The Deal With Add-Ons

One advantage of the 2008 Primary: Director’sCut: Extended Edition is the opportunity to learn every last minutiaof the Democratic nomination process. What used to be a complex, impenetrablesystem no one had time to examine has now become a complex, impenetrable systemwe have way, way, too much time to examine.

Under the microscope today is a group of superdelegatescalled add-on delegates . The 76 add-ons are unbound,just like other superdelegates. The difference is that they’re not named until thespring, when the states hold their conventions. Most states have just oneadd-on delegate, but some bigger states have more. (Pennsylvania,for example, has three; Californiahas five.) Each state has a different process for selecting add-ons—sometimesthe state party chair picks them, sometimes it’s a committee, sometimes anentire convention. If you’re curious, DemConWatch listswhen and how each state picks.

So although you can’t predict their behavior perfectly,there’s at least some logic behindwhich way the add-ons swing. For example, Californiastate party chairman Art Torres says he plans to pick the state’s five add-ons proportionally according to theprimary results—three for Clinton,two for Obama. In Washington, D.C., where Obama wonoverwhelmingly, one add-on has declaredfor Obama and one is still undecided. In states with only one add-on, thedelegate is likely to go to the candidate who won the state. That’s why Clinton won Arkansas’ singleadd-on delegate. In general, the add-ons seem to break down roughly accordingto which candidate won where.

So here’s an experiment: What happens if each of the statesthat haven’t yet selected their add-ons pick proportionally according to theprimary results? (For the sake of argument, we’ll go with ABC’s prediction yesterday that Clinton wins Pennsylvania,Indiana, West Virginia,Kentucky, and Puerto Rico, and that Obama winsGuam, North Carolina,Oregon, Montana,and South Dakota.)In states that have more than one delegate, we split them roughly proportionally.For example, each candidate gets one of South Carolina’stwo add-ons; likewise, each gets two of New York’s four. In Pennsylvania,we’ll give Clintontwo and Obama one.

In this scenario, of the 63 add-on delegates that have yetto be selected, Obama gets 35 and Clintongets 28. Factor in the add-ons who have already declared their allegiances, andObama gets 41, Clinton29. (Another five have been selected, but not yet announced whom they’rebacking.) In other words, it’s likely the add-ons will split in Obama’s favor,or at worst roughly 50-50.

The takeaway point being that the superdelegatewall we discussed yesterday is evenhigher if you take the add-on situation into account. Seeing as the add-onsare likely to favor Obama, Clintonneeds even more than 70 percent of the remaining uncommitted superdelegates(our ballpark prediction) in order to reach 2025. One Frayster, “Independent Don,” crunched thenumbers (an excellent post I encourage you to read) and concluded that ifthe two candidates hypothetically split the remaining pledged delegates 50-50, then,given the likely allocation of add-ons, Clinton would need to win a whopping 90 percent of the remaining 230 or so superdelegatesto get the nomination.

It’s not often we use the word”impossible” around here. But it’s starting to look like there is no other way to describe Clinton’s chances.