Patterico's Pontifications


The MSM Is Busy Filling In Their Electoral Maps Based On State-By-State Polling That, If Accurate, Means the National Tracking Polls Are Off Nearly Double

Filed under: 2008 Election — WLS @ 6:02 pm

[Posted by WLS Shipwrecked]

I continue to be fascinated by the issue of whether Obama will prevail by a large enough nationwide vote total to overcome the advantages of small states in the Electoral College, when most of those small states will be going for McCain — while all but one of the most populous states will go for Obama.

Today I’m trying to reconcile nationwide tracking polls that show a 4-5 point race, with the projections of a blowout in the electoral college, in light of the inherent advantage that small states have over large states.

I broke down the vote totals in each state being predicted by the latest polling posted at CNN, using the 2004 vote totals for each state increased by 10%. I’ve previously used projected increases as high as 25%, but I heard one analyst on CNN last night predict that the nationwide increase over 2004 will probably be around 10%.

The math comparing the state polls to the national tracking polls doesn’t add up – one or the other is way off. The national tracking polls are bouncing around anywhere from 3 to 7. Let’s call the nationwide result next Tuesday at 4.5%, and make the outcome of the election at 51.5 to 47.5, with 1% going to third parties.

At 51.5%, Obama would receive 68 million votes, and McCain would receive 62.7 million votes at 47.5%, for a overall win by Obama of 5.3 million votes.

I looked through all the state polls in all the states that are being used by CNN on their interactive electoral map, which can be found here.

I wanted to include all the math in this post, but WordPress doesn’t incorporate columns of numbers very well, and I can’t get it to line up. But based on current polling, here are some samples of what Obama’s winning margins would be in certain specific states:

CA — 3.1 million votes

IL — 1.4 million votes

MI — 1.2 million votes

MN — 600k votes

NJ — 900k votes

NY — 2.5 million votes

NC — 230k votes

PA — 1.2 million votes

McCain wins by more than a million votes in only one state — TX  by 1.5 million.

When you go state by state, estimate the vote total from 2004, and then apply the current poll, Obama would win the nationwide vote by 10.1 million votes. The percentages would be 53.8 to 46.1, a spread of 7.7%.

If you average the 6 nationwide tracking polls at Rasmussen, split the undecideds, and give 1% to 3rd parties, you end up at 52-47.  So the state-by-state polls being used by the MSM are saying OBAMA is going to OUTPERFORM his national polling by 2%, and McCain is going to under-perform by 1%.  That’s a net 3 point swing of nearly 5 million voters.

These state-by-state polls are being used by the MSM to fill in their electoral maps and declare Obama a likely landslide winner with 350+ EVs, taking FL, IN, NC, OH, and VA.

Don’t be fooled. It’s going to be 3-4 points on Sunday, and Obama is going to under-perform just as he has consistently.  Undecideds are going to vote better than even for McCain. If they haven’t been convinced to drink the Kool-Aid yet, given all that has happened with him since the Iowa caucuses, there is a reason why they haven’t.

For the first time since mid-September, Rasmussen had McCain and Obama even on the issue of the economy. McCain was down 9 points on that issue at the time of the debates. That is going to be reflected in the balloting only on election day.

— WLS Shipwrecked

45 Responses to “The MSM Is Busy Filling In Their Electoral Maps Based On State-By-State Polling That, If Accurate, Means the National Tracking Polls Are Off Nearly Double”

  1. Shipwrecked, the link to the state polls is missing.

    What I find interesting is how many polls that claim that Colorado is clearly Obama’s and yet that Obama keeps finding the need for coming back repeatedly to visit.

    SPQR (26be8b)

  2. The new CBS/NYT poll has Obama up by 11 points over McCain 52%-41%:

    Looks like a 4% gap on Sunday is a pipedream, the gap is widening, not shrinking.

    snuffles (677ec2)

  3. 13% of the poll is new voters. Yeah, like that’s going to be the way the real vote comes out.

    Chuck Simmins (e7938e)

  4. snuffles — when was the last time the NYT-CBS poll was accurate? Oct. 14 they had Obama up 14.

