Electoral College Prediction and Analysis (long and detailed)

[Dynasty]

I thought it would be fun to make a comprehensive election prediction and post it. For those who only want the final result, I'm going with Bush 286, Kerry 252 with thirteen states being decided by 4% points or less (147 Electoral College votes).

Clearly, the country is divided about 50-50 right now and either candidate can win. The divide is especially noticable in the Red State-Blue State nature of the Electoral College. Very few states should switch sides compared to the 2000 election.

This election would be a blowout if it weren't for one decision by President Bush. That's the decision to invade Iraq. At the time of invasion, Bush was still enjoying job approval rates in excess of 70%. If Bush had simply played it politically safe, I think he'd cruise to a small landslide victory (56-44). But, Bush, unlike most Presidents before him, seemed much more willing to spend "political capital" and push a neo-conservative agenda when the country wasn't ready to go so fast. From a political perspective, Bush energized both his own base and the Democratic base whose charge of "Anybody but Bush" has resonated for nearly two years.

Bush's ideological approach to government was by design not only to further his agenda but also to get re-elected. I think the dynamics of the campaigns have been shaped by Karl Rove, Bush's campaign manager and senior political advisor. After the 2000 election, Rove embarked on a re-election approach which was radically different from anything we've seen before. Rove's 2000 post-election analyis showed that nearly four million social conservatives stayed home instead of voting (presumably for Bush). Rove decided not to expand the President's constituency by appealing to moderate voters which is the standard approach. Instead, he wanted to appeal to the Republican base and motivate a big turnout. That apporach was hugely succesful in the mid-term 2002 elections which put the Senate back in Republican control. In 2004, it's going to butt heads with the "Anybody but Bush" army the Democrats have behind them.

The winner of this election is going to be the side which turns out the vote better. I'm giving the nod to Bush and Rove since the voters they targeted are more likely to vote and more likely to vote for Bush.

Here's my state by state breakdown. I tried to give a % vote to third party candidates where appropriate.

18 completely safe Bush states (144 Electoral College votes):

Utah (5) 69-29-2
Wyoming (3) 69-30-1
Idaho (4) 63-34-3
Nebraska (5) 63-35-2
Oklahoma (7) 62-38-0
North Dakota (3) 61-38-1
Kansas (6) 61-39-0
Alaska (3) 58-37-5
Montana (3) 59-38-3
Texas (34) 60-39-1
Kentucky (8) 60-40-0
Alabama (9) 59-41-0
Indiana (11) 59-41-0
South Carolina (8) 58-41-1
South Dakota (3) 57-42-1
Mississippi (6) 57-43-0
Georgia (15) 56-44-0
Tennessee (11) 56-44-0

10 completely safe Kerry states + D.C. (153 Electoral College Votes):

District of Columbia (3) 84-12-4
Massachusetts (12) 65-32-3
Rhode Island (4) 59-38-3
New York (31) 48-40-2
Illinois (21) 56-43-1
Connecticut (7) 55-43-2
Maryland (10) 55-44-1
Vermont (3) 54-43-3
California (55) 54-44-2
Delaware (3) 54-45-1
Maine (4) 54-45-1

I can't envision any scenario where the 28 above states and D.C. don't go to the candidate I have them listed under.

Current tally: Bush 144, Kerry 153.



6 safe Bush states (61 Electoral College votes):

Arizona (10) 54-46-0
Louisiana (9) 54-46-0
North Carolina (15) 54-46-0
Colorado (9) 53-46-1
West Virginia (5) 53-46-1
Virginia (13) 53-47-0

3 safe Kerry states (33 Electoral College votes):

New Jersey (15) 53-46-1
Washington (11) 52-46-2
Oregon (7) 52-47-1

The 9 safe sates are only going to switch colluns if something strange happens. Kerry would need a completely unprecedented colossal turnout to challenge Bush's 6 safe states. Bush would need the same or a strange Nader effect to win in Washington or Oregon.

New Jersey is something of an oddity this year. Bush has gained and Kerry has been hurt by three issues.

