moldybluecheesecurds 2

Wednesday, January 23, 2008

Nominating a candidate: only 80% democratic

The process of selecting a party's candidate typically draws little attention, with around 10% of eligible voters participating in primary elections. And given the complex rules that govern the selection of presidential candidates, it may come as no surprise.

While around 80% of delegates to each party's convention are selected by the caucuses and primaries currently underway, the remaining delegates are sent to the convention "unpledged." For the Democrats, these unpledged folks are called "superdelegates": party elders, state party chairs, and other party leaders. For Republicans, the same folks are simply "unpledged."

The role of these special delegates is to help consider the party's best interest (or halt the rise of insurgent candidacies), and they can make a big difference. For example, in the Democratic presidential race, the pledged delegate count reads: Obama 38, Clinton 36, Edwards 15. But if you count the leanings of superdelegates (as reported by CNN), the totals shift: Clinton 202, Obama 116, Edwards 51. In other words, the party faithful are leaning toward Clinton.

You can learn more about the strangeness of superdelegates here.

1 comment:

rick said...

Your CNN example represents 89 (38+36+15) pledged delegates and 280 (202+116+51-89) superdelegates. That is, about 76% of the delegates in the example are superdelegates.

The example supports your conclusion that the party faithful (at least those surveyed) are leaning toward Clinton, however it is not the best example of the "big difference" the superdelegates can make, since in the example they are given a weight of about 76%, whereas you state that their true weight is more like 20%.

We don't dispute that the superdelegates can potentially make a big difference. I argue this example is not representative of *how big* of a difference they might make. It is unfortunate that this is point is not emphasized on the CNN page. As long as we're reporting the "leader" we might as well count a different way: If we suppose all delegates in each category (normal and superdelegates) will vote in the same proportion as those reported in the example, then Clinton is ahead with 1788 to Obama's 1611 and Edwards's 548. The superdelegates still make enough of a difference to put Clinton in the lead, but not the overwhelming lead represented on the CNN example. (Of course, both methods have advantages and disadvantages.)