Final West Virginia primary polls:

Elliot predicts Clinton 65%, Obama 31%, Others 3% with a delegate split of 19 to 9 in Clinton’s favor. This sounds perfectly reasonable to me, placing the result exactly between Arkansas (70 to 26) and Oklahoma (55 to 31). So, I suppose I’ll make that my official prediction: Clinton will win West Virginia by 24% to 44%, with a median of 34% and a net gain of 10 delegates. On the other hand, my own own rule of thumb suggests “Arkansas minus 5%” as a better rule, so if the margin is closer to 39%, I’ll know I should stick with my model rather than adopt Elliot’s. The likelihood of anyone getting within 5% is pretty slim due to the difficulty in predicting the exact margin on a blowout, and I usually just stick to saying “25% or more”. However, the primaries are almost over and I decided to just go for it, albeit with a very large possible range.

We’ll keep an eye out for any exit polls (which I’m duty-bound to remind you are ONLY USED FOR DEMOGRAPHIC PURPOSES – they are notoriously bad at predicting actual results). We’ll be liveblogging the results for this race and the House race in Mississippi tonight as they come in.