As I've been saying since February, I think that the great question of the 2008 campaign will be whether Barack Obama can unite Democrats to a great enough degree to win. (Corollary to that is the question of whether John McCain can unite Republicans enough to take advantage of any Clinton-Obama fallout, and to overcome the massive partisan advantages of Democrats this year.)
Real Clear Politics takes a stab at answering the question of how many Clinton supporters will support McCain, using current (very early!) polling:
Luckily, we have some data from Quinnipiac's recent swing state survey that gives us some indication of how Clinton primary voters in Florida, Ohio, and Pennsylvania are feeling. As you can see, Florida closely mirrors the national numbers but the trend is more pronounced in Pennsylvania and especially Ohio, where the net change (drop in support for Obama/increase in support for McCain) reaches 22 and 29 points respectively:
Again, it should be noted that Obama increased his lead over McCain in every Quinnipiac state poll, indicating that despite whatever lingering fallout exists with Clinton supporters, Obama is more than making up the difference with gains among other groups.
Still, it should give the Obama campaign a bit of a pause to see that, at least for the moment, one in four Democrats who voted for Hillary Clinton in the primaries in both Ohio and Pennsylvania indicate they will crossover to vote for McCain in the fall.
I really need to sit down and do my own analysis on this. I keep waffling on whether Obama is doing a great job winning Clinton supporters and I just haven't had a chance to compile and drill down into any real data yet. But I wanted to share this anyway.
Going off the 2004 exit polls it looks like roughly 11% of Dem voters crossed over to vote for Bush... roughly unchanged from 2000. Repubs generally split off at around 6%... a little lower than in 2000 from what I can tell.
I could see that trend reversing this year... or having higher cross over on both ends depending on how all of these first time voters vote, if they turn out much at all. While politically the last breakdown was roughly even for dem/repub voter turnout with independents generally evenly split as well, it looks like we'll be seeing a lot more Dem voters than Repub this year... the hitch is when it comes to ideology. There were far more self-described conservatives than liberals in 2004... and nearly twice as many self-described moderates. Kerry still won over a larger share of the moderates in 2004... but it could be much more difficult for Obama to appeal to moderates, especially in contrast with the guy who is notorious for being one.
It'd still be surprising if Obama lost given all of the other advantages going into the race. But this election has generally been surprising all around... so I guess I'd be surprised, but not shocked (does that even make sense?) if he snatched defeat from the jaws of victory. I doubt too many dems would agree with me though as few seemed to think Richardson was the ideal candidate for the times for their political perspective. To me that was their first step towards making an unlikely defeat far less unlikely.
--
Glock21 Op/Ed
The problem with these numbers is that they may not really differentiate just how open or closed the primary is, or the interest in the race at the time, or how Florida is a train wreck of flawed elections (the Dem primary this year being no exception). For example, I'm an independent, but Illinois has a pretty open primary and it's not like the election was really close on the Dem side. This was, however, before McCain wrapped it up, and I liked him more than Mitt, Huck, etc., so I declared GOP and voted for McCain. In Pennsylvania, however, you need to declare your party ahead of the election when registering / getting your drivers license, so it's far more closed. That said, they had that situation where they were being hyped as the state that was going to finally decide the Dem race (whoops!), and that may have driven indy / GOP voters to register as Dems. So barriers to entry were high, but so was interest, as opposed to Illinois, where barriers were low, and interest was mainly in a high turnout to make the local guy look good. How many of these "new Dems" are going to stay Dems? Who knows? We don't even know how many were doing it just to be able to vote in an interesting election as opposed to people that are really Dems and just got the extra measure of motivation. I know that I don't feel more Republican, even though I suppose that I'm technically a Republican at this point.
There is one measure that should be useful - whenever people say that they are all ticked off and that they won't vote for eventual primary winner (Ford, Carter, Gore, Bush in 2000, Kerry), they almost all come around in the end. How many true believer Dems will vote for McCain? More than I can count on my fingers and toes, but fewer than will actually make a difference.