in 08 and they truly believe that firewall will hold as well. i truly believe they believe that their numbers will exceed 2008 on their ground game and micro-targeted plans alone.

        Obama and has team have a deep rooted HATRED of Mitt Romney, they have no respect for him at all. He at least repsect McCain. There is zeo there for Romney. I think part of boils down to the old SNL skit with Jon Lovitz as Dukakis where he turns looks at the camera and says “i can’t believe I’m losing to this guy” they cann fathom losing to this guy they hate so much.

      • Dave
        Posted October 26, 2012 at 6:36 pm | Permalink

        “they will swing into action and get hoards of democrats at or exceeding 08 levels to the polls, period end of story for them”

        They could. Don’t underestimate that possibility. The establishment has voted for O. Disney, comcast, time warner, ge, and viacom (to a lesser extent) and others all have been staunchly pro Obama supporters throughout the year/summer and this election. Banks like Goldman and other wall street elite (JP, Goldman, BOFA, USB, etc. — do you remember all those earnigns where they bragged about zero days trading losses — biggest cronies market in history — nobody has zero days trading loses unless you are the market or are in touch with the guy who is the market – Bernanke/Geithner)…all benefitted. The Fed timed QE infinity just right for Obama. They could have stopped at QE1 they didn’t they kept going and now have committed to infinity. In fact, it will probably become regular fed policy — the spending — any time the market threatends to go down. That’s great you say, my 401K will do wonders. Well, therer’s going to be a bill for that. The question is how is it paid for? If the spending stops, the market likely corrects back to whre it should be a few hundered SP points lower until a real recovery happens, which will happen at some point, but the stocks of corporations go down. If spending continues then likely go on to new highs and corporate stock and stock options of executives skyrocket (and 401Ks do OK too but those are just scraps comparatively) but there’s a bill for that. Who pays for it? And how? If you’re a dem you want an excuse to raise marginal rates for everyone while increasing entitlements (that keeps you in power and that’s how they’ve done it in CA). If you’re a repub, you cut taxes and cap deductions and limit spending (yes, the math works). This is a much more palatable option for those who make a decent amount. It’s hard to know why there is such a push to keep O in office but I think most will agree there has been one. With all the largeese he has given corporations through the stock market, and with the threat of that stoping under Romney, I would imagine they’re wanting to keep him in office — keep spending, keep jacking the market, and spead the tax burden to everyone in the form of higher marginal rates. With so much $$$ at stake I wouldn’t put it past them if 100,000 dead voters showed up in early voting or some shenanigans to push him over. I am skeptical they’ll let him lose. I hope that concern is unfounded.

    • allthingsgeography1
      Posted October 26, 2012 at 5:30 pm | Permalink | Reply

      As an Obama guy, I haven’t gotta freakin’ clue. I wish he would just say, “Hey, Mitt’s got momentum, we’re falling behind, let’s GOTV and stop him in his tracks!” But he’s not. I started to get clues not everything was ok just from his performance in debates and the increasing antagonistic views in polling by people towards him on his management of the economy. Hence, why I started looking for other views outside the news and left-wing talkers. Now I’m realizing how bad things are and they seem to match up well with Obama’s behavior and demeanor in debates and the campaign. He knows he’s in trouble but doesn’t want to admit it I think. I think he’s more afraid of the panic as he’s built this strong aura around him from 2008. I re-watched his acceptance speech the other day on YouTube and all of the nostalgia came back. I’m afraid he thinks he’s built an aura of invincibility and dreams so strong, that to admit it’s all crumbling would be crushing. And sadly, it kinda is. So much for Hope and Change.

      • Posted October 26, 2012 at 5:39 pm | Permalink

        to expand on what i said above…I think he still sees himself as the IPhone/IPad it is the cool, it product still several years later. There have been lots of challengers, lots of supposed next big things and they have all fallen before him….he will succeed.

