It’s been a general tendency throughout this election campaign, but it is quite striking how well Kerry is doing among independent voters in the various pre-election polls that have been recently released. If Kerry’s lead among independents holds up, it will be very, very difficult for George Bush to win this election.
Here’s why. Averaging the last eight national polls for which I can obtain relevant data, Bush is running about a 7 point higher margin among Republicans than Kerry is getting among Democrats. That’s actually worse than Bush did in 2000, when he ran an 8 point higher margin among Republicans than Gore did among Democrats.
Then bring in our independent voters. Right now, Bush is averaging about a 7 point deficit among independent voters, compared to the 2 point advantage among these voters he had in 2000, a 9 point swing against him among independents.
So Bush seems likely to do slightly worse than 2000 in terms of receiving higher margins, relative to the Democrats, from his own partisans and very likely to do much worse among independent voters than 2000. Given that he actually lost the popular vote in 2000 by half a percentage point, he therefore has no chance of carrying the popular vote, or even coming close, unless he can change something else in his favor to counter these negative trends.
That can only be accomplished by erasing the Democrats’ traditional turnout advantage in Presidential election years (Democrats generally run 3-5 points higher as a proportion of voters). In fact, these data suggest that, given the deficit Bush is likely to run among independents, simply erasing the Democrats’ turnout advantage would likely not be enough; Republicans would actually need a turnout advantage of their own–perhaps a point or two–to prevail.
How likely is this? Not particularly. Such a radical shift in turnout patterns is only possible with a high mobilization of Republicans that is not counterbalanced at all by mobilization of Democrats. Does that sound to you like a description of this year? Not to me either; on the contrary, Democrats and Democratic institutions are exceptionally mobilized this year and it seems likely they, not the Republicans, will win the turnout wars.
And, if that‘s true, there is no way the Republicans can recover from the way independent voters are swinging against them–not only on the national, but also in key states like Ohio and Florida.
Ohio: In 2000, Bush carried Ohio independents by 16 points (!); this year, he is dead-even with John Kerry among this sector of the Ohio electorate (Gallup Ohio poll, October 29-31)
Florida: In 2000, Bush and Gore were about even among Florida indpendents; this year, Kerry leads Bush by 5 among these voters (Gallup Florida poll, October 28-30).
Election day seems likely to be independents’ day–to Bush’s bitter regret.

By Alan Abramowitz
Florida and Ohio are the keys to the 2004 election. If John Kerry carries one of them, he has an excellent chance of winning. If he carries both of them, as I believe he will, he is almost certain to win. Instead of just averaging all of the polls, let’s combine them all, based on the actual numbers of respondents in each sample of likely voters. Remember, I’m looking at likely voters here, but Kerry generally does better among registered voters and in this high turnout election, the registered voter results probably give a better picture of what is going to happen. But many polls do not report results for registered voters, so we’re stuck with the likely voter numbers.
Here’s what they show. In Florida, there have been 12 polls since October 15 with a combined n of 9526 likely voters. George Bush is supported by 47.3 percent, John Kerry by 46.3 percent. So Bush is still well below the crucial 50 percent mark generally needed by an incumbent.
In Ohio, there have been 11 polls since October 15 with a combined n of 9034 likely voters. John Kerry is supported by 47.9 percent, George Bush by 47.4 percent. Again, Bush is not close to the 50 percent mark and is actually trailing slightly in this combined sample of likely voters.
With a huge turnout expected tomorrow and with even a small undecided break toward the challenger, we should all be celebrating tomorrow night.

