Too-safe vs. too-safe.

[Shira]

A too-safe state is a state where a party can easily afford to lose 3% and still the state will remain a safe one.
In this respect the parties have wasted votes in these states. The total national number is going up without increasing the EVs. Needless to say that the Dems would have been very happy if 100,000 people had moved from DC to VA, and the Reps would have liked some migration from IN to OH for example, etc.

On this topic of a better distribution, the Reps have a big advantage.
Their too-safe states are WY, UT, ND, SD, NE, ID, MT, OK, AK, TX, KS, AL and KY
The total number of EVs is 93.
The too-safe states of the Dems, on the other hand are: DC, RI, MA, NY, CT, NJ, HI, MD, DE, VT, CA and IL and the total number of EVs is 168.

This, certainly diverts the numbers (more accurately – the conclusion from the numbers) in a national poll in favor of Kerry. In my estimate we have to make adjustments by subtracting 1% from Kerry and adding it to Bush.

Does anyone know whether there are national polls which exclude the too-safe states?

[Gabu]

I don't think there are any, though I might have just missed them.  I just spent a good ten or so minutes looking in Google and found nothing current.  I did find this, which lists the popular vote among the so-called battleground states (but which is dated April 22).

I personally wouldn't pay too much attention to the popular vote projections, anyway.  As has been shown numerous times in the history of the United States, it is not really a very good indication of how an election will turn out.  For example, in 1932, Franklin D. Roosevelt only won the popular vote by a margin of 11.9% but won the election with a margin of 413 electoral college votes (77.8%).  John F. Kennedy won the popular vote by 0.1% but won with a fairly comfortable margin of 84 electoral college votes (16.1%).  Four presidents have even lost the popular vote but still won the election.

[khirkhib]

Though it is true the difference in the Popular Vote against the electoral vote is already factored in.  Small states have more leverage, of course, because they have a higher voter to Electoral Vote ratio.

Wyoming.  The least populous state (including DC) has population of about 500000 and they have 3 electoral votes.  Now 500000 people don't vote in Wyoming of course but the electoral votes are decided of raw population.  So Every man, woman and child in Wyoming has about about .000006 of an electoral vote.  Fut it is pretty hard to see contrast in small numbers like that.
It is easier to think of it as for every about 166,666 people they get one electoral vote.

Where as in California with a population of  about 34 million they have 55 electoral votes or about .0000016 of a vote per person or one EV for every 618,181 people.  
So essentialy a vote in Wyoming is about 3 times more powerful than a vote in California.

So from the raw numbers the states have so many head per EV.

Wyoming                  164594
DC                            190686
Vermont                   202942
Alaska                      208977
North Dakota           214067
South Dakota          251615
Delaware                 261200
Rhode Island           262080
Montana                  300732
Hawaii                     302884
New Hampshire       308946
Mississippi                316073
Maine                       318731
Idaho                       323488
Nebraska                 342253
West Virginia           361669
New Mexico             363809
Nevada                   399651
Iowa                       418047
Arkansas                 445567
Utah                        446634
Kansas                    448010
Colorado                 477918
Connecticut            486509
Oregon                   488771
Minnesota              491948
Oklahoma               492951
Alabama                 494122
Louisiana                496553
South Carolina       501502
Kentucky                505221
Missouri                  508656
Arizona                   513063
Tennessee              517208
Massachusetts        529091
Maryland                 529649
Washington            535829
Wisconsin               536367
North Carolina        536621
Virginia                   544501
Georgia                  545763
Indiana                  552771
New Jersey            560957
Ohio                       567657
Michigan                 584614
Pennsylvania         584812
Illinois                   591395
Florida                   591940
New York               612144
Texas                    613289
California               615848

Now this clearly shows that their are red small states and blue small states that pack more of an EV whallop however if average it in terms of the current election using the Gore/Bush election as the template.

The average of the Bush states is 500449 people per EV and 541522 people per EV for the Gore states keep in mind that this is from raw population.  

Now the EVs in Bush's states have increased for this election by 10 votes because of demographic changes that have now been acounted for but from the vote number Bush recieved 271 EVs with a popular vote of 50,460,110 and Gore recieved 266 EVs and 51003926 votes.  So each 186889 votes was worth 1 EV for Bush and each  191744 votes was worth 1 EV for Gore.  Essentially a vote for Gore was worth only 97% of a vote for Bush.  

But those are the rules and you have to use what you got.  What it means however though is since a vote for Gore was worth only 97% of what a vote for Bush is than Kerry has to get a measured precentage higher than Bush to win the electoral college.   Suppose for a second that their are no third parties and the EVs don't come in clumps but are more directly related to popular vote.  Though genearlly with a popular vote the winner would need 50% + 1 to win.  But because Kerry's votes per EV are worth somewhat less than Bush's he would about 52.5% of the Vote to win the 269 votes that he would need in the electoral college.

Now you can put third parties and everything back in the equation of equations and states .  Basically their are two bars for the election Bush needs about between 47% - 50% to win and Kerry needs between 50% - 52% to win.  Kerry has a bigger challenge but with Polls  you don't need them to calculate and redistribute you just need to know that for Kerry to win he needs to be polling 3% or better than Bush to win.  If Kerry and Bush are polling dead even on election day or Kerry has even just a slight advantage than Bush wins.  For Kerry to win this election, which he will, he needs to blow Bush out of the water.

[Gabu]

Regarding Shira's question about national polls that exclude the "too-safe" states, what you could do is go to a site like this one, write down the percentages for each of the states that are not the "too-safe" states (if you scroll over the states in the picture on that site it'll tell you who's leading and by how much), and then average them out to get a pretty good idea of what you're looking for.

Just a thought.

[Gustaf]

This is heavily exaggerated. When Gore won the PV in 2000 he was a grand total of 0.01% in Florida from also winning the EV. So the difference is very marginal, at best.

[Shira]

The derived conclusion is that after the Nader‘s and “others’” votes are dumped,
Kerry: 50.4 vs. Bush: 49.6 will make a tie in the EV.
Take any reasonable combination that makes a 269:269 EV tie, and you’ll see that the number of citizens (and consequently voters) in Kerry’s states are significantly higher then in Bush’s states.
 An example of a reasonable tie is: Kerry wining all Gore’s states + NH + NV