1. Many New Jersey families lost loved ones in the 9/11 attacks.
2. The Republican convention got huge local coverage since it was in New York.
3. The New Jersey state Democratic party is mired in scandal, including a Governor who is resigning from office.

Gore won New Jersey by 16 points. These three factors shouldn't be enough to eliminate such a huge starting advantage for Kerry.

Current tally: Bush 205, Kerry 186



3 semi-safe Bush states (22 Electoral College votes)

Arkansas (6) 52-48-0
Missouri (11) 52-48-0
Nevada (5) 51-47-2

2 semi-safe Kerry states (14 Electoral College votes)

Minnesota (10) 51-47-2
Hawaii (4) 51-47-2

Kerry made a late bid for Arkansas by spending some campaign funds on TV advertising there. Also, President Clinton campaigned there on Monday. But, no poll has shown Kerry up in that state. Usually, the polls have Bush up by 5 points or more but one showed a tie.

Kerry gave up on Missour early, probably too early. If Dick Gepardt had been his Vice Presidential running mate, this could have been a real fight. This seems like the one state where Kerry dropped the ball.

Nevada and the southwest in general was where Kerry hoped to pick up some Bush states rather than going to the traditional south. But, this is still Republican territory and Kerry is the wrong Democrat to make gains here.

Minnesota is a long time Democratic stronghold. It was the only state the didn't vote for Reagan in 1984. But, the demographics of the state have been changing. When they elected Jesse Ventura as Governor in 1998, Minnesotans demonstrated they're very willing to break from tradition. In four years, the Republicans might have this state.

Hawaii is the big late surprise. Everyone had it as a completely safe Kerry state until two recent polls showed Bush with a 1% point lead. I'm giving it to Kerry based on historic trends. But, Hawaii did elect a Republican Governor in 2002. So, maybe I'm just unwilling to change.

Current tally: Bush 227, Kerry 200



That leaves us with the Crazy Eight Battleground states worth a total of 111 Electoral College votes. This is where I expect the election to be decided. In truth, all these states are not really predictable. The public polls and the actions of both campaigns have made it clear all these states were in play to the very end. But, since a prediction has to be made, here I go.

4 battleground Bush states (59 Electoral College votes):

Iowa (7) 51-48-1
New Mexico (5) 50-48-2
Ohio (20) 51-49-0
Florida (27) 50-49-1

4 battleground Kerry states (52 Electoral College votes):

Michigan (17) 51-48-1
Pennsylvania (21) 51-49-0
New Hampshire (4) 50-49-1
Wisconsin (10) 50-49-1

What can be said? The public polls have all these states within the margin of error. Both campaigns have been visiting these states in the last days of the campaign. They'll be won or lost based on turnout.

In the end, I only have three states changing collumns. Kerry is picking up New Hampshire while Bush is picking up Iowa and New Mexico.

Final tally: Bush 286, Kerry 252

A lot of these battleground states are in the east so we could know who wins the election before 11pm EST. But, if it's very close, we may have to wait hours after polls have closed in New Mexico, Nevada, and even Hawaii before the networks can make a projection.

[andyfox]

Terific analysis, clearly argued and written.

[lorinda]

Thanks Dynasty, nice post.

Lori

[J.A.Sucker]

I think you are correct. I hope you aren't thought. We'll see tomorrow (maybe...)!

[LinusKS]

You're missing two important elements: the incumbent effect, and new voters.

Historically, challengers pick up around two points when voters actually go into the booths. Pollsters tend to explain this in terms of undecideds breaking for the challenger (or against the incumbent, however you want to look at it).

Also, large turn-outs tend to favor Democrats. Small turn-outs tend to favor the incumbent.

Kerry has both working in his favor.

In FL, for example, Gallup shows early voters (30% have already voted) breaking for Kerry 51%-43%.

http://www.usatoday.com/news/politicselections/nation/polls/2004-10-31-poll-x_x.htm

[LinusKS]

I'm guessing Kerry will win both FL and OH.

The chances that Bush will win both of them - I think - are remote.