Today’s tracking polls, like yesterday’s, are a mixed bag in terms of movement favoring Bush or Kerry–which is the typical pattern, as Alan Abramowitz has noted. (You can find all the relevant data and links at NowChannel.com.)
Here are today’s results:
Fox 2way RV: 48-45 Kerry, from 47-45 Kerry yesterday
Fox 2way LV: 48-46 Kerry, from 48-47 Kerry yesterday
Fox 3way RV: 47-45 Kerry, unchanged from 47-45 Kerry yesterday
Fox 3way LV: 48-46 Kerry, from 46-46 tie yesterday
TIPP 2way LV: 45-44 Bush, from 47-42 Bush yesterday
TIPP 3way LV: 47-45 Bush, from 48-43 Bush yesterday
Rasmussen: 48.8-47.4 Bush, from 48.1-47.1 Bush yesterday
WaPo RV: 48-48 tie, from 48-47 Kerry yesterday
WaPo LV: 49-48 Bush, from 48-48 tie yesterday
Zogby: 48-47 Bush from 48-48 tie yesterday
Note: Over the last three days, Fox has moved steadily toward Kerry; over the same time period no other poll has exhibited steady movement toward either candidate. It may be significant that the Fox data are the most recent, since they conduct their tracking poll only over the last two days, while the rest do three days, except for WaPo, which does the last four days. Therefore, if there has been pro-Kerry movement very recently (as suggested, for example, by results of today’s CBS News poll), the Fox poll would be more sensitive to it.
Note: At this stage of the election in 2000, Bush was ahead of Gore by one point in the Zogby tracking poll, by 3 points in the ABC/WP tracking poll and by 6 points in the TIPP tracking poll (3-way).
Note: It is not clear which of these organizations have now done their last polls though, based on past practices, we can expect at least Zogby, TIPP and Rasumssen to release their final results tomorrow–but Fox and WaPo may also release results. I just don’t know.
Note: Today’s results, whatever the movement from yesterday, all imply a close election, with results ranging in a narrow band from +2-3 Kerry to +1-2 Bush.
Note: Taking the most common horse race result whereever possible (3-way LV), these polls average 47.8 percent for Bush to 47.1 percent for Kerry. Not an auspicious result for Bush in light of the “50 percent rule“.

Yesterday, two days before the election, CBS News reported a 3 point Bush lead among LVs. That survey covered October 28-30. Today, they released survey data covering October 28-31 which show Bush and Kerry now tied among RVs (46-46) and Bush up by only a point (47-46) among LVs. Since these data only differ from the data released yesterday in the inclusion of interviews from Sunday, that suggests Sunday was a strong day for Kerry in this poll–and, perhaps, in others.)
It’s interesting to note that, in 2000, Bush was also leading (by 5 points) in the CBS News poll released 2 days before the election. But that lead also pulled a disappearing act between then and election eve. History may be repeating itself.

By Alan Abramowitz
1. The National Polls
In the 12 most recent national polls listed on pollingreport.com, among likely voters, Bush is leading in 7 polls, Kerry in 2, and 3 are tied. Average support was 48.2 percent for Bush, 46.7 percent for Kerry, and 0.8 percent for Nader. In the 7 polls that provide results for registered voters, however, Kerry is leading in 4, Bush in 1, and 2 are tied. Average support was 47.0 percent for Kerry, 46.7 percent for Bush, and 0.9 percent for Nader.
Bottom line: Even in the samples of likely voters, Bush is well below the 50 percent mark generally needed by an incumbent. In fact, when Gallup allocates the undecided vote, their likely voter sample goes from a 49-47 Bush lead to a 49-49 tie. In the broader samples of registered voters, Bush is actually trailing in most of the recent polls. With a very high turnout expected tomorrow, the registered voter samples are probably more representative of the actual electorate than the likely voter samples.
2. The Four Major Battleground States
In Florida, there have been 11 polls since October 15. Bush led in 5, Kerry led in 5, and 1 was tied. Average support was 47.5 percent for Bush, 46.5 percent for Kerry, and 1.2 percent for Nader. Turnout in the early voting has been enormous, with a clear advantage for Democrats. Expect a huge turnout tomorrow as well that will put this state in the Kerry column.
In Ohio, there have been 11 polls since October 15. Kerry led in 7, Bush led in 3, and 1 was tied. Average support was 47.2 percent for Bush and 48.3 percent for Kerry. Ralph Nader is not on the ballot. Turnout is going to be enormous and two federal judges ruled this morning that Republican political operatives cannot challenge voters in minority precincts. That was Karl Rove’s last gasp in Ohio. The Buckeye state will go Democratic this year and no Republican has ever won a presidential election without carrying Ohio.
In Pennsylvania, there have been 11 polls since October 15. Kerry led in 8, Bush led in 2 and 1 was tied. Average support was 46.8 percent for Bush and 48.7 percent for Kerry. Ralph Nader is not on the ballot. Pennsylvania looks solid for Kerry.
Finally, in Michigan, there have been 5 polls since October 15, including only the most recent release of the Mitchell tracking poll. Kerry led in all 5 polls. Average support was 44.2 percent for Bush, 47.2 percent for Kerry, and 1.0 percent for Nader. Michigan also looks solid for Kerry.
Bottom line: George Bush’s situation in all four of these key battleground states is dire. His support is well below 50 percent in all of them and he is currently trailing John Kerry in 3 of the 4. A clean sweep of all four states by John Kerry is a distinct possibility.

