Shifting Seats

For much of this year, it seemed a near mathematical impossibility that Republicans could score the 10-seat net gain needed to flip the Senate, which is split between 59 Democrats (including two independents who caucus with Democrats and largely vote with the party) and 41 Republicans. As recently as six weeks ago, I wrote in a CongressDailyAM column that a GOP win was "certainly possible" but "still fairly unlikely." Although the "fairly unlikely" part is still valid, the possibility of a GOP takeover is growing.

To be sure, a 10-seat gain for Republicans remains hard. Eighteen Senate seats could plausibly turn over -- a dozen held by Democrats and six by Republicans. Looking first at the five open seats -- Florida, Kentucky, Missouri, New Hampshire, and Ohio -- that the GOP is defending, the Republican challenger holds the lead in each race. Granite State voters won't select nominees until Sept. 14, but former Attorney General Kelly Ayotte, the Republican with the best chance of defeating Rep. Paul Hodes, the presumptive Democratic nominee, is increasingly favored to win the GOP nod. None of the Republican leads in these five states is insurmountable, but at this point, you would rather be the GOP nominee than the Democratic one in each place.

There are few signs that the contest of scandal-plagued Rep. David Vitter, R-La., is getting competitive. Meanwhile, Rep. Richard Burr's numbers suggest real vulnerability in North Carolina, but it looks exceedingly unlikely that either Southern incumbent is in any real danger of losing. In a better year for Democrats, maybe, but not this year.

Suffice it to say that Republicans have a good shot at holding all their seats. If that's true, then the GOP would need to win 10 Democratic-held seats to win the majority.

Turning to the Democratic-held seats, the open seats in Delaware, Indiana, and North Dakota are pretty much goners, and it appears increasingly remote that Sen. Blanche Lincoln can make a successful comeback in Arkansas. This would bring Democratic losses to four.

If I had to make a wager today, I would bet that the open seats in Illinois and Pennsylvania will also fall to Republicans, although both races remain quite competitive and are hardly over. If my hunch is correct, Republicans would gain six seats. That brings us to Democratic incumbents Michael Bennet (Colorado), Barbara Boxer (California), Russell Feingold (Wisconsin), Patty Murray (Washington), and Harry Reid (Nevada), who are all roughly even-money bets. Boxer, Murray, and Reid have statistically insignificant leads over their challengers, while Bennet and Feingold trail their opponents by similarly insignificant margins.

In Connecticut, where the seat is open, Democrats are watching their once huge lead erode rapidly. Some Republicans are also eyeing the West Virginia open seat, noting that President Obama's job-approval ratings in the Mountaineer State are among his lowest in the country, and they speculate it could get interesting as Democratic Gov. Joe Manchin pivots from state issues to more polarizing and ideological national ones.

With this many races in play, Democrats may have to perform triage and focus their resources on those that remain winnable. That means giving up on the rest.

Of the 18 competitive Senate races (this number doesn't include Vitter, Burr, or the seat in West Virginia), Republicans would need to win 16 to secure a majority, and certainly logic suggests that the odds of achieving this would be long in any remotely normal year. But the operative term is "in a normal year," which this is most certainly not.

The Senate editor of The Cook Political Report, Jennifer Duffy, notes that the toss-up races don't always break evenly. She points to the Democratic wave year of 2006, when the party won 89 percent of the nine races that The Cook Political Report rated as toss-ups before the election. In 2008, Democrats won 78 percent of the toss-up races, while in 2004, a good year for Republicans, the GOP won 89 percent of the most competitive races. In other words, these wave elections produce a cascading effect in which the close races often break disