Now that we have post South Carolina and post Edwards polling from several places we can get serious (OK amateurish) about playing the post Super Tuesday delegate count game.
The polls are coming in quickly now and the doubts and questions about their competence and honesty are coming in even faster around the blogiverse ( and not just from Obama supporters trying to keep calm). The bookies are ready to take your bet. Can you beat the pollsters?
This is a follow up to my previous post which uses the same methods. If you want to know more about my figuring have a look there. This time I've added in specific numbers for Georgia and Illinois.
I've put the latest polls though my own humble little model and come up with the following
If we use Real Clear Politics' rolling averages, which at the moment include a mix of earlier polls pre-Edwards' departure and later ones that look better for Obama, we get the following
Clinton Obama
National 47.3% 40.5%
California 41.5% 40.0%
NY 53.8% 33.5%
NJ 46.8% 38.8%
GA 35.5% 49.5%
IL 23.3% 55.3%
That gives us a delegate count from the Super Tuesday states of
Clinton 870
Obama 818
which if we add in the delegates decided before Super Tuesday comes to a total of
Clinton 918
Obama 881
However, if we accept the trend shown in today's polls, which includes some numbers from Zogby that smack more of showmanship than statistical credibility, they show Obama closing the gap. Using those numbers we get
Clinton Obama
National 46.0% 44.0%
California 44.0% 45.0%
NY 53.8% 33.5%
NJ 43.0% 42.0%
GA 35.5% 49.5%
IL 24.0% 55.0%
Which gives a Super Tuesday delegate count of
Clinton 841
Obama 847
plus delegates decided before Super Tuesday comes to
Clinton 889
Obama 910
All of which looks good for Obama closing the gap and a brokered convention. But, the pollsters have a lousy seacon so far. Do we trust them? What's your prediction?