Campaign 2004

Gallup: Race A Dead Heat

The latest Gallup poll shows Bush and Kerry tied at 49%, but Bush ahead slightly with registered voters. It’s clear that Bush’s poor debate performance has hurt him in the polls. However, the Gallup poll was taken Friday through Sunday. For whatever reason, weekend polls tend to dramatically depress Republican numbers, especially polls on Sundays. (Given that GOP voting is strongly correlated to church attendance, this isn’t a big surprise.) It would be interesting to see how the numbers play out over a larger sample set. So far we have Newsweek using skewed numbers to show a major Kerry surge, Rassmussen’s daily tracking poll showing the race largely unchanged, and Gallup’s weekend poll showing the race dead even. However, based on the ABC and Gallup snap-polls taken before and after the debates, it would appear that the number of voters that changed their minds post-debate is relatively small. By mid-week there should be enough polling to give a clear indication of how accurate those snap polls were.

The question is now how Bush will respond. Bush did an abysmal job of countering Kerry in the debates, but Kerry provided plenty of fodder for the Bush team to work with in attack ads later. From giving the Iranian mullahs nuclear fuel to bilateral talks with North Korea for subjecting American action to a “global test,” Kerry’s foreign policy plans are foolish at best and dangerous at worst. Unfortunately Bush didn’t call Kerry on many of his assertions, but they will not go unchallenged for long.

11 thoughts on “Gallup: Race A Dead Heat

  1. Yes, but Jay, isn’t the best indicator of trends during an election is to follow one pollster through the race?
    Certainly this practice reduces the number of measured variables and improves the accuracy of claim.

  2. Bush is being absolutely skewered in the post-debate analysis, which counts for as much if not more than the debate itself (remember when Gore’s “win” in the first debate turned into loss within 48 hours?). As for the whole idea that this Gallup Poll is being mean to Bush by polling on Sunday, it’s time to face reality. Your boy was badly battered. Gallup has consistently shown the most Bush-friendly poll numbers this whole campaign, often as much as seven or eight points to the right of the poll averages for any given two-week period. If even Gallup can’t produce anything better than a Bush tie, that thud you just heard was the lid of the casket slamming shut on Bush’s chances.

  3. Karen: Not necessarily. Different pollsters have different biases and as with any statistical calculation it’s best to have the widest possible sample set in order to draw inferences from. For instance, Zogby applies an artificial weighting mechanism in his polls based on the partisan balance in 2000. If the electorate has changed since then, his national polls are going to be skewed.

    Mark: Any poll taken over a weekend will be skewed towards the Democrats. Even Democracy Corps shows Bush ahead by 2, and they’re James Carville’s pollster. If you think that Bush’s chances are through when Bush is trouncing Kerry in the EC vote and after a disastrous performance by Bush the best Kerry can do is tie him, you’re deluding yourself.

  4. Jay, are you suggesting Kerry can get a nine-point poll bounce and it won’t affect the Electoral College numbers game? Isn’t much fun playing defense it? I’ll enjoy watching the endless spin, manipulation and contradiction coming from you and other GOPers in the days to come as poll after poll validates Kerry’s bounce. You’re correct that a false move by Kerry or a shrewd move by Bush and his puppetmasters could quickly change the dynamic, but if Bush doesn’t do better in the second and third debates than he did in the first (which involved foreign policy, his perceived strength), the guy will lose and lose relatively soundly.

    As I stated before, Bush’s battering on Saturday Night Live and other post-debate analysis coupled with reports of Kerry’s big bounce are certain to amplify Bush’s problems. If Kerry’s smart (as he proved he was Thursday night), he’ll use this momentum to correct his campaign’s biggest strategic error of the race thus far and insist they start advertising hot and heavy in Missouri and Arkansas again.

    Bottom line: if Bush doesn’t do demonstrably better or Kerry demonstrably worse this Friday night, Bush is in serious trouble. I’m expecting soft Kerry states (New Jersey, Minnesota, Pennsylvania) to become much harder in the next round of state polling, and for Florida and possibly even Ohio to start narrowing again (your buddies at Rasmussen are already showing it a one-point race again in OH).

  5. There was no nine-point bounce. Newsweek and Gallup are both relying on dodgy figures. Rassmussen, Democracy Corps, and the ABC and Gallup snap polls all showed at most a two-point bounce for Kerry – if that.

  6. Gallup had Kerry behind by eight as recently as last week. Now they’re tied. That’s close enough to a nine-point bounce for me. You look nervous, Jay. Calm down a little bit… 🙂

  7. The Bush campaign made two fundamental mistakes in the debate. First, Bush should simply have picked up more on the many Kerry gaffes and contradictions — Saturday Night Live did an excellent job at this! Second, the campaign should have agreed to debates only on a Tuesday (Monday would have been better, but MNF is clearly the issue here). Weekend polling — Saturday in particular — consisently shows a much smaller Bush lead than polls taken Monday through Thursday because conservatives are underrepresented in these polls. By having the debate on a Thursday, the only polling that came out right after the debate was conducted Friday through Sunday, yielding the worst possible set of results for Bush.

  8. Mr. Mike, Republicans are less likely to have jobs than Democrats (think housewives versus single working mothers). Whatever disadvantage you claim the GOP has on the weekend, presumably because they’re either in church for 48 hours or out partying on their yachts, the Democrats would have an equal disadvantage on the weekdays when they’re in the factory, at least the ones that haven’t moved to China yet.

  9. Mark, so you are using speclatory evidence to support the claim that democrats have more jobs than republicans. Since when does one’s party affiliation effect one’s job placability? Can you link to any substantial evidence that proves democrats are have more jobs than republicans?

    And the Yachts joke simply parrots Al Franken’s on his Yahoo! commerical. Lame.

    I would consider it damning if we say a quick and large turn around in the polls against Bush. But we have not. I think the overall effect of individual debates will either be mis-measured (due to the delay in truely observing its effects) and the cross-pollentation of having many debates in a short period of time. For example, by time we truely start seeing the effect of Thursday’s debate, we’ll all be talking about the Vice Prez. debates.

  10. Chris, how is my evidence any more “speculatory” than the preposterous charge that weekend polls favor us Satan-worshipping Democrats who spend our Sunday mornings burning bibles next to the telephone while pious Republicans are spending the day in church and missing all the calls from Newsweek pollsters? I was merely trying to dispel the false credibility of Bush shills convincing themselves that every poll making their guy look bad is some sort of vast left-wing conspiracy undertaken by pollsters.

    As for my yachts remark, it was made without any knowledge of “Al Franken Yahoo ads”. I didn’t know such creatures existed, but would happy to see them if you led me in their direction.

  11. Actually it also has a lot to do with marriage as well. Married couples tend to be far more Republican than unmarried people, and you’re far less likely to get someone who is married with kids to talk to a pollster for 20 minutes on a weekend.

    The facts don’t lie – weekend polls shift more Democratic than do weekday polls.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.