10/23/2004 10:54:12 AM|||Nathan Moore|||

Email from Bob Krumm

I'm having trouble understanding these internals vs. the national vote.
Blacks last time went 9% for Bush. This time polls are showing between 14%
and 18%. Jewish voters last time 20% for Bush, this time 30%. With women
Gore beat Bush by +11. Kerry's margin is +3. How then is Bush running only
3 points higher than he did last time (-.5% vs. +2.4%)? When you ask people
if they're certain to vote, 82% of them say yes. That's exactly the same
response as 4 years ago.

Therefore, if turnout is on par with 2000--and I think it will be closer to
that than to 60%--and Bush does better with blacks, women, and Jewish voters
than he did 4 years ago, that means Kerry has to pick up white male votes
that Gore didn't get. Haven't seen that breakdown, but can't imagine it's
happening. In short, I think these likely voter models are over-emphasizing
turnout of first-time voter turnout to keep the race looking close. If they
don't turnout in the numbers these polls assume, it's going to be a short
night.

I agree entirely. I've said before that the phenonomenon of first-time voters is being blown out of proportion. They either won't show (which is standard), or they'll show in equal numbers for both candidates. With the amount of effort being put into new voters on both sides, this would be the default, and the new voters would be a wash.

|||109854709200404763|||Good Point