    Take a look at their history.

    WLS (26b1e5)

  5. The new CBS/NYT poll has Obama up by 11 points over McCain 52%-41%:

    Yeah, let’s all pay strict attention to those paragons of objectivity and accuracy. Your rampant idiocy and firm denial to ever using objective sources continue to amaze us all.

    Dmac (e30284)

  6. Tell me the turnout, I’ll tell u the winner.

    The historic levels of turnout, including the two most recent presidential cycles, show that McCain can count on a 2-4% bump from the legitimate weekend polling – Gallup traditional/LV among these.

    I still think BHO ekes it out, though.

    Ed (d7cda1)

  7. Funniest moment was when Gunga Dan declared the election for Kerry, even before the polls closed. Hilarious.

    Dmac (e30284)

  8. According to the CBS/NYT poll, 93% of CBS employees thought Dan Rather was not lying through his teeth about the bogus 2004 TANG “story,” and 89% of New York Times employees thought that fired reporter Jayson Blair was a righteous brother who spoke troof to power!

    Official Internet Data Office (184273)

  9. Thanks for the link Shipwrecked. Now look at Colorado, if indeed Obama has an 8 point lead as CNN claims, then why did he come here so much in the last week? Makes no sense.

    SPQR (26be8b)

  10. I see no way that McCain wins without PA. Going by the RCP map, McCain must win all “his” states (142) all the tossup states (MT, ND, MO, IN, FL, GA, NC) (142+85 = 227) plus two of (PA, OH, VA) and at least one of (NV, CO, NM).

    For example, he gets to 270 with the above 227, plus PA, VA and CO. If he loses PA, he needs not only OH and VA, but TWO of the others to get past 269.

    Since it would tear the country apart if he won without winning the popular vote, I surely hope that your scenario is wrong. While I’d much rather see McCain that Obama, it isn’t at any cost, and the cost of serious division and (at worst) armed insurrection and (at best) the end of the electoral college is too much to pay.

    So, not only do I not see this “small state strategy” working, but it would be awful if it did.

    Kevin (0b2493)

  11. Democratic early voters are out in higher numbers than before but we can’t tell if those are:

    (1). New Democrats who haven’t voted before, or
    (2). Democrats who usually vote but are voting early because they are energized, or
    (3). Independent and Republican voters who have switched to Obama.

    (1) and (3) are bad for McCain but (2) will be a wash, and right now I don’t think anyone can make a credible prediction who they are.

    DRJ (cb68f2)

  12. WLS Shipwrecked, the state polls added together should logically show a wider margin than the national polls, just as you have found. But that’s because the state-by-state polls are less accurate than the national polls, have a wider margin of error, and because they also lag the national polls.

    However, based on the history of the Republican winning small states, we know that the state polls added together will give a popular vote result closer on Election Day than the curerent national polls, or may even give the electoral victory to the trailing popular-vote candidate. Remember that the major pollsters don’t make a big deal about geographical distribution on their national polls, and would give the same weight to a New York Republican as a Tennessee Republican.

    If you want to do some math, tell us by how many points in the popular vote you think Obama can win by (in theory) yet still lose the Electoral College to America’s 44th president, John McCain.

    By the way, North Carolina has as much chance of going for Barack Obama as they do for Eleanor Roosevelt.

    Official Internet Data Office (184273)

  13. Why are both campaigns going to Iowa this weekend? National polls show large margin. Has to be wrong. They ain’t spending a second in Iowa on the last weekend unless there’s some doubt about the outcome. Meaning that all the national polls could be way off.

    Dantana (205f0f)

  14. 11- by that logic Bush should have just conceded 2000 election to popular-vote winning Gore. But now because Obama is black and there might be rioting and other violence, we should just say give it to him and ignore the constitution? F**k that.

    Part of me is ok with the bIg nUll in the white house with a congress running amok because maybe then the illiterate/elite masses combined might wake up. Of course by then those 12 million newly minted citizens would show gratitude to the dems and vote the party line for more manna from tax heaven.

    madmax333 (0c6cfc)

  15. OIDO,

    The last CBS poll of the 2004 election showed Bush up by 2%, the margin that he won by:

    This time CBS has Obama up by 11%

    Science don’t lie.

