It is a seductive thought for anyone who has spent days/weeks/months trying to run all of the possibilities for who could win the early states and how, only to be left with a three-way deadlocked race and too many variables to predict the sunrise with any confidence. If Hillary wins, does that mean more than if Obama finishes the deal? If Edwards surprises the pollsters, does that matter? Stack the outcome any way you want, does it really matter who wins 35% or 40% of Iowa’s handful of convention delegates, and who only gets 30%?
Maybe it’s an epiphany of convenience, but late last night I decided the answer was no, this year it does not matter who wins Iowa.
No, this year, the ultimate question is who loses. Tell me who is going to come in third next Thursday, and I think I can tell you who will win the Democratic nomination.
More after the jump...
Can the Democratic nomination still be a competitive, three-person race after Iowa? I don’t think so.
John Edwards obviously has to win in Iowa, and win pretty big, to give him a chance in New Hampshire. A narrow victory won’t be enough to dethrone both Hillary and Obama from their lead in New Hampshire. Any second-place finish will, at best, put him on a repeat of his 2004 path where a close second in Iowa wasn’t enough to lift him past fourth in New Hampshire. Without a win in New Hampshire he won’t get any traction in Nevada or South Carolina. For Edwards, his campaign is Iowa. If he finishes third, he doesn’t even need to go to New Hampshire.
But what about Obama and Clinton? If Edwards doesn’t finish third in Iowa, one of them will.
What will it mean for either the Clinton or Obama campaign after either holding the lead for most of the last six months, or taking it away in the last weeks of the race? The short answer is that it will be a disaster. Neither campaign is likely to make the huge blunders in Iowa like Howard Dean’s in 2004, but the perception of Obama and Clinton being the frontrunners is the same, and a third-place finish is going cause a lot of damage with New Hampshire falling just five days later.
The damage could be magnified because while Iowa often has close races finishes between the top to vote getters, the level of competitiveness drops off quickly after second place. In the thirteen contested caucus races since 1972, five have been decided by five points or less. But only two caucuses have seen the top three separated by less than ten points, and the typical spread is over fifteen. Finishing a close third in Iowa would severely hurt the Clinton or Obama campaign in New Hampshire just five days later, but finishing a distant third could be fatal. Like Edwards, if either Clinton or Obama lose both Iowa and then finish second in New Hampshire, it’s hard to imagine recovering to be a serious challenger by February 5.
That’s why I think the race will be a two-person contest next Friday morning. Whoever comes in third in Iowa will still be around, but it would take a clear win in NH after finishing third in Iowa to stay competitive and none of the candidates is running a New Hampshire firewall strategy like Kerry did as his primary strategy in 2004.
When this becomes a two-person contest the dynamic changes totally. If Edwards finishes third it will give Obama a major boost regardless if he wins Iowa or finishes second to Clinton. If it’s Obama who falls in Iowa, Clinton will benefit because Edwards is unlikely to make up the ground or the money difference to be competitive after Iowa. If Clinton comes in third next week it will set a path for Obama to run the table up to February 5.
The compressed schedule is going to increase the importance of losing momentum. In 1992 there was two months between Iowa and Super Tuesday, with only New Hampshire in between. This year there is just 33 days between Iowa and an even bigger Super Tuesday, and there are two more primaries crammed in between. Any one of the three who loses momentum will have a difficult time getting it back before the next primary. String a couple of losses together and I don’t see how recovery is possible.