Mason-Dixon Polls From Blue Country

    Over the weekend, we got Mason-Dixon’s results from the red states. Tonight, we get the numbers from the blue states and the news is not good if you are a Kerry fan.

    Bush 48%
    Kerry 42%
    Nader 2%

    Kerry 47%
    Bush 41%
    Nader 2%

    Kerry 46%
    Bush 45%

    Kerry 46%
    Bush 44%
    Nader 1%

    Bush 47%
    Kerry 43%
    Nader 1%

    Kerry 45%
    Bush 44%
    Nader 1%

    Bush 47%
    Kerry 43%
    Nader 2%

    Bush 46%
    Kerry 44%
    Nader 1%

    The polls were done September 14-16.

    UPDATE: Oops, forgot Pennsylvania. Has been added above.

    Posted by Dave at 7:27 pm
    Filed under: Election 2004 Polls - 09/04 | Comments (43)

    43 Responses to “Mason-Dixon Polls From Blue Country”

    1. Tina says:

      Iowa and New Mexico I moved to Bush.

      Oregon, I’m shocked about that result.

    2. Zoltan says:

      Also polled Pennsylvania

      Kerry +1

      Sample looks a bit Dem frioendlky however.

      Had 6% more Dems than GOPers (2000 exit polls GOP and Dems exactly even in party ID)

    3. Scotsmac says:

      Reagrading the Minnesota Poll… The breakdown completely ignores Central Minnesota. Specifically the St.Cloud Area. This is the 2nd most Bedrock Republican Area of the State. Methinks someone at Mason Dixon should do more research on the state. I would like to see a better Break down on responses from these different areas. If I could I could give a Solid state of the Race in Minnesota.

    4. David says:

      Maybe that is why I could not find the breakdown betwwen Dems/Reps on the MN poll.

    5. MikeKS says:

      As I said, I believe while these polls are excellent, that they probably actually favor the Dems. PA, now, in several polls show bush up by 1-4 points, so I think that you can probably assume that’s off. Same thing for Minnesota..though the # is consistent with others, I think pollsters are assuming the state is Democratic when Republicans have made huge inroads. Also, look at the internals for Wisconsin and see Kerry’s unfavorables. My guess is the lead is larger than 2 points, and probably more along hte line of 6-8 you’ve seen in other polls.

      Oregon, again, we’ve gotten different numbers but Gore barely won it in 2000!! there is no reason to believe it wouldn’t be in play, given that result. If you think Bush has an 8 point lead nationally, then Oregon would have to swing his way. I am prepared to say right now, that I believe if the election were held today, you’d have bush at 350+ in th EV category. MD, MI, NY, IL, etc…would hold on for Kerry but by tight margins.

    6. Alcon says:

      Because every single poll ever released has good internals for Republicans? Sheesh, man.

      In any case, my count is now Bush 313-Kerry 195-Undecided 31 (OR, WV, NJ). If I had to give states, I’d give all three of those to Kerry, though, so Bush 313-Kerry 226.

      And I just realized that added up to 539. I hate the world. Off to find my mistake.

    7. MikeKS says:

      By the way, i would personally like to thank Senator kerry for ruining his campaign today. Can you believ ethe gift he gave when he said he wouldn’t have removed Saddam, that the world is not safer? Did his adivosrs not know he made that stupid comment back in December, when he was running against Dean? Bush just had an absolute field day with it. I can’t believe Kerry actually said that…is he that incompetent?

      I thought Kerry might be able to dig into Bush’s lead some on Iraq, but now i’m not so sure. He just has no credibility and his statements make no sense.

    8. Scotsmac says:

      No Party ID #’s shown just this…

      Male 310 (50%)
      Female 315 (50%)

      Hennepin/Ramsey Counties 210
      Twin Cities Suburbs 150
      Rochester/Southeast 70
      Southwest Minnesota 60
      Northwest Minnesota 65
      Duluth/Northeast 70

      I also found interesting that they have it 46-44-1 for Kerry with Nadar. Yet 46-45 Bush in a Head to Head.