[vulturesrow]

[ QUOTE ]
Historically, challengers pick up around two points when voters actually go into the booths. Pollsters tend to explain this in terms of undecideds breaking for the challenger (or against the incumbent, however you want to look at it).

[/ QUOTE ]

People keep saying this as if it were gospel. Its not. It didnt break that way in the last 3 incumbent re-elections. In the last 3 elections it didnt break this way.

As for the large turnout, I think people are also too willing to give this as a major advantage to Kerry. Im not so sure. In the 2002 election cycle, the GOP 72 hour GOTV campaign paid huge dividends. Also many people expect a much larger showing among evangelical Christians. However, I dont have data handy on the history of these so I will concede the point.

[largeeyes]

[ QUOTE ]
I'm guessing Kerry will win both FL and OH.

The chances that Bush will win both of them - I think - are remote.

[/ QUOTE ]

Actually, the odds are about 37-40%

[Nepa]

[ QUOTE ]
Also many people expect a much larger showing among evangelical Christians.

[/ QUOTE ]

I think you might want to replace the word expect with the word hope or pray for.

[Rick Nebiolo]

Dynasty I'm going to echo the chorus but this is great stuff. Better than anything I've seen by the major pundits.

My fear is that the several states will be close and we end up with Florida 2000 times 6.

Belaboring the obvious but that won't be good for the nation /images/graemlins/frown.gif.

~ Rick

[vulturesrow]

[ QUOTE ]
I think you might want to replace the word expect with the word hope or pray for.

[/ QUOTE ]

Hmmm, let me think about it......nope I dont want to. The GOP knows this group didnt show up in 2000. They are confident this year that they will. Is it a certainty? Close to it I think.

[Nate tha' Great]

[ QUOTE ]
[ QUOTE ]
I think you might want to replace the word expect with the word hope or pray for.

[/ QUOTE ]

Hmmm, let me think about it......nope I dont want to. The GOP knows this group didnt show up in 2000. They are confident this year that they will. Is it a certainty? Close to it I think.

[/ QUOTE ]

Indeed, it is close to a certainty that the GOP is confident that the evangelicals will turn out.

[banditdad]

[ QUOTE ]
This election would be a blowout if it weren't for one decision by President Bush. That's the decision to invade Iraq. At the time of invasion, Bush was still enjoying job approval rates in excess of 70%.

[/ QUOTE ]

Good analysis. However I disagree with the above. The only thing keeping Bush in the race is his invasion of Iraq and the war on terrorism. If there had been no 9/11 attack Bush would be far behind at this point. I also think that if Bush had kept his eye on the ball after 9/11 and gone after Bin Laden rather than a foolish invasion of Iraq based on neo-con fantasy he'd be way ahead at this point.

Of course my crystal ball is no better than the next guys.

[SinCityGuy]

[ QUOTE ]
This election would be a blowout if it weren't for one decision by President Bush. That's the decision to invade Iraq.

[/ QUOTE ]

Colin Powell got overruled by Dick Cheney and Donald Rumsfeld.

[LinusKS]

[ QUOTE ]


People keep saying this as if it were gospel. Its not. It didnt break that way in the last 3 incumbent re-elections. In the last 3 elections it didnt break this way.

[/ QUOTE ]

Yes, it did:

[ QUOTE ]
There have been four incumbent presidential elections in the past quarter-century. If we take an average of the final surveys conducted by the three major networks and their partners, we find that in three of these the incumbent fell short of or merely matched his final poll number, while exceeding it only once, and then by just a single point (Ronald Reagan). On average, the incumbent comes in half a point below his final poll result.

The numbers for challengers look quite different. In every case, the challenger(s) -- I include Ross Perot in 1992 and 1996 -- exceed their final poll result by at least 2 points, and the average gain is 4 points. In 1980, Ronald Reagan received 51 percent, fully 6 percentage points above his final poll results.

[/ QUOTE ]

That's an average net gain of 4.5% for the challenger.

http://www.prospect.org/web/page.ww?section=root&name=ViewWeb&articleId=8694

[mikech]

[ QUOTE ]
The winner of this election is going to be the side which turns out the vote better. I'm giving the nod to Bush and Rove since the voters they targeted are more likely to vote and more likely to vote for Bush.

