Post South Carolina, a quick run through of the delegate numbers in play in the weeks ahead and some projections of what Obama would need to do on Super Tuesday and beyond to get ahead of Clinton.
It's going to be quite a ride. It will be fascinating to see how it comes out. The most likely scenario remains a brokered convention. About which nobody knows anything much.
Warning. Some math may be encountered along the way.
First, a few preliminaries. I'm going assume for the sake of brevity and that everyone is familiar with the rules of the primary process. For in depth description of how the primary process works see Wikipedias excellent primer on the Democratic primary process. The main distinction that is important in delegate math is that between pledged delegates and super delegates. A pledged delegate is "pledged" to a particular candidate as a result of voting in a primary or caucus. A super delegate is some form of party dignitary, congressperson, senator, governor etc. who has a vote at the convention and is at liberty to vote any way they choose. There are 3,253 pledged delegates at the convention and 796 super delegates for a total of 4,049 (including those from Florida and Michigan who won't get seated because they been bad).
The polling numbers I'm going to use are Real Clear Politics' averages of recent polls. If you don't trust RCP's methods or think they use the wrong polls for their averages, or that they are in thrall to a vast MSM conspiracy, you should save yourself some aggravation and stop reading now.
As to estimates of super delegates. I'm using numbers compiled by CNN which they claim to be based on clear public declarations by congresspoeple, senators, governors etc. that they will vote for a particular candidate. As we all know, the solemn promises of such office holders are subject to change without notice. Also, the vast majority of super delegates are yet to show their hand. So, super delegate numbers should always to be treated as a guess, no matter how well sourced. If you think CNN's work in this area is not to be trusted or that they are the conveners of a vast MSM conspiracy, you should stop reading now.
So, on to the starting point, the hard(ish) numbers. As we stand, still peering through the dust settling around the South Carolina landscape at the departing silhouette of John Edwards, the delegate count is as follows
Pledged Super(est.) Total(est.)
Clinton 48 184 232
Obama 63 89 151
Edwards 26 36 62
At this stage the total estimates are disproportionately distorted by the fact that, as of now, we have more super delegates committed (so they say) than pledged delegates. This is due to the slow accumulation of declarations and endorsements from super delegates over the past year. This means that while Obama is ahead on voted-for delegates he is still well behind Clinton. She apparently has a strong lead amongst early committing super delegates.
So, what happens next? We now go straight to super Tuesday where there are 1688 pledged delegates up for grabs. As I write we are a little bit in limbo as regards polling for Super Tuesday. We haven't had enough polling results come in since South Carolina, and the Edwards retirement, to know what effect either of these developments have had on the voters.
The polling that we do have for the three big Super Tuesday states and nationally is as follows
Clinton Obama Edwards
National 42.7% 33.2% 12.2%
California 45.0% 33.3% 10.3%
New York 50.2% 27.3% 9.8%
New Jersey 46.0% 28.0% 10.5%
So, how can we use these numbers to estimate delegate counts? Obviously we allocate Clinton and Obama their share based on their projected vote. We can then allocate the unknowns (100% minus the total known, about 12-15%)in the same ratio as the candidates projected votes. As an approximation we can use the national poll numbers for the other 19 Super Tuesday states. That leaves the Edwards vote, which is up for grabs. We should start by allocating those 50/50. So, on that basis, if the vote on Super Tuesday is as the current polls predict the delegate count on the morning of Post-Super Wednesday would be
Pledged Super Total
Clinton 965 184 1149
Obama 723 89 812
Edwards 26 36 62
Which looks like Clinton would be well on her way, with some work left to do. These figures contain two key assumptions, that Obama gets little bounce from South Carolina and that the Edwards voters spilt 50/50.
Having established this scenario as our "no change" base point we can explore a couple of possible scenarios. First the question of post South Carolina bounce. If it was me I would want to at least come out even on pledged delegates after Super Tuesday and start to close the gap with super delegates included. To do this it turns out Obama needs a 7.5% bounce. That would deliver results as follows
Pledged Super Total
Clinton 818 184 1002
Obama 870 89 959
Edwards 26 36 62
If we continue that scenario, through the remaining primaries, which decide 1,378 pledged delegates we would get the following result
Pledged Super Total
Clinton 1342 184 1516
Obama 1845 89 1934
Edwards 26 36 62
That would leave Obama 318 delegates ahead at the convention with 487 super delegates and 62 Edwards delegates still open to persuasion. So, even an good bounce of 7.5% for Obama post South Carolina would likely leave us with a brokered convention. To put the matter beyond doubt Obama would need something like a 10% or better bounce that is sustained through April.
In comparison to the results from even a modest bounce for Obama, any realistic breakup of Edwards voters makes a smaller difference. Even in the completely unreal scenario where 100% of Edwards voters go to Obama the count, based on current poll results, still comes out with Clinton ahead on pledged delegates by 869 to 819 after Super Tuesday, and therefore still well ahead with super delegates included.
Realistically it's hard to see the super delegates breaking Obama's way if the pledged delegates do not. That is, whilst Clinton can hope to make up a modest pledged delegate deficit with super delegates, Obama probably can't.
So what's all this telling us? Well, given Clinton's lead in super delegates to date, Obama needs at least a break even result on Super Tuesday to stay competitive. That would require a 7.5% bounce after South Carolina, which is a big ask in national numbers in 1 week. Whilst a good break with Edwards' voters and Edwards' existing delegates would help it won't make up the difference if there is no bounce. In fact if Edwards were to endorse Obama the main benefit would be in giving Obama a bounce in voting on Super Tuesday by moving undecided and Clinton voters.
Long story short, Obama needs to have vote getting on Tuesday as his first priority, but, given that it is likely be close, he needs to back it up by winning over Edwards and his supporters and as many super delegates as possible. Taken all together, such a three pronged effort will take lots of resources in a short time. Which is why it takes $100M to get nominated, and why campaign staff don't sleep much.
It's to going to be fascinating to see what happens. If the polls move substantially between now and Super Tuesday I'll do an update of this post with a new set of calculations.