      They state Quotas were used to divide it weighted by area yet No Party Id is examined at all.

    9. Alcon says:

      Major calculation error there…it’s actually Bush 333-Kerry 177-Undecided 28. Sorry.

    10. McCain says:

      Alcon, don’t hate the world. You are too young to be a committed liberal.

    11. Scotsmac says:

      Yeah Alcon You seem very intellagent. hehehe There is hope you will be converted to the Truth of Conservatism yet. LOL

    12. Scotsmac says:

      Gah my spelling. Intelligent… pfff

    13. Tina says:

      Alcon, I have Bush ahead 321 to 217.

      I won’t move Oregon or New Jersey (from a couple days ago_, nor would I move PA back to Kerry yet.

    14. Craig says:


      Do you promise to hold the same opinion nine months from now, about something that is sure to change everyday?

    15. Alcon says:

      Well, thanks, I guess, although I must admit it has always bothered me when people discount my political opinions soley because I am what age I am. In fact, I come from a family that is mostly Republican. I’m a dedicated liberal, and I can’t see how I would not remain so throughout my life.

      But time will tell.

    16. bob says:

      New Mexico poll is 54-36 D but Bush is still up 4

      Mike Moore says these are wrong and young people with cell phones are gonna save Kerry

    17. McCain says:

      Alcon, your opinions are not discounted. You are one of the most articulate bloggers here. But you know, when you say that you hate the world, it was just a fastball right in the zone — you know the rap on liberals.

      About age, studies have shown that people do indeed grow more conservative as they get older. Interesting I think. There is a cute cliche which captures this reality nicely — umm, but I forget it, another sign of aging.

    18. bob says:

      I like Al.Disagree with him/her but at least he is honest with himself and results

    19. Rachel says:

      Alcon, you seem very intelligent, especially for your age. I guess that’s why we can’t help hoping to have you on our side someday. 🙂

    20. Tina says:

      I’m waiting for Hawaii to go to Bush now. They have the lowest unemployment rate 2.9%. I think it is a possibility since a Republican Governor was elected in 2002. Can this happen? Probably not.

    21. Alcon says:

      Well, thank you in any case. I was just having a bit of fun, of course (as were you). Didn’t mean to sound serious or anything. 🙂

    22. McCain says:

      Fine, you just proved you are not a liberal. Liberals don’t have a sense of humor.

    23. David says:


      As discussed last night, Hawaii is not quite Republican yet.

      …andy Alcon will be on our side by Nov. 2.

    24. joe shmoe says:

      Nah, Alcon seems like a smart lad. He’ll stick with the thinking man’s party. 🙂

    25. Alan R. says:

      Of these Mason-Dixon polls, the only ones that would be attention-grabbing (to me) would be Iowa, Oregon, and New Mexico.

      Let’s look at some state polls at

      OREGON: Of the four most recent polls — Zogby, M-D, Rasmussen, and ARG — Kerry leads in three of the four, albeit sometimes by close margins.

      NEW MEXICO: Little recent polling, just M-D and Zogby (which is big for Kerry).

      IOWA: A much more mixed picture than Dave portrays. Zogby has Kerry up 3, Rasmussen has Bush up by 1, and Strategic Vision (GOP pollster) has Bush up 1 or 2 (depending upon whether Nader’s included).

      Some of the other M-D polls are actually pretty good for Kerry:

      *Kerry’s lead is showing resilience in Michigan.

      *Minnesota, Pennsylvania, and Wisconsin are 1-2 point races (with Kerry leading in two of the three), but Bush is consistently getting only around 45%.

      (Despite what some have written, incumbents DO NOT tend to pick up the undecided, most dramatically illustrated by Carter in ’80 and Bush in ’92; the latter getting 38% of the vote, the lowest for an incumbent in EIGHTY YEARS.)

    26. Polaris says:


      You should completely discount Zogby interactive polls since they are self-selecting polls and based on exit polls for demographics. They have been consistantly bad (when compared with other polls) all election cycle.