[/ QUOTE ]
The first statement is spot-on. The second, well, I'm not sure I understand why you think that. I don't think that "the evangelicals who didn't show up in 2000" (not your phrase, a subsequent poster's, but the same group you were pointing to, I believe) are an X-factor, they're not an unknown quantity; they've been sampled and represented in the polls as they stand--and even so, the candidates are in a virtual dead heat. So the polls already take into account those Bush supporters; however, the polls under-represent new and young voters. Zogby conducted a survey of 18-29 year-olds who exclusively use cell-phones (without landlines, this demographic was entirely ignored in other polls), and found an overwhelming 55%-40% advantage for Kerry.

And then there's that 8-point lead Kerry currently enjoys in Florida after THIRTY-PERCENT of ALL registered voters have already voted. I wonder how Bush is going to overcome that...

[SinCityGuy]

[ QUOTE ]
I think you are correct. I hope you aren't thought. We'll see tomorrow (maybe...)!

[/ QUOTE ]

I disagree with Dynasty, although his analysis is extremely well written.

Here's why I disagree. In the ten most recent national polls, President Bush is not at 50% in any of them. Not a single one. For an incumbent president, this is very troubling on the eve of the election.

But, as we know, the election will be settled in nine battleground states. In those states, there have been 57 polls in the past four days. President Bush is above 50% in 3 out of those 57 polls. Once again, this is very ominous for an incumbent.

I think Kerry gets there with over 300 electoral votes.

[mikech]

[ QUOTE ]
I think Kerry gets there with over 300 electoral votes.

[/ QUOTE ]
I think the chances of Kerry not simply winning, but winning going away, are looking better and better.

[MMMMMM]

As usual.

[largeeyes]

[ QUOTE ]
[ QUOTE ]
I think you are correct. I hope you aren't thought. We'll see tomorrow (maybe...)!

[/ QUOTE ]

I disagree with Dynasty, although his analysis is extremely well written.

Here's why I disagree. In the ten most recent national polls, President Bush is not at 50% in any of them. Not a single one. For an incumbent president, this is very troubling on the eve of the election.

But, as we know, the election will be settled in nine battleground states. In those states, there have been 57 polls in the past four days. President Bush is above 50% in 3 out of those 57 polls. Once again, this is very ominous for an incumbent.

I think Kerry gets there with over 300 electoral votes.

[/ QUOTE ]

300? That leaves bush with at, 240? The odds on that are like +260. If you are serious and think that, plenty of money to be made.......but you just threw that out there and don't really buy it, right?

[largeeyes]

All these people who keep saying Kerry will win going away, put your money where your mouth is. I did.

[Moozh]

I think this is a very well thought out and clear analysis. I am in complete agreement up until the final 8.

Of the 8, I strongly agree with the top two for each candidate. I would be surprised if Bush lost either Iowa or New Mexico at this point. I would also be very surprised if Kerry lost Michican ot Pennsylvania. This puts our candidates at Bush 239, Kerry 238.

That leaves New Hampsire (4), Wisconsin (10), Ohio (20) and Florida (27). It's effectively two little states and two big states.

If either candidate wins both big states, they win. If they win both little states, one big state will be enough.
The combo of NH and Florida for Kerry, Wisconsin and Ohio for Bush gives us the dreaded tie. Swap the states and you get Bush 270, Kerry 268.

Based on the recent polls, I'm guessing Wisconsin will go to Kerry. Thus, that leaves it that if Kerry wins Florida, or if he wins New Hampshire and Ohio, he wins. As a democrat, I'm heartened to see this.

[Dynasty]

[ QUOTE ]
[ QUOTE ]
The winner of this election is going to be the side which turns out the vote better. I'm giving the nod to Bush and Rove since the voters they targeted are more likely to vote and more likely to vote for Bush.