Sunday night, Gallup released their final national poll, based on an unusually large sample (over 2,000 adults), plus six polls in key battleground states (OH, FL, PA, IA, WI and MN). The results indicate that Kerry is finishing strongly and should be in a good position to pull off a victory on Tuesday.
In the national poll, Kerry is ahead by a point among RVs in a 2-way race (48-47) and by 2 points in a 3-way race (48-46). In 2000, it’s worth recalling, Gallup’s RV result was a better predictor of the final outcome than their LV result, as it has been in three of the last four presidential elections.
After allocating undecideds, Gallup’s LV result is a dead heat, 49-49. That’s a considerable improvement from their last poll, where Bush was running a 5 point lead, 51-46, among LVs. And keep in mind that in 2000 Gallup’s final LV result gave Bush a 2 point lead, while Gore went on to win the popular vote by half a point. In that context, a dead heat final estimate from Gallup makes Kerry look pretty good heading into election day.
Alan Abramowitz below has discussed the significance of Kerry’s solid lead among independents in the national poll (8 points among RVs; no LV breakdown available). The poll also shows Bush’s approval rating at just 43 percent in the battleground states and Kerry beating Bush by 10 points in those states (52-42).
Speaking of the battleground states, it struck me as quite significant that Gallup’s state polls showed Kerry with solid leads in both Ohio and Florida among both LVs and RVs, since those were the two states in the “big three” (OH, FL, PA) that seemed most competitive and were red states in 2000. And, while Gallup’s PA poll did show Bush with a lead among LVs, it also showed him trailing among RVs in a state where polls have very consistently shown Bush behind. The most reasonable assumption, it seems to me, is that Kerry is still the odds-on winner in that state.
The other results–a strong Bush lead among LVs in WI (but a small lead among RVs), a big lead for Kerry among both LVs and RVs in MN and close to a dead heat in IA–don’t change my impression that this is a good set of polls for the Kerry campaign. But I was surprised about how CNN and other media outlets played these polls, implying that they were too much of a mixed bag to be good news for either candidate. Ohio and Florida to Kerry? Ho-hum. Just another symptom of a dead-locked race, etc.
Reader Lawrence Becker shared my surprise and contributed this entry in what could be a new game: “How can CNN…….”

Again, I am amazed. But this time, it is the interpretation of the Gallup poll that amazes me, not the poll itself. As you know, Gallup just released a set of battleground state polls that (if accurate) are remarkably positive news for John Kerry and pretty devastating news for George Bush. And yet, CNN’s interpretation of these polls is, “President Bush and Democratic Sen. John Kerry were almost evenly split among likely voters in six major
battleground states the weekend before the election.” Well, that’s
techinically true. Among likely voters, Bush is ahead in 3 of these states and Kerry is ahead in 3. But CNN, perhaps knowing Gallup’s “likely voter” model is very suspect goes on to say, “the figures were mostly the same among registered voters, except in Pennsylvania, where Kerry had a 2-point lead.” Well that’s a pretty big “exception,” wouldn’t you say?
Let’s take a look at how “split” the race really is in these states among BOTH likely voters AND registered voters.
Florida (27)
(LV) Bush 46, Kerry 49
(RV) Bush 45, Kerry 49
Iowa (7)
(LV) Bush 48, Kerry 46
(RV) Bush 47, Kerry 46
Minnesota (10)
(LV) Bush 44, Kerry 52
(RV) Bush 43, Kerry 51
Ohio (20)
(LV) Bush 46, Kerry 50
(RV) Bush 44, Kerry 51
Pennsylvania (21)
(LV) Bush 50, Kerry 46
(RV) Bush 47, Kerry 49
Wisconsin (10)
(LV) Bush 52, Kerry 44
(RV) Bush 49, Kerry 46
In these six states, 95 electoral votes (as well as the presidency itself) are up for grabs. I distribute the all the other states to Kerry and Bush exactly as they voted in 2000 with one exception. I give New Hampshire to Kerry based on countless polls that show Kerry with a lead there. That leaves Bush with 227 electoral votes and Kerry with 216. If we just take Gallup’s likely voter results (a very risky proposition, indeed), we find that Kerry would win 57 of these 95 electoral votes (Florida, Minnesota, and Ohio). Kerry would end up with 273 electoral votes and the presidency. And that assumes Kerry really would lose Pennsylvania, a possibility I find very
hard to believe since Kerry has led EVERY ONE of the last 20 polls reported by NowChannel.com. But okay, we’ll settle for 273 electoral votes if we have to. It isn’t horseshoes and it isn’t hand grenades, right? So Kerry wins even with Gallup’s likely voter data -data we already know to be biased against the Democrats.
Now if we just take Gallup’s registered voter results, we find that Kerry would win 78 of 95 electoral votes (including the entire trifecta commonly known as “FLOHPA”). That would put Kerry at 294 electoral votes giving him some room for error in New Hampshire, New Mexico, etc. But wait, the news gets worse for Bush and better for Kerry. Bush is not at at 50% or above in any of these six states among registered voters. By now, we all know that Bush is highly unlikely to improve at all on his showing in poll numbers on the eve of Election Day. Gallup seems to acknowledge this fact by pointing out in their interpretation that while their national poll shows Bush at 49 and Kerry at 47 among likely voters nationwide, “Using voting behavior data from previous elections, the Gallup organization attempted to estimate how the undecideds would vote Tuesday. The result was a tie of 49 percent each for Bush and Kerry …”
Can Bush win if these registered voter numbers are accurate? Simply put …no. Can he even win among Gallup’s biased formulation of likely voters? Probably not. Are these poll numbers accurate? Who knows? But I ask the key question of our new game. How can CNN … read these numbers and come up with the headline, “Poll: Bush, Kerry split six key states?“? To use a favorite term of the President’s, that seems like a bit of an “exaggeration.”