    27. Alan R. says:

      Quite the contrary on Zogby. I did a detailed analysis (written up as a Daily Kos diary) on how Zogby Interactive (ZI) fares in comparison to other polling outfits that conducted traditional telephone surveys in the same states and at the same times as ZI. The bottom line: ZI acquitted itself well.

      Further, in my Message 25, I drew upon other polling outfits besides ZI. I would suggest neither accepting nor rejecting ZI in a knee-jerk sort of fashion. ZI should be viewed as one piece of data for a given state, among other pieces of data.

    28. Alcon says:

      If Alcon is on your side by Nov. 2, David, Alcon will personally PayPal you $150. 😉

    29. Brandin says:


      I would give Bush back Pennsylvania due to a new Quinnipiac poll. I saw it today in the Philly newspaper…even though I live in DC…I read Philly papers because of Melissa Brown and the fact that I am a Pittsburgher

      Pennsylvania: Bush 47 Kerry 44 MoE-4
      with a Keystone Poll showing Bush 47 Kerry 45 MoE-4.4

      I think even PA politicos say Bush owns PA

    30. Polaris says:


      You trust the Daily Kos? I don’t. I also don’t trust any self-selecting poll. Zogby Interactive has been all over the map. Up 12 in WV for Bush while Kerry leads AR.


      Mason-Dixon says Bush +6 OR while Z.I. says +12 Kerry?



    31. Alcon says:


      I think he probably trusts Daily Kos when it is in a diary he wrote himself. 😉

    32. Polaris says:


      Sorry, misunderstood you. That was your article. I still say you are off base. How do I know? First of all ZI has been pro-Kerry and consistantly off base in NM (by a long ways), OR, FLA, AR (see above), and now in IA and WIS as well. It is also Prokerry in TN which is not a battleground state no matter how much pro-Kerry people would like to pretend otherwise.

      Furthermore, I suggest you go to and look at the Electoral College graph and overlay it with the days the ZI results are posted. You will find a “sawtooth” pattern where the poll goes pro-Kerry for about 2-3 days and then goes pro-Bush as the ZI polls are replaced with other polls.

      If ZI was a good poll, you would not see that sawback pattern.


    33. Tina says:

      David, I’m just joking about Hawaii.

      I have kept PA to Bush, and Oregon/ New Jersey to Kerry. I have moved IA to Bush.

    34. joe shmoe says:

      Alan, interesting analysis.

      According to this site:

      Zogby is actually quite a bit less biased than other polls some folks here do find valid.

    35. Polaris says:


      That’s Zogby not Zogby Interactive. There is a difference. It is a iron law of polling that you never trust a polling sample that is self-selecting. That is why internet polls including ZI are garbage.


    36. Alex says:

      I was just thinking about this. Arkansas has had Kerry and Bush seperated by at most two points. What makes people so sure that Arkansas is safe Bush? I think we need to see a new poll out of there come out soon.

    37. McCain says:

      Isn’t the gay marriage amendment on the ballot there?

    38. Polaris says:


      Demographics tell me that AR is safe for Bush. If LA is safe for Bush (and it is) and TN and MO are safe for Bush (and they are), then AR is as well because of similiar demographics.

      Now there may be some variance, but given the situation as it is, there is no way that Kerry is ahead right now in AR. Hell, Kerry doesn’t think he has a chance in AR (look at where he is spending his time and money!)