[/ QUOTE ]
The first statement is spot-on. The second, well, I'm not sure I understand why you think that. I don't think that "the evangelicals who didn't show up in 2000" (not your phrase, a subsequent poster's, but the same group you were pointing to, I believe) are an X-factor, they're not an unknown quantity; they've been sampled and represented in the polls as they stand--and even so, the candidates are in a virtual dead heat.

[/ QUOTE ]

It's not fair to say they've been represented in the polls accurately. Polls, like Zogby, weight their results by using party demographics. It's usually a 39-D, 36-R, 25-I breakdown. But, if the Republicans had a massive get-out-the-vote effort and the Democrats didn't have a comparable one, then the 39-36-25 weighting would fall apart. The actual Republican % would go up while the Democratic % would go down. So, while they may be getting sampled more often, their greater #'s (if the turnout is big) isn't being represented.

[Dynasty]

[ QUOTE ]
Of the (battleground) 8, I strongly agree with the top two for each candidate. I would be surprised if Bush lost either Iowa or New Mexico at this point. I would also be very surprised if Kerry lost Michican ot Pennsylvania.

[/ QUOTE ]

I understand what you're saying. The public polls have usually put Kerry ahead in Pennsylvania (http://www.realclearpolitics.com/Presidential_04/pa_polls.html) and Michigan (http://www.realclearpolitics.com/Presidential_04/mi_polls.html) by 2-4 points. But, the behavior of the two campaigns made me wonder what is going on. The internal polls of the campaigns are supposed to be much better than the public polls. The Kerry campaign supposedly thought they had Pennsylvania wrapped up. But, 7-10 days ago, they saw something which worried them and decided to continue to devote resources in Pennsylvania when they thought both sides would be taking them out.

The same goes for Michigan. The Kerry campaign had to call on Al Sharpton to make some "emergency" campaign stops in Detroit (where Sharpton is from) to rally the black vote like Clinton did in Philidelphia. Meanshile, the Bush campaign visted Michigan over and over in the final week.

So, I projected those states conservatively but still gave them to Kerry.

[ QUOTE ]
Based on the recent polls, I'm guessing Wisconsin will go to Kerry.

[/ QUOTE ]

The polls in Wisconsin (http://www.realclearpolitics.com/Presidential_04/wi_polls.html) have been crazy. Zogby has Kerry up by +6. But, CNN/USA Today/Gallup has Bush up by +8. The other reliable posters (FoxNews/Opinion Dynamics, Mason-Dixon, ARG, and a local paper) all have +3/2/1 figures but for different candidates.

In September, this look like the best candidate for a Bush Blue State pickup. So, I was very reluctant to give it to Kerry at all. In the end, I went with Mason-Dixon who probably has the best track record of state polling. They have Kerry winning by 2.

[AleoMagus]

I agree...

Oh, except that I see Kerry taking Florida.

So that's 279 for Kerry
and 259 for Bush

Lets hope so anyways

Regards
Brad S

[Dynasty]

John Zogby has posted his Electoral College prediction on his website. But, the wimp is projecting Bush 247, Kerry 264, and a tie in Florida. (http://www.zogby.com/news/ReadNews.dbm?ID=923) C'mon. You can't project a tie.

Zogby and I agree on all states except Iowa, New Mexico, and Florida. Zogby gives Iowa and New Mexico to Kerry.

The only state Zogby predicts will be different than 2000 is New Hampshire which went for Bush in 2000 but is projected for Kerry in 2004. His tied Florida prediction is basically the same as 2000.

Zogby will make his final predicions on Election Day at 5pm EST.

[Mason Malmuth]

Hi Dynasty:

I enjoyed reading this.

One note. I watched the Beltway Boys on Fox last night and while they predicted Bush would win the election, they had Ohio going to Kerry.

Best wishes,
Mason

[El Barto]

Good analysis and I agree with it. The only difference I had (posted in Nate's prediction thread) was Wisconsin, but that is the one state that seems to have changed the most in the past 3 days and really does seem to be a toss up now.

[Toro]

Great analysis. I know you wanted to do a post where you made an actual prediction and it would have been whimpy to call the election a toss up but you should have called Florida "too close to call". Again, it will come down to Florida. Whoever wins Florida, wins the election.

