For as long as I can remember (which is a long time), the alpha and omega of UK election analysis has been the Swingometer. This is a BBC graphic that translates the shift in the national two-party vote--the swing--into parliamentary seats won and lost: a pendulum that sweeps across a list of constituencies ordered by size of the two-party majority. So simple. I recall Peter Snow gyrating idiotically beside ever more elaborately animated versions of this concept, election after election.
I've often wondered why I've seen no US version. (Maybe CNN has one. I don't watch CNN as often as I should.) The idea has obvious game-show appeal. And it's illustrative power is tremendous. Perhaps the idea of a roughly uniform national swing is unAmerican. Maybe it's a states' rights thing.
Anyway, the excellent Sean Trende alerts me to the fact that there is indeed a US swingometer, put together by Neil Stevens at Unlikely Voter. Trende, who understands these things, says that a uniform national swing from one party to the other "isn't a horrible assumption" because of something to do with PVIs. Give the thing a try. Plug in a swing and see what happens. Feel a bit British at the same time.
I see that Rasmussen's likely-voter tracking poll today has Romney at 49 and Obama at 45--a 4 point advantage. Obama's advantage over McCain in 2008 was 7 points. That's an 11-point swing. Plug it into the swingometer and Romney wins the electoral college 315 to 223, gaining Colorado, Florida, Indiana, Iowa, Minnesota, Nebraska (2nd District), New Hampshire, North Carolina, Ohio, Pennsylvania, and Virginia.
Well, now I see why Obama's people are so confident.