Jacob Rosen has an interesting take on the candidates chances over at his blog The Sports Report. According to his math, if self identified Republicans vote for McCain in the numbers that they are currently polling at (79.5%), then Obama will only need to win self-identified Democrats by a margin of about 70.96%.
Now this level of statistics is way over my head (I just make the pretty pictures), and he seems to make a few assumptions (such as turnout) that may not end up being accurate. But even if his conclusion is off by 1 or 2%, the current political climate suggests that the final results will be close to what Rosen suggests.
Of course, his numbers do have one flaw... They only look at that interesting but fundamentally irrelevant number "Popular Vote". It would be very interesting to have a similar breakdown of each state. Unfortunately, with less than two months to go before the election, I doubt that he would have time for such a massive undertaking. But even with that flaw, his numbers are worth a look.
Subscribe to:
Post Comments (Atom)
1 comment:
Very good point you make about popular vote. I had previous tables on my site that analyzed state-by-state averages from FiveThirtyEight.com, and I should have some new tables coming up soon.
A friend asked me recently what I thought the margins of error on my graph were considering the possible changes to my spreadsheet that could occur come November. My guess is that +/- 1% could possibly occur. It is a logical and very well-documented assumption to state that more Democrats will vote than Republicans. Thus, the turnout effects cannot be too serious, with somewhere around 750,000 votes being up for grabs.
Thanks for referencing me!
Post a Comment