[ericd]

This was really well done.

The only area I question is the voter turnout. My understanding is that "getting the vote out" is a long time Democratic strength. The Democratic effort is run by paid professionals whereas the Republican is not. Also, the Democrats have a much larger base to draw from.

Regardless, I truly hope that whichever way it goes the winner is clear cut.

[Nicholasp27]

[ QUOTE ]

It's not fair to say they've been represented in the polls accurately. Polls, like Zogby, weight their results by using party demographics. It's usually a 39-D, 36-R, 25-I breakdown. But, if the Republicans had a massive get-out-the-vote effort and the Democrats didn't have a comparable one, then the 39-36-25 weighting would fall apart. The actual Republican % would go up while the Democratic % would go down. So, while they may be getting sampled more often, their greater #'s (if the turnout is big) isn't being represented.

[/ QUOTE ]


actually, most of the polls have OVERSAMPLED Republicans if u look at their internals (including Gallup, Fox News, etc) and they still show a dead heat...so if Dems/Reps come out in same numbers, then Kerry would push ahead slightly...if Dems/Reps follow normal proportions that u listed, then Kerry could win by a decent margin

[Analyst]

To add to the chorus, outstanding job and a good read. To add to a somewhat smaller chorus, I hope you're wrong.

The election hinges almost exclusively on Ohio and Florida. If Bush wins both, you can turn off the TV and head to bed (or the card tables). If Kerry wins either state, then it would take some unusual - though not impossible - circumstances for Bush to win.

As a Michigan grad, it is pretty scary that the election may very well depend on how many Ohio State students show up to vote!

[Analyst]

[ QUOTE ]
Hi Dynasty:

I enjoyed reading this.

One note. I watched the Beltway Boys on Fox last night and while they predicted Bush would win the election, they had Ohio going to Kerry.

Best wishes,
Mason

[/ QUOTE ]

If Ohio does go to Kerry, then I think he has a very strong chance to win the election regardless of Florida.

[TomCollins]

I bet you think Kerry will take Texas too.

[TomCollins]

The beltway boys had Minneosta, Wisconsin, Iowa, and NM going to Bush.

I don't think he has a shot at Minnesota.

[WDC]

if either wins both florida and ohio its over

[wacki]

Do you still think your prediction is good?

[andyfox]

Looks like Dynasty may have scored a bullseye.

[goldcowboy]

Damn, you are good. What are the chances of my getting you to analyze the cattle and grain markets for me?

[Dynasty]

[ QUOTE ]
Looks like Dynasty may have scored a bullseye.

[/ QUOTE ]

Damn. If the current numbers hold up, I'm going to go 50 for 50. Iowa is really close, though.

[MEbenhoe]

[ QUOTE ]
[ QUOTE ]
Looks like Dynasty may have scored a bullseye.

[/ QUOTE ]

Damn. If the current numbers hold up, I'm going to go 50 for 50. Iowa is really close, though.

[/ QUOTE ]

Congratulations Dynasty. Awesome analysis, and as of right now its looking like you got them all. Maybe you;ve got a future career in political analysis? You did much better than anybody on the news stations did.

[Doubling12]

I printed this and had it with me while watching. You absolutely nailed it. Thanks for all the time and effort this took!!!! Bravo!!!!!

[El Barto]

I also congratulate Dynasty for getting it right on target.

Note also that the pollsters as a group did very well too. The Real Clear Politics Average turned out quite reliable state by state.

Zogby however zig zagged at the end, then crashed and burned just like 2002. I hope this is the end of the nonsense praise of him.

[Men the Master]

Do they award Pulitzer Prizes to internet posts? This post sure deserves one.

[goldcowboy]

Absolutely. Dynasty, how about using your analytical skills and insight to assess what changes are coming in the Bush cabinet?

[Men the Master]

Powell bye bye.

[goldcowboy]

Maybe so, but I think Rummy heads for the exits too.

[ilya]

Do you guys realize that Dynasty got it almost exactly right? Bow down to his wisdom!