    But a little risk-free wishful thinking never hurt anyone, right?

    snuffles (677ec2)

  16. Now that I’ve seen the NYT/CBS poll, I’m not even going to bother voting. There’s no reason to cast my little vote if Obama and his leftist illuminati ideals are a lock for the White House. Thanks, liberal MSM, for giving me the accurate, unbiased facts once again.

    Jeff (7ff0a7)

  17. WLS, I’d be interested in your thoughts on how your method of analysis contrasts/compares with that of the Republican blogger at Election Projection, who isn’t giving McCain much of a chance. His formulas are at:

    Peccator Dubius (0a6237)

  18. “Obama is going to under-perform just as he has consistently.”? Yeah, just like you predicted there was no way in hell he’d make $80m+ in the last couple of months, when he ended up rolling in $150m?

    If McCain does win Iowa, he deserves extreme kudos, given his honorable stance on cutting farm and ethanol subsidies.

    Slade (e7d80d)

  19. Since it would tear the country apart if he won without winning the popular vote,


    JD (5b4781)

  20. So, not only do I not see this “small state strategy” working, but it would be awful if it did.


    JD (5b4781)

  21. I think the polls are going to either tighten this weekend as pollsters try to save some of their credibility or go nuts for Obama in a last ditch effort to discourage McCain voters. The polls have over sampled Democrats and under sampled Republicans all year. And of course they have bought into the “historic” heavy turnout of young voters. As is always the case, by the time of the election, the massive outpouring of young voters turns to a trickle.

    Obama has a massive lead among the dead, house pets, cartoon characters, illegal aliens and random collections of letters ( especially those registered by ACORN ). Otherwise I think it’s just about even.

    Ken Hahn (092084)

  22. I think the media hype and harping on the polls is causing the opposite of what is intended. A lot of McCain voters may be thinking “at least I can say I voted and made the effort” whereas the “youth for Obama” may see it as in the bag and not show up.

    voiceofreason2 (45f71e)

  23. Whatever happens, you will all still be racists.

    JD (5b4781)

  24. “…the end of the electoral college is too much to pay…”

    Kevin, I don’t think you have much to worry about on this item.
    There are just too many small states, and they will prevent any constitutional amendment from receiving 38 “Aye” votes. I think it would be pretty much a lock that almost every state between the Mississippi River and the Sierra Nevadas would vote “Nay”.
    There are nineteen states in that expanse of country, and I would think that not more than four would vote to abolish the Electoral College.
    Amendment defeated.

    Another Drew (3eb642)

  25. JD, I denounce you for being racist toward racists.

    Eric Blair (51924c)

  26. This new cartoon, by the Pulizer-prize winning Michael Ramirez, reminds me of when Obama went to Germany and announced, “Ich Bin Ein Beginner.”

    Official Internet Data Office (184273)

  27. People who don’t make Joe the Plumbers list of “real” Americans:

    Puerto Ricans
    “Mixed” peoples

    JOE THE PLUMBER (28bd41)

  28. 16, Pure science may not lie, but you do, and as often as you can.

    PCD (7fe637)

  29. This new cartoon, by the Pulizer-prize winning Michael Ramirez, reminds me of when Obama went to Germany and announced, “Ich Bin Ein Beginner.”

    I think he misstated his intent – it should have read “Ich Bin Ein Bullsplitter.”

    Dmac (e30284)

  30. NOW this appears. Certainly explains many of the Trolls this election season.

    PCD (7fe637)

  31. How many polls predicted the 1994 election results ? I think it is unknowable at this point. I’m with Gingrich who believes that there are currents that flow through the electorate and the Rove method of targeting small groups is not the way to run election campaigns. I also think the Obama people know how close it is but are trying to hold the lid on until it is over. They hope to depress turnout by Republicans but risk doing the same to their own youthful supporters. There are, for example, some statistics about absentee ballots in Colorado. College age kids were registered and encouraged to vote by mail but it may be backfiring. Those absentee ballots are not being returned.