        This is where his inexperience in politics comes in…the attention span of the country is as long as a gnat’s ass. All these young people, new voters TRULY TRULY TRULY believed he would sweep in make Washington perfect, the country would be humming, wars woudl end, peace would reign, global warming would stop, he and the US would be looked upon as saviors of the world. 4 years later in the eyes of most of them he is just another damned lying politicians who failed them by promising the world and delivering them the bill. He is not cool anymore to anyone other than the die hards and hollywood. If you watch TV or movies or read magazine you would think it was still the summer of 08 and Obama was the IPAD wrapped in a Furby with a couple Beany Babies tucked under it’s arms…..the facts are American doesn’t care. They have an American Idol mentality…NEXT

      • fab4gal
        Posted October 26, 2012 at 5:43 pm | Permalink

        allthingsgeography1, I have seen you posting on here and now that I’m responding to you directly, I have to tell you how much I admire you and am glad you are here. I don’t think it’s a secret that most people on here are for Romney and your logical perspective and looking at the pure facts as they are is extremely admirable.
        I’m glad you’re as confused by this as I am, because if it were me I’d do the “we need to GOTV!” thing like you said, not just act like everything is fine. You bring up a good point, the arrogance of the entire campaign is probably their downfall. Us on the right have made jokes about “The Annointed One” because the main stream media has acted like he walks on water for the last 4 years and every TV show he goes on kisses his butt big-time. I think he was in a bubble for 4 years where everyone thought he was the most amazing thing and now he truly can’t grasp that he can lose after so many years of that.

      • PeterJ
        Posted October 26, 2012 at 5:45 pm | Permalink

        The problem for O supporters like yourself is that even if he wins and Dems retain control of the Senate, the House will still be republican and in the senate still be able to filibuster without 60+ democrats. So no repeat of the first two years of Obama getting his way in his first term. All that is left then for the next two years is more playing chicken with the House on fiscal cliffs and hoping he can bully them into raising taxes, which he won’t be able to do. So then you have to hope he can convince voters in a mid-term election to give him solid control of both houses so that he can do something in his normally lame duck last two years. The only thing you can really hope for is another Supreme appointment or two. Ironically democrats might be better off if Mitt wins and doesn’t deliver and they take both houses of congress in 2014 and the presidency back in 2016.

      • fab4gal
        Posted October 26, 2012 at 5:46 pm | Permalink

        Shane, I think you are right, I said it in another post but Obama was “trendy” in 2008. It was the cool thing to do to vote for him. I was 22 in the year 2008 and you would’ve had to put a gun to my head to get me to vote for a socialist, but I knew full well my peers were going to, and honestly, I knew he was going to win that year. I even attended an art school at the time so you can imagine what that was like being a Republican during an election year at not only a school but an art school…ugh.

      • fab4gal
        Posted October 26, 2012 at 5:49 pm | Permalink

        Peter, I see your point, but Obama is the most partisan President I have ever even heard of. Republicans were literally locked out of discussions on healthcare, that has never happened before in history, to completely lock another party out. Something tells me Mitt won’t let things like that happen on *either* side. Even if Dems have control of the Senate, I am confident Mitt will make them all play nice.

      • fab4gal
        Posted October 26, 2012 at 5:52 pm | Permalink

        I realize my post above made it sound like I went with the trend, in fact I voted for McCain in 2008 (but it was really a vote against Obama).

      • beach91
        Posted October 26, 2012 at 5:57 pm | Permalink

        allthings I cannot understand your support for obama in the very least. Why in the world do you still support him? The guy’s ideas are marxist economically speaking and they do not work. We have yhe last 4 years to provide the evidence of his failed ideas and policies…

      • PeterJ
        Posted October 26, 2012 at 5:58 pm | Permalink


        With the level of partisanship in the last couple decades it is by no means sure that Mitt can reach out if the Dems in the senate bully the caucus into conformity and they insist on tax raises as the price of bi-partisanship. 2-party Parliamentary democracies with their total effective control (upper house cannot truly impede) of government, get leeway to do what they want mostly and take all the credit or all the blame. Our divided government via ticket-splitting allows both sides to claim the other is blocking progress with the status quo mostly remaining unchanged on so many issues. Even on a state level, I think it would be better if the repubs could not block monster tax hikes in Cali and they went off the cliff sooner and thus sooner got back on the road to sanity.