    39. Jeremy says:

      Ok Alan,

      Let’s look at REAL data from past Presidential elections:

      BTW, that evil Gallup poll data

      Poll 1 month out: Bush 48, Gore 43
      Last poll: Bush 47, Gore 45
      Result: Bush 48, Gore 48

      Undecideds break toward incumbent VP

      Poll 1 month out: Clinton 48, Dole 39
      Last poll: Clinton 52, Dole 41
      Result: Clinton 49, Dole 41

      Broke evenly

      Poll 1 month out: Clinton 47, Bush 29
      Last poll: Clinton 43, Bush 36 (hmm, looking interesting)
      Result: Clinton 43, Bush 38 (Wow)

      Broke toward incumbent

      Poll 1 month out: Bush 49, Dukakis 43
      Last poll: Bush 53, Dukakis 39
      Result: Bush 53, Dukakis 46

      Broke toward incumbent VP from poll to poll, challenger after last poll

      Poll 1 month out: Reagan 58, Mondale 38
      Last poll: Reagan 56, Mondale 39
      Result: Reagan 59, Mondale 41

      Broke evenly

      Poll 1 month out: Carter 47, Reagan 39
      Last poll: Carter 47, Reagan 39
      Result: Reagan 51, Carter 41

      Broke toward challenger (there’s the one you’re looking for, Alan) Of course, John Kerry is no Ronald Reagan.

      But wait, let’s look at these:

      1 month poll: Ike 51, Stevenson 41
      Last poll: Ike 51, Stevenson 41
      Result: Ike 57, Stevenson 42 (hmmm)

      1 month poll: Dewey 46, Truman 40
      Last poll: Dewey 50, Truman 45
      Result: Truman 50, Dewey 45 (hmmm)

      1 month poll: FDR 47, Dewey 45
      Last poll: FDR 47, Dewey 45
      Result: FDR 53, Dewey 46

      Ok, I’m going to stop right there. Please, Please, PLEASE, stop making this incorrect statement that undecideds will break for the challenger in a Presidential election.

      Thank you.

    40. Bernie says:

      Saying undecidedes will vote for challenger is false. By the time the election nears voters have usually been swayed to change or to just stay the course. Kerry is doing nothing to see why the masses would change to him. He seems to be looking at what he said in the primaries about Iraq and then doing a 180 on everything.

    41. Alan R. says:

      A few points:

      1. I only consider races with an incumbent PRESIDENT to be valid for this type of analysis. A sitting VP does not get anywhere near the exposure of the president, so will have had less chance to win voters over than a president. Voters will still be making up their minds about a sitting VP late in the campaign, thus permitting some volatility in the polls.

      2. In ’92, if you want to claim great credit for the idea that undecideds break toward the incumbent from H.W. Bush going from 29% in the polls to 38% on Election Day, that’s your prerogative. I find such an argument unpersuasive, to say the least. Each party has a rock-hard core of around 40% (landslide losers Goldwater [38.5%], McGovern [37%], and Mondale [41%] each bottomed out at around 40%). H.W. Bush’s “rise” to 38% likely represented nothing other than a consolidation of the Republican base. I see no evidence that truly undecided, swing voters broke to H.W. Bush at all!

      3. The other examples go really far back, where the polling methods were very different from what they are now. Even the 1948 election, where it looks like there was movement toward the incumbent (Truman), provides no evidence of that. More likely is that the polls showing Dewey ahead were just plain wrong. Earl Babbie writes the following in his textbook, “The Practice of Social Research”:

      “Quota sampling — which had been effective in earlier years — was Gallup’s undoing in 1948… For national political polls, such information came primarily from census data. By 1948, however, World War II had produced a massive movement from the country to cities, radically changing the character of the U.S. population from what the 1940 census showed, and Gallup relied on 1940 census data.”

      In conclusion, I submit that if one restricts the field to only those elections with an incumbent PRESIDENT and one looks at ’92 in the proper context, there is NOT A SCINTILLA of evidence for movement toward the incumbent in the last 24 years.

    42. McCain says:

      Well, Alan, there are three possibilities:
      1) break for incumbant
      2) break for challenger
      3) break for nobody, or even

      There is at least as many examples that undecideds break evenly than break for the challenger or incumbant. In 1992, they broke for the 3rd party candidate. The timeline one uses is important. People who are undecided on election day break for the challenger usually, an exception being 2000.

    43. leaddog2 says:

      The Albuquerque Poll is from the largest paper and the largest city in New Mexico. It is a Plus for Bush. I suspect that it is MUCH more accurate than Zogby.