By Alan Abramowitz
No Republican in the modern era has won a presidential election without carrying the independent vote. In the 2000 election, despite losing the national popular vote, George Bush led Al Gore by a narrow margin among independent voters. However, according to the final, supersized Gallup Poll, John Kerry now has an 8 point lead over George Bush among independent voters. In fact, Kerry led Bush among independents in all four Gallup polls this month–by an average of 6 points. This is one more piece of evidence that George Bush’s situation is extremely dire.

By Alan Abramowitz
The Pew Research Center has just released its final numbers on the 2004 presidential race and they show George Bush with a 3 point lead among “likely voters” but John Kerry with a 1 point lead among registered voters. Not to worry. There are good reasons for skepticism about those likely voter results.
First of all, here’s Andrew Kohut’s track record at Pew and, previously, at Gallup. In 2000, the final Pew Poll had George Bush leading by 3 points. We all know what actually happened. In 1996, Pew had Bill Clinton ahead of Bob Dole by 14 points–the actual margin was 9. In 1988, Kohut, then at Gallup, had George H.W. Bush ahead of Michael Dukakis by 12 points–the actual margin was 8. And in 1980, Gallup had Ronald Reagan ahead of Jimmy Carter by only 3 points in their final poll. Reagan won by 10. Notice that in all three of the elections involving incumbents, the final poll overestimated the incumbent’s support.
In addition, their equal allocation of undecided voters between the incumbent and challenger is contrary to the normal tendency of undecided voters to break in favor of the challenger by a wide margin.
And there’s more. The current Pew sample of likely voters consists of 37.3 percent Republicans and 33.6 percent Democrats. That’s right–a 3.7 percent Republican identification advantage. Of course we know that every national exit poll since 1988 has had a Democratic identification advantage of between 3 and 5 points. It is highly unlikely that in a very high turnout election in 2004, Republicans will outnumber Democrats.
Finally, here are a couple of numbers from the Pew Poll that are revealing. John Kerry is leading among independents by 48 percent to 44 percent and Kerry is also leading among those who have already voted by 48 percent to 47 percent.

Today’s tracking polls show more a mixed bag than yesterday in terms of movement favoring Bush or Kerry–which is the typical pattern, as Alan Abramowitz has noted. (You can find all the relevant data and links at NowChannel.com.)
Here are today’s results:
WaPo LV: 48-48 tie, from 49-48 Bush yesterday
WaPo RV: 48-47 Kerry, unchanged from 48-47 Kerry yesterday
Fox 3way LV: 46-46 tie, from 47-45 Bush yesterday
Fox 3way RV: 47-45 Kerry, from 46-46 tie yesterday
Zogby: 48-48 tie from 47-46 Kerry lead yesterday
Rasmussen: 48.1-47.1 Bush, from 47.9-47.1 Bush yesterday
TIPP 3way LV: 48-43 Bush, from 46-44 Bush yesterday
Note: over last two days, WaPo and Fox have moved steadily toward Kerry; TIPP steadily toward Bush.
Note: Today’s WaPo RV result makes 6 out of 7 days that Kerry has been ahead in their RV sample.
Note: At this stage of the election in 2000, Bush was ahead of Gore by 2 points in the Zogby tracking poll, by 4 points in the ABC/WP tracking poll and by 9 points in the TIPP tracking poll.
Note: All the polls, save TIPP, imply something very close to a dead heat. Chris Bowers of MyDD has a good analysis today of how such a result, if correct, fundamentally favors Kerry.