      • Dave
        Posted October 26, 2012 at 6:50 pm | Permalink

        I’m splitting my ticket so we get a divided government. I would like to see a divded government. Whenever one side takes over completely the extremists come out in droves and we go flying off the deep end with constitutional this and tax and spend on that.

      • Porchlight
        Posted October 26, 2012 at 7:16 pm | Permalink

        I like the concept of divided government, except in this case…Obama likes to do end runs around Congress, and…Obamacare, and….SCOTUS. Mostly SCOTUS.

      • allthingsgeography1
        Posted October 26, 2012 at 8:34 pm | Permalink

        beach91, I have my reasons for supporting Obama and not supporting Romney that are much too numerous to this blog. Everything from Romney’s economic policies, the nature of his positions overall as well as social issues. Hence why I said on another posting, me voting Republican for President is a pretty tough sell. Here, I mostly stick to the numbers and data as arguing over politics would be…well, endless and pointless. Not saying you wanted to argue, but that’s why I don’t throw my politics out there.

      • Dave
        Posted October 26, 2012 at 10:11 pm | Permalink

        “Even on a state level, I think it would be better if the repubs could not block monster tax hikes in Cali and they went off the cliff sooner and thus sooner got back on the road to sanity.”

        Cali will not go off the cliff until every last voter is drained through higher taxes. You’re seeing them attempt that now. You don’t want to let it get that far. So called temporary tax increases have a way of becoming permanent.

  13. PeterJ
    Posted October 26, 2012 at 5:29 pm | Permalink | Reply

    A nice Red State article on the Swingometer concerning party ID this cycle:


    • Dave Ped
      Posted October 26, 2012 at 7:00 pm | Permalink | Reply

      I have been looking at Rasmussen party ID since early summer and also the swingometer. I totally agree that this is what will play out on election day. What is interesting about that analysis in red-state is that it does not really consider the equally huge swing from O to R in the independents. That will only add to this IMO. Plus you have the entheusiasm gap too.

  14. Blackcloud
    Posted October 26, 2012 at 5:34 pm | Permalink | Reply

    Josh Jordan (@numbersmuncher on Twitter) points out that of the last 17 polls in Ohio, only the two CNN/ORC polls and the Rasmussen poll have Obama winning independents. In the fourteen other polls, Romney’s lead with independents is 11.4. There’s lots of screwy stuff going on with the polls, and in Ohio the early voting seems to be playing the most tricks on pollsters.

  15. hunter
    Posted October 26, 2012 at 5:38 pm | Permalink | Reply

    Obama’s road to reelection: RCP has him at 201; he takes PA (20), OH (18), MI (16), CO (9) and NV (6) for @270.
    Let’s hope you’re right about OH.

  16. Dave
    Posted October 26, 2012 at 5:40 pm | Permalink | Reply

    The CNN poll, do they release the breakdown (not that that matters, they can lie their a$$ of about that too). The only piece of info you can go by is what they give on their website is that for those voting early it’s 59-38 O and for those voting on election day it’s 51-44 R. The question is what % of the electorate is voting early? That number determines if R is in front or O. Say they sampled 750 people, what % of those people said they were voting early? If it’s 10%, O’s going to be behind, if it’s 50%, R is behind according to this poll.

    That said many of the comments on CNN (and I’m sure MSNBC) mirror the comments here about hope for their candidate. You guys try to break it down a little further but let’s be honest the bias is not insignificant here.

    Whatever ever happens the biggest winners will be the companies that manufacture prozac as half the natilon will be in dire need of some.

    • PeterJ
      Posted October 26, 2012 at 5:47 pm | Permalink | Reply

      More jobs outsourced overseas when folks buy the cheap omeprazole generic made somewhere else. 🙂

    • Brad
      Posted October 26, 2012 at 5:56 pm | Permalink | Reply

      @adriangray – CNN OH poll: Obama leads 59-38 among 1.4 million that voted early. Romney leads 51-44 among 4.4 million have yet to vote. You do the math.

