nimh wrote:- Moreover, the states [Hillary] would win [against McCain] and he wouldnt are bigger than the ones he would win and she wouldnt, so in the Electoral College she does a lot better than him according to this collection of polls. She would score a convincing victory against McCain, while Obama's stuck in a virtual tie. More about that in a minute, featuring maps.
Here are the maps to go with that.
These maps, again, represent the average of the polling out there on the match-up in question (Hillary vs McCain or Obama vs McCain) in that state from the period March-May.
There are of course a number of problems associated with this method. It weighs a March poll as heavily as a May poll, and a poll by an obscure local pollster as heavily as one of a reputed national one. Poblano's models at
www.fivethirtyeight.com are way more sophisticated in those respects.
Then again, it isn't, at least not in the case of Obama, as if there's been much dynamics in the race, and reputed national pollsters have also regularly gone off the mark this season.
Moreover, though, there are 14 states for which no match-up polling from the last three months are available. Luckily, they are overwhelmingly safe Democratic or safe Republican states, so I could just color them in accordingly. But a few are iffy: Delaware on the Dem side (where Kerry won by 8%) and West-Virginia on the Rep side (where, judging on how she's polling in Kentucky, Hillary would have a real chance). No recent polling for Maryland, Illinois, Hawaii, Louisiana and South Carolina either. But again, there's plenty of polling for the most hard-fought battleground states.
Anyway, here are the maps. Do compare
with those from last month. I have chosen, this time, to leave states where the average of polls has McCain or the Dem up with just 1% or less grey on the map, as true toss-ups.
Advantage Hillary (with the number of Electors between brackets):
Florida (27), Hillary would win, Obama would lose
Arkansas (6), Hillary would win, Obama would lose
Pennsylvania (21), Hillary would win, tossup for Obama
Ohio (20), Hillary would win, tossup for Obama
Kentucky (8), tossup for Hillary, Obama would lose
West-Virginia (5), tossup for Hillary, Obama would lose
New Hampshire (4), tossup for Hillary, Obama would lose
Advantage Obama (with the number of Electors between brackets):
Colorado (9), Obama would win, Hillary would lose
Wisconsin (10), Obama would win, Hillary would lose
Iowa (7), Obama would win, Hillary would lose
Michigan (17), toss-up for Obama, Hillary would lose
Indiana (11), toss-up for Obama, Hillary would lose
In both categories, Hillary's states "outweigh" Obama's.
It's also interesting, in the light of long-running speculations that an Obama nomination might entirely change the structure of the US electoral map, to see that Hillary would actually change it more - at least right now. Compare these maps with
the 2004 map (watch out for the colour reversal on that site):
- Compared with Kerry, Obama would win CO, IA and WI, and maybe IN or OH; but he would lose NH and maybe MI or PA.
- Compared with Kerry, Clinton would win NM, AR, OH and FL, and maybe KY or WV; but she would lose MI, WI and maybe NH.
None of which, obviously, is intended to say that Hillary should have been the nominee. There are a range of reasons to think that Hillary would have been a worse campaigner and thus less likely to win in the end, not to mention be a worse President.
But on the basis of recent state-level polls, as with national polls, she currently does have a legitimate claim to doing better against McCain than Obama -- one that she notably did not have earlier in the race. It still depends on the metrics used: an Obama supporter can still point to how Obama actually performs better against McCain in 25 out of 37 states that have been polled. But Hillary supporters can respond that her candidate does indeed do better where it counts most: in delegate-heavy states that she looks better placed to pick up from the Republicans than him, and that make all the difference in the Electoral College count.