    By the way, weekend polling always favors Democrats because people with families are not home to be polled.

    We will all know in a few days. The Danforth doesn’t lie.

    Mike K (2cf494)

  32. I see the Obama supporters are back to their slandering ways with the #28.

    SPQR (26be8b)

  33. Comment by JOE THE PLUMBER — 10/31/2008 @ 5:32 am

    What’s the matter Joe? Not getting enough attention to your trolling at Ace’s?

    X_LA_Native (7afef3)

  34. The race must appear close in order to sell advertising.

    Just like they said with Kerry, they’re saying with McCain

    “It’s close, folks, tune in at 11 to see just HOW close. Now here’s a commercial…”

    i like america (d2f951)

  35. WLS, weren’t you predicting over the summer that at any minute, Obama’s fundraising numbers would fall through the floor? Because if you were, I’d say you’ve been pretty significantly discredited. Why should we believe you about this, given how spectacularly wrong you were a few months ago?

    Bathe her, and bring her to me (54d03f)

  36. Bathe — #37:

    I’ve commented on that subject. At first I conceded that the Obama campaign did what I thought they couldn’t do. And, I’d note in that regard that the NYT ran two stories in Aug and Sept., that commented on uneasiness in Dem circles with Obama’s summer numbers, one of which dealt with the fact that Hillary supporters were contributing far less than the Obama campaign had planned on.

    But, earlier this week, when the news came out that 2/3 of the money raised by Obama in September came from credit card contributions over the internet, and that the Obama campaign had chosen to take no steps to prevent fraudulent or illegal contributions from being made by using AVS or other fraud detection devices typically employed by merchants, I qualified my concession with the comment that I hadn’t anticipated that the Obama campaign would abandon all notions of ethics and good governance in its pursuit of power.

    But, I should not have been surprised. David Fredosso’s book quotes Obama as sayin that the lesson he learned from his first political campaign, when he won by having all his DEMOCRAT opponents disqualified from the ballot, was that “If you can win, you should do what you have to do to win.”

    Reminds of of Sean Connery in “The Man Who Would Be King.”

    WLS Shipwrecked (26b1e5)

  37. “If you can win, you should do what you have to do to win.”

    Anymore, with the huge support networks these candidates need, they have to go to the wall to win if they can or face serious blowback from their own side. But Mavericky McCain probably does not care that much and might pull a Nixon-in-1960. But Obama won’t, not with his backers.

    cassandra (5a5d33)

  38. And right ready to assuage your fears… Gallup daily comes out with leads between 8 and 11 points, pick your model.

    But I think its not very accurate to add up results like this, without following margins of error and adding them up too.

    imdw (513533)

  39. Too many dwell on the popular vote vs electoral vote strawman.

    If you win Ca by one vote or 3 million votes the electoral vote outcome is exactly the same.

    The bluest of blue states have some large population states that go heavy in the Dem column pushing the national popular vote total up but doing exactly nothing for the electoral college vote.

    That is why we have an electoral college to keep 4 or 5 states from dominating the process.

    daytrader (ea6549)

  40. Here’s McCain’s task. Not impossible, but very difficult.

    Obama will win all the Blue states Kerry won in 2004 by large margins. None of them are in play.

    What’s in play, according to respected Republican blogger,, and others, are nine Red states that voted for Bush in 2004, but in which Obama is leading today.

    McCain needs 96 Electoral Votes from the 112 these states represent to win the election. If he loses just Iowa and New Mexico, the two states Obama has the biggest leads in, but gets all the others, he wins, because the rest of the states total 100.

    However, if he loses just one more state from the list below in addition to Iowa and New Mexico, say Nevada with 5 votes, he loses because at best he then only gets 95 votes and that won’t do it.

