Predicting the Superdelegates

One of our readers (thanks Andy) pointed us to a pretty darn good website for those following the twists and turns of the roughly 300 uncommitted Democratic superdelegates who will end up deciding for the rest of us who the party's nominee will be this fall.

It's the blog of the Center for Congressional and Presidential Studies at American University, and they have a lot of neat stuff there, not the least of which is a statistical model that predicts how each of the remaining uncommitted superdelegates will vote. (For the current iteration of the model, see here.)

Based on the model, Obama is highly likely to get the votes of most of the remaining superdelegates, including an 80% probability that Virginia superdelegates Jim Webb, Joe Johnson and Jerome Wiley-Segovia will go with the Illinois senator.

Indeed, the CCPS model predicts that 171 delegates are at least 60% likely to go for Obama, compared to just 19 for Clinton, with 78 superdelegates in the "too close to call" mid-range.

It will be interesting to see how this model works in the coming weeks. We hope they'll update the predictions soon.

The site contains a wealth of other interesting statistical information, such as comparisons of Pennsylvania to various other states on measures such as gun ownership, Walmart shopping, pick-up truck use and PBS watching, in an effort to see which states Pennsylvania most mirrors for electoral purposes. There are also analyses of anomalies in polls and other tidbits of wisdom, so if you're a numbers junkie like the Curmudgeon, check it out!