2004 Electoral Snapshot

[Stuckinthebush]

I have started to collect state level polling data to track the 2004 electoral vote possibilities. Every so often, I will update a spreadsheet and post it here. As we know too well, the general election hinges on electoral votes not popular votes, so it is imperative that we track the electoral possibilities.

The spreadsheet linked below shows each state with electoral votes, 2000 presidential results, and 2004 possibilities based on current polling trends.

For the 2004 data, there are four columns - Dem, GOP, Tossup, and Source. If the polling data show that the race is within the margin of error, then the electoral votes go in the tossup column. If there is no source listed, the electoral votes are based on the 2000 results. As we get further into the year, we should see polling data from most states.

I only used results based on direct horse race questions, therefore, there is no assumption based on Bush favorability ratings. If there was no horse race question for any poll, then the 2000 results were used.

As it stands at this slice in time, the possible electoral votes look like this:

DEM - 254
GOP - 227
Tossup - 57

If you have any information on state polls not shown, please PM me and I'll include it.

Feedback is encouraged! Have fun and keep it bumped up for a bit.

Thanks!

2004_Electoral_Snapshot

[mdmc]

1. bookmark

:dem:
thanks!

[Stuckinthebush]

2. No problem...

let me know what you think and how it can be improved.

[LeftCoast]

3. Perused it quickly

Seems to make sense, though I was a little surprised at a few of the "toss-ups". Not challenging your data, just surprised. :)

[beyurslf]

4. Very nice. I think we will see a lot of shifting in that in the coming

months with many more toss up states in play. I would say we will have as many as 10-15 states close and in play.

[Stuckinthebush]

15. I agree

I look forward to more state polling data to see how we stand.

[Sagan]

5. I still don't see Ohio going Repug this year....

That's 21 and the win right there.

[Stuckinthebush]

16. That would be sweet

Do you have any information on Ohio polling, Sagan? I couldn't find any recent horse race data.

[mobuto]

6. A few oddities

1. KY is are definitely not in the Democratic column. It could be a toss-up, but if anything, I'd say its leaning Bush.

2. LA is definitely a tossup.

3. FL is probably also a tossup, but you may be right to say its leaning Bush.

4. MO is a tossup.

5. SC is no tossup. Bush.

6. West Virginia and Arkansas. Leaning Democratic, but call them tossups. Definitely not for Bush.

7. AZ is now a tossup.

8. Ohio is a tossup.

9. Indiana is a tossup.

10. New Hampshire is a tossup.

[Stuckinthebush]

10. I'd agree on some of those

but I'm basing the determination on polling data. So far, many of these states don't have up to date polling information that I have found, so I am reluctant to make a determination.

If no polling data are available, I am using the 2000 results for the state with no data.

[Perky]

7. A great link

http://www.uselectionatlas.org/USPRESIDENT/GENERAL/CAMPAIGN/2004/pred04.php

[Brian Sweat]

8. baleted

[City Lights]

9. Arizona has 10 electoral votes, not 8.

Did you update your numbers after the 2000 census?

[Stuckinthebush]

11. Thanks - I got these from the electoral web site

I'll go in and make sure the new electoral votes are in there.

[City Lights]

13. Try this link.

I didn't check other states, but it shows 10 for Arizona.

http://www.thegreenpapers.com/G04/ElectorAllocation.phtml?sort=Alph

[Stuckinthebush]

12. Recalculation based on 2000 census

Thanks DesertDem...

The new totals based on current polling data are:

Dem: 247
GOP: 232
Tossup: 59

[City Lights]

14. You're quick.

;-)

[Stuckinthebush]

17. One thing that got me excited

Was the Shapiro poll in Georgia showing a dead heat between Bush and Dem.

I won't get too excited just yet...but I'll remain hopeful...

[Dems Will Win]

18. You can play with this cool interactive map!!

http://www.johnedwards2004.com/map/

[snippy]

19. Interesting and bookmarked.

I notice you have no polls from SurveyUSA. Is that deliberate? I also am curious how you apply the margin of error.

[Stuckinthebush]

20. Excellent!

I'll update based on some of their results. Thanks!

As for the MOE - it depends on the poll in question. If the results of the horse race overlap based on the MOE, then I consider it a toss up.

[snippy]

22. Do you apply the MOE to each candidate's percentage?

I have a vague recollection that the entire MOE can't be applied to each candidate's percentage for a mathematical reason which avoids "overlapping." It has been many years since I took statistics and I could be wrong. The number that sticks in my mind is that something like 1.8 times the MOE, rather than 2 times, is more accurate when determining whether the results are within the MOE.

An example would be Bush 54% Kerry 46% with a MOE of 4%. I think this is actually slightly outside the MOE with the numbers being Bush 50.4% and Kerry 49.6% after applying the MOE to each. But like I said I could be wrong, just ask my stats instructor.

[Stuckinthebush]

23. That rings a bell...

I'll check into it.

Thanks!

[Poiuyt]

21. Excellent - One Suggestion

It would be nice if you included the date of the most recent update so we can see how current it is.

Thanks for your work. I've been looking for something like this.

[Stuckinthebush]

24. Good suggestion

I'll put an update date at the top of the page.

[woo me with science]

25. Thank you very much for this.

I'm bookmarking it, too, and will be looking forward to updates. Great work.

[Jersey Devil]

26. New Jersey a tossup? Not nearly

If the Dem nominee loses NJ it is all over. IMO Bush doesn't have a prayer in NJ.

[Stuckinthebush]

27. Probably not

but it is based on recent state level polling data.

[Stuckinthebush]

28. Updated to reflect Pduck's and Snippy's suggestions

Thanks!

The SurveyUSA poll numbers give us different totals now, but they will be ever changing anyway.