      Ok, I will….I get R 48, O 47.

      • Brad
        Posted October 26, 2012 at 5:58 pm | Permalink

        ..and note, yet again O 47.

      • fab4gal
        Posted October 26, 2012 at 6:00 pm | Permalink

        But 1.4 million people have not voted early. From the state of Ohio itself, 1.6million have been mailed out, but only 800,000 have actually voted early: http://www.sos.state.oh.us/SOS/mediaCenter/2012/2012-10-23.aspx
        I realize this was released 3 days ago but I can’t imagine 600,000 people have suddenly stampeded the polls in 3 days.

  17. JGS
    Posted October 26, 2012 at 6:05 pm | Permalink | Reply

    OK, so the polls says that the sample includes 741 likely voters. So let’s do a little algebra. X = % who reported that have already voted, Y = % voting on election day (or who have not yet voted but are likely to vote, either early or on election day, hereafter).

    Obama’s overall 50% is comprised of .59X + .44Y.

    Romney’s overall 46% is comprised of .38X + .51Y.

    A classic simultaneous equation.

    .59X + .44Y = .50
    .38X + .51Y = .46

    .59X = .50 – .44Y, which means that X = 1.695(.50 – .44Y) = .8475 + .7458Y

    If X = (.8475 + .7458Y), then in the second equation, we get [.38 (.8475 + .7458Y)] + .51Y = .46, or .32205 + .283404Y = .46, or .283404Y = .13795. This means that Y = .13795 divided by .283404, or .4867609.

    If Y = .4867609, then X = 1 – .4867609, or .5132391.

    If all of this is correct, this means that CNN’s sample includes 51.32391% who reported that they have already voted, and thus assumes only 48.67609% who will be voting on election day.

    That cannot possibly be right, so let me test it again by back-filling my answers.

    If Obama leads by 21% among the early voters and the early voters are 51.32391% of the overall sample, that’s a lead for Obama of 10.778% based solely on these early voters.

    If Romney leads by 7% among those who have not yet voted and those comprise 48.67609% of the overall sample, that would give Romney a 3.40732% lead among this group.

    10.778% – 3.40732% would be an overall 7.37268% lead for Obama, not a 4% lead, so something is wrong here.

    If you assumed 40% early voters and 60% later voters, you’d get closer to the 4% overall margin:

    .40 X 21 point Obama lead among that group = 8.4 lead on account of that 40%
    .60 X 7 point Romney lead among that group = 4.2 Romney lead on account of that 60%
    These two net to 4.2% which is pretty close to the reported 4% margin between them. So it must be about a 40/60 split which is way too high.

    • AJ
      Posted October 26, 2012 at 6:25 pm | Permalink | Reply

      omg my head hurts

    • Michael
      Posted October 26, 2012 at 6:42 pm | Permalink | Reply


      I posted the math in a comment above. The range is 37%-42% early voters. Plug in anything in that range and the rounded numbers come out to 50-46.

      • Evan3457
        Posted October 26, 2012 at 9:55 pm | Permalink

        40% is the middle of the range in my quick calculation. No way 40% have already voted, however, the CNN poll doesn’t say “already voted”. It says those who’ve voted early or are going to vote absentee.

    • Dave
      Posted October 26, 2012 at 6:43 pm | Permalink | Reply

      JGS, thanks. That’s what I wondered. It didn’t seem right.

  18. Pete
    Posted October 26, 2012 at 6:54 pm | Permalink | Reply

    Lying liars lying

  19. zang
    Posted October 26, 2012 at 7:00 pm | Permalink | Reply

    Here’s the big media’s take on those baffling MSM Ohio poll numbers:


    • fab4gal
      Posted October 26, 2012 at 7:17 pm | Permalink | Reply

      According to that article, Obama got 89% of the Democrat vote in 2008, but this year he will get 92-94% of the Democratic vote. With voter enthusiasm in the Democratic party way down, MORE Democrats are going to vote for him? I’m sorry, but I’m not buying it.

