Predictions

Ohio and Texas tomorrow, and it's been a while since I've had the opportunity to make predictions that turn out to be wrong, so I'm excited.

Let's start with Texas.

In the past 10 days there have been 21 publicly available polls taken of the Democratic race in Texas, and the bottom line is that almost every one has suggested a tight race. 17 of the 21 polls have the two candidates within four points of each other. 13 have Obama ahead, 5 have Clinton ahead, and 3 have the race tied. According to the weighted average, these 21 polls have the race as follows:

Obama: 47.2%
Clinton: 46%
Undecided: 6.8%

There doesn't appear to have been a whole lot of movement over the past ten days. Polls taken before the 27th of February have the race essentially in the same place as polls taken after. This race is going to come down to turnout, organization, and undecideds. I'd say that, on balance, those three favor Obama, so my prediction is as follows:

Obama: 54%
Clinton: 46%

Now to Ohio. There have been 20 polls taken since February 21, and 18 in the past 10 days. Like the Texas polls, they have been pretty consistent. Of the 20 polls, all but one show a Clinton lead, and 15 show a lead of 5 points or more. Despite news reports of "momentum shift" and whatnot, the polls from 12 days ago show about the same thing as polls from the last few days. Here's what the weighted poll shows:

Clinton: 48.9%
Obama: 42.7%

As with Texas, there don't appear to be any trends within these past 12 days. The first five polls have Obama at 41.3 and Clinton at 48.9 (a spread of 7.6 points), the last five have Obama at 43.6 and Clinton at 50.4 (a spread of 6.8 points). So it appears to me that Clinton has a small, but real lead in Ohio and that it would take a big break of undecideds and some serious organizational muscle for Obama to pull it out. My prediction:

Clinton: 52%
Obama: 48%

Until tomorrow night...