    Obama +6.3
    EV’s 9

    Obama +3.4
    EV’S 27

    Obama +11.4
    EV’s 7

    Obama +0.1
    EV’s 11

    Obama +6
    EV’s 5

    New Mexico
    Obama +8.7
    EV’s 5

    North Carolina
    Obama +1.7
    EV’s 15

    Obama +5
    EV’s 20

    Obama +5.5
    EV’s 13

    Peccator Dubius (0a6237)

  41. PD — you’re wrong about him not getting a Kerry state — McCain is going win PA. He’s going to run ahead of Bush there, and Obama can blame himself (“bitterly clinging to guns and religion”) and Murtha (“a bunch of racists”). Those chickens are coming home to roost.

    PA is the state where early polling means the least. The traditions of central and western PA being heavily Democrat go back generations and are associated with ethnicity and high levels of union membership in the steel and coal industries which dominated the region.

    But, those areas are the epitome of the term “Reagan Democrat”. They register as Democrats, they run for office as Democrats, locally they vote for Democrats.

    But in Presidential elections, they vote for Republicans.

    Dems have won the state based on large margins in Philly and a couple of its suburbs. That won’t be enough this year with the damage done to Obama by himself and Murtha.

    WLS Shipwrecked (c1b09d)

  42. Here’s an interesting scenario:

    Take a look at the CNN interactive map. Mark all the “toss up” states for McCain. Mark Virginia and Colorado for McCain. Give McCain 1 electoral vote in Maine. McCain 270, Obama 268.

    cimics (9a3aad)

  43. Even the harshest critics of National Popular Vote start with the premise that it would take the 11 (not 4 or 5 as claimed here) largest states to win the National Popular Vote.

    And even then, the political reality is that the 11 largest states rarely act in concert on any political question. In terms of recent presidential elections, the 11 largest states include five “red” states (Texas, Florida, Ohio, North Carolina, and Georgia) and six “blue” states (California, New York, Illinois, Pennsylvania, Michigan, and New Jersey). The fact is that the big states are just about as closely divided as the rest of the country. For example, among the four largest states, the two largest Republican states (Texas and Florida) generated a total margin of 2.1 million votes for Bush, while the two largest Democratic states generated a total margin of 2.1 million votes for Kerry.

    Moreover, the notion that any candidate could win 100% of the vote in one group of states and 0% in another group of states is far-fetched. Indeed, among the 11 most populous states, the highest levels of popular support were found in the following seven non-battleground states:
    ● Texas (62% Republican),
    ● New York (59% Democratic),
    ● Georgia (58% Republican),
    ● North Carolina (56% Republican),
    ● Illinois (55% Democratic),
    ● California (55% Democratic), and
    ● New Jersey (53% Democratic).

    In addition, the margins generated by the nation’s largest states are hardly overwhelming in relation to the 122,000,000 votes cast nationally. Among the 11 most populous states, the highest margins were the following seven non-battleground states:
    ● Texas — 1,691,267 Republican
    ● New York — 1,192,436 Democratic
    ● Georgia — 544,634 Republican
    ● North Carolina — 426,778 Republican
    ● Illinois — 513,342 Democratic
    ● California — 1,023,560 Democratic
    ● New Jersey — 211,826 Democratic

    To put these numbers in perspective, Oklahoma (7 electoral votes) alone generated a margin of 455,000 votes for Bush in 2004 — larger than the margin generated by the 9th and 10th largest states, namely New Jersey and North Carolina (each with 15 electoral votes). Utah (5 electoral votes) alone generated a margin of 385,000 votes for Bush in 2004.

    Under a national popular vote, a Democratic presidential candidate could no longer write off Kansas (with four congressional districts) because it would matter if he lost Kansas with 37% of the vote, versus 35% or 40%. Similarly, a Republican presidential candidate could no longer take Kansas for granted, because it would matter if he won Kansas by 63% or 65% or 60%. A vote gained or lost in Kansas is just as important as a vote gained or lost anywhere else in the United States.

    Although no one can predict exactly how a presidential campaign would be run if every vote were equal throughout the United States, it is clear that candidates could not ignore voters in any state. The result of a national popular vote would be a 50-state campaign for President. Any candidate ignoring any particular state would suffer a political penalty in that state.

    susan (4f64ca)

Powered by WordPress.

Page loaded in: 0.3749 secs.