Bad Analysis of NH Numbers

Let us crunch some New Hampshire numbers (with a heavy reliance on The Green Papers). Via here we see the latest NH poll results:

Braun 0%
Clark 16%
Dean 36%
Edwards 3%
Gephardt 6%
Kerry 13%
Kucinich 2%
Lieberman 7%
Sharpton 0%
Other 0%
Undecided 17%
Sample size 613

Democrats 437
Undeclared 176

For the sake of argument, let us assume that the above numbers are at least roughly accurate, and that the undecided voters will more or less vote according to the orginial pattern. That gives us new percentages of (very roughly)

Braun 0%
Clark 20%
Dean 45%
Edwards 3%
Gephardt 7%
Kerry 15%
Kucinich 2%
Lieberman 8%
Sharpton 0%
Other 0%

Eliminate all the under 15% (the Democrat's threshold for delegates) and you get

Dean 45% - 56.25% of pledged delegates, or 12
Clark 20% - 25.00% of pledged delegates, or 6
Kerry 15% - 18.75% of pledged delegates, or 4
Other 20% - N/A
Total pledged delegates: 22

Compare this to 2000: Gore got 13 delegates and Bradley 9 pledged delegates, but Gore did this through what was just short of a majority (76,897 votes to 70,502 to 7,240). In other words, this will likely be a win for Dean, but not an unambigous one. Also note that in order for Dean or any other candidate to win they will need at least 2,162 delegates out of the 4,322 available. 3,520 of these delegates will be chosen by caucus or primary; if our hypothetical Dean continues to pick up a hypothetical 56.25% of the available delegates he will end up with only 1,980... which means that he'll have to cut a deal with somebody to get the nomination.

Of course, I made about umpteen different assumptions on this thing, so I'm essentially guessing, but it looks generally right and it's not too unreasonable to think that Kerry can barely pull out a 15%. At any rate, the end result of all of this is: don't expect majorities, and do expect horse-trading. Maybe lots of horse-trading.

Comments

[Mark]

Doesn't that depend on Kerry staying in?

[Moe Lane]

"Doesn't that depend on Kerry staying in?"

That's the funny part about this: any candidate who'll be able to consistently pull 15% may find it in his (no offense to Braun, but she won't) best interest to stick it out. Ask again after March, when all shall be clearer.

[seth]

Moe is right - if I'm Kerry or Edwards, even if I fully intend on pulling out down the road, it still behooves me to have a handful of pledged delegates I can use to be a player at the convention.

[praktike]

Moe, what kind of horse trading do you think would happen? Do you have any insight as to how those behind-the-scenes discussions happen?

[Catsy]

Keep in mind that the longer the individual candidates stay in, the more establishment pressure there is going to be on the lower-polling candidates to drop out and throw their weight behind the front-runner, and the more momentum the front-runner is going to gain. People like to back a winner. The more endorsements and donations Dean picks up and the further he pulls ahead, the more that people on the fence are going to fall into his camp. The analysis here assumes that the undecideds will distribute themselves based on current patterns in one fell swoop, when in actuality it will happen over a period of time, with more people going towards the front-runner the further he seems to pull ahead.

All of this, of course, assuming all else remains equal--that none of the candidates say or do anything which works substantially for or against them--and ignores current trends. Kerry, for instance, has been trending quickly and steadily downward in the polls for a few weeks now. Clark has been mostly steady, with a recent upward tick commeasurate with Kerry's drop. Dean's support has remained mostly unchanged, with all movement within the MOE and no directional trends.