Wednesday, March 05, 2008

It's not Karl Rove's math...

So, last night was a good night for the Clinton campaign. They learned that going negative gets them votes, and votes get them the media, and the media gets them the story. Congrats to them, and thanks for all the wonderment and joy.


Did yesterday change anything? If so, how? And what has to happen in the remaining 12 races (WY, MS, PA, IN, WV, KY, ND, MT, NC, OR, Guam, and Puerto Rico) for her to win this thing?

Well, over at Daily Kos, a diarist named PocketNines has done a little bit of math (no, don't run away!) and we learn some interesting things from this diary:

Because of the way that the Democratic party splits delegates proportionally, merely winning a state or a district doesn't necessarily grant the winner a large margin of delegates.

In order to cross all thresholds except the initial break that give you a +2 delegate swing, you need to win by an extra 200/X%, where X = the number of total delegates at stake. Let's see how this works by easy example - West Virginia and its 10 statewide delegates. 200/10 = 20%. To go from 5-5 to 6-4 there you have to win by over 10% (55-45). But to get ANOTHER +2 you need to add 20% to your win and win by 30% (65-35).

What do we see here? In a district/state with 10 delegates, the "winner" gets 5 if they get less than 55% of the vote, 6 if they get between 55 and 65, 7 if they get between 65-75, 8 if they get between 75-85, 9 if they get between 85-95, and 10 if they win more than 95%. So, a 10-point win by either candidate only nets a +2 delegate margin, and to increase that they have to win by 30%. Trivia: In how many states has Hillary gotten more than 65% of the vote? Answer: One - Arkansas (70%)

What's the upshot of this? If Hillary was able to win every race by 10% and every district by 10%, she'd get a net gain of 58 delegates (Obama's lead is somewhere around 140-150 right now in pledged delegates). If we raise that number to 16.5% (58.25-41.75), then she gains 64 delegates on Obama. If she won every district by 24.9% (62.45-37.55), then she gains 110 delegates on him (and still trails by 30-40).

Putting it this way, there is no realistic chance that she'll catch him in pledged delegates, and likely not even get close. If she's still down by 80+ delegates, her means of victory are in SuperD's and/or Florida and Michigan. Of course, if she wins every state from here on out by 10%+, then her case to get the SuperD's is reasonable (not airtight, but it's certainly an argument with value). But if she splits the states reasonably evenly (which is still a good scenario for her), she won't gain any appreciable number of delegates on him, which means that she'll need a huge number of the remaining SuperDelegates to go her way.

No comments: