I was asked recently to provide regularly comment for The Hill's The Big Question. Last week I posted thoughts on what success might look like in Afghanistan, focusing on the value off building stronger democratic institutions, and President Barack Obama's delicate dance of political positioning.

Yesterday's question was whether Obama should campaign hard for his party's gubernatorial candidates in Virginia and New Jersey. Here were my comment:

President Obama hasn't shied away from uphill risks if there's a chance of success: witness his trip to Copenhagen to seek the Olympics for Chicago, his ambitious domestic policy program, and his own underdog campaigns for Congress (loss), U.S. Senate (win) and the presidency (win). With Creigh Deeds and Jon Corzine both having a chance to win, I doubt he'll give up on them, particularly given that analysts will tie their results to his presidency regardless of what he does.

Furthermore, Obama's political team will want him in Virginia for two reason. First, Virginia is one of the dozen swing states that will be key to his prospective re-election bid in 2012. During the final two months of the 2008 campaign, the major party candidates spend more than 90% of their time and resources on those swing states, and this year the White House has sent the president overwhelmingly to them again. Second, the state legislators elected this year are going to draw the congressional district lines that govern U.S. House elections in 2012-2020. It's no exaggeration to say that how Virginians are represented in Congress in 2014 may have more to do with how votes are cast this November than in the actual 2014 elections. Even if Deeds can't be saved, Democrats will want to hold onto the state senate and compete for the the House of Delegates.

For more on the rather startlingly political nature of the president's travels this year, see our Innovative Analysis The Swing States of America: Has the Electoral College Pushed Barack Obama Away from a 50-State Strategy.

For more on how gubernatorial elections and presidential election often generate different results in states, see our Innovative Analysis How Presidential Elections Make Poor Predictors of Gubernatorial Races.

For more on Virginia and its remarkable history of non-competitiveness in state legislative elections in the 1990s, see this FairVote study.