11:14 PM ET: After a seesaw night, in which the Signal was drafting explanations for why we don't expect to get every prediction right, we're calling Ohio for Romney. The markets are now 97 percent in Romney's favor, in large part because the last 15 percent of the vote to be reported is primarily from areas that favor Romney.
10:52 PM ET: After briefly flipping to Santorum's favor, the prediction markets have reversed back to Romney in Ohio. This is likely because the regions with the most votes yet to be reported are expected to favor the former Massachusetts governor.
10:18 PM ET: The markets have flipped to Santorum's favor. (Hey, if you predict things enough times, eventually you're correct!) Santorum is hanging on to a 2-point lead with returns of just over 60 percent.
9:23 PM ET: With 27% reporting Santorum is leading Romney by 7,500 votes, but our model still predicts a Romney victory. The reason is Cuyahoga County (Cleveland), Franklin County (Columbus), and Hamilton County (Cincinnati). They have reported 0, 14, and 0 percent and are likely to go towards Romney.
9:16 PM ET: In the last hour there have been 757 tweets about Santorum, 1255 about Gingrich, 3021 about Romney, and ... 14,065 about Peyton Manning.
9:10 PM ET: Santorum takes the lead in the results with 20 percent reporting in Ohio, but markets still favor Romney to win this one, albeit less confidently than earlier in the night.
7:59 PM ET: The night has been going as expected so far. And, in our estimation, the most interesting state to most observers has gotten a bit less interesting. Ohio is now 92.6 percent likely to go for Romney.
The chart shows the progression of the likelihood of victory in Ohio since the morning after the last round of primaries. This is a great example of the value of prediction markets. The markets were able to adjust quickly, in real-time, to new information. They are also able to incorporate more information, like Romney's upward momentum in Ohio and his superior organization.
Sources: (market-based) Betfair and Intrade
5:40 p.m. ET: As expected, Mitt Romney has won Virginia and Newt Gingrich has won Georgia. Here's what to watch for tonight:
Ohio is the election that everyone is discussing, because the polls are tight, the delegate count is meaningful, and the outcome could be some sort of indicator of the general election in that perpetually pivotal swing state. We predict Romney to carry the state with a nearly 90 percent likelihood. There are 66 delegates at stake, second only to Georgia. Ohio is a crucial swing state this year, as always; our model indicates Ohio as a key pickup if the Republicans are going to capture the presidency.
Alaska, North Dakota, and Tennessee are the closest elections for us going into the evening. Romney is favored at 70.8 percent likelihood in Alaska; Ron Paul trails him at 21.5 percent likelihood. Similarly, Romney is 64.0 likely to carry with North Dakota; Paul trails him there at 22.7 percent. Rick Santorum is 71.3 percent likely to carry Tennessee; Romney trails him there with 24.1 percent likelihood of victory.
David Rothschild is an economist at Yahoo! Research. He has a Ph.D. in applied economics from the Wharton School of Business at the University of Pennsylvania. Follow him on Twitter @DavMicRot and email him at email@example.com.
Want more? Visit The Signal blog, connect with us on Facebook, and follow us on Twitter. Handy with a camera? Join the Yahoo! News Election 2012 Flickr group to submit your photos of the campaign in action.