      • zang
        Posted October 26, 2012 at 7:21 pm | Permalink

        His vicious partisan attacks on Romney can gin up Dem enthusiasm, or at least that’s the aim. The tradeoff is you piss off swing voters. If there are really 9% more Dems in Ohio like the big media polls claim, that’s a good tradeoff.

  20. Brad
    Posted October 26, 2012 at 7:08 pm | Permalink | Reply

    11 days is an eternity in politics. I don’t think it’s time to panic about OH yet.

    • Medicine Man
      Posted October 26, 2012 at 7:38 pm | Permalink | Reply

      Whose panicking? That is exactly their game plan…or the only one left.
      I’m a realist. I think the chance in NV is tough with the numbers, so I’m not a dreamer, but the data doesn’t point for a reason to “panic”.

      If our hopes were pinned to Penn, Michigan..I would be panicking…

  21. NHConservative
    Posted October 26, 2012 at 7:23 pm | Permalink | Reply

    Didn’t Kerry have a 3-4 point lead in OH heading into Election Day?

    • JGS
      Posted October 26, 2012 at 7:33 pm | Permalink | Reply

      Not heading into Election Day, no — by then most polls had Bush with a small lead. But if you look 1-2 weeks before the election, a majority of polls did have Kerry in the lead. See:


      Colmbus Dispatch* | 10/20-29 2880 LV 2.0 50 50 – TIE
      Survey USA| 10/23-10/25 831 LV 3.5 47 50 – Kerry +3
      ARG | 10/23-10/25 600 LV 4.0 47 49 – Kerry +2
      LA Times | 10/22-10/26 585 LV 4.0 44 50 – Kerry +6
      Strategic Vision (R) | 10/22-24 801 LV 3.0 47 45 – Bush +2
      Rasmussen | 10/20-10/26 537 LV 4.0 50 46 – Bush +4
      Scripps | 10/17-10/21 358 LV 5.3 46 50 – Kerry +4
      Gallup | 10/17-10/20 706 LV 3.5 47 48 1 Kerry +1
      FOX News | 10/17-10/18 800 LV 3.5 49 44 – Bush +5
      SurveyUSA | 10/16-10/18 698 LV 3.8 47 49 – Kerry +2
      Mason-Dixon | 10/15-10/18 625 LV 4.0 46 45 – Bush +1
      ABC News | 10/14-10/17 789 LV 3.5 47 50 – Kerry +3
      Rasmussen | 10/12-10/18 537 LV 4.0 47 47 – TIE
      Ohio Poll | 10/11-10/17 757 LV 3.6 46 48 – Kerry +2

      • Porchlight
        Posted October 26, 2012 at 7:56 pm | Permalink

        Yes, it was very back and forth between Bush and Kerry in Oct 2004. Plus RCP was new on the scene and under the radar, so the idea of tossing the occasional skewed outlier into the mix to game the average (*cough* PPP *cough*) hadn’t occurred to anyone yet.

  22. Evan3457
    Posted October 26, 2012 at 11:16 pm | Permalink | Reply

    Posted this at the end of the Ohio Romney Rally Thread, thought I’d repeat it here:

    ARG poll in OH showing Obama up 49-47….D/R/I is 43/34/23 and Early voters are 28% with Obama leading 55-44, and Romney only up by 1 with Election Day voters, so it would appear both fudge factors (too many D’s; too man Early’s) are in play on this one.

    Purple Strategies poll in OH showing Obama up 46-44….D/R/I is 34/27/38, and early voters are 26% with Obama leading 58-32, so again, both fudge factors in play.

    Somebody, I say, somebody is a-lyin’, and come a week from Tuesday, we’re gonna find out who.

One Trackback

  1. […] Obama but so many votes banked away for Obama that it wasn’t enough. Today, according to the same Adrian Gray: “220,000 fewer Democrats have voted early in Ohio compared with 2008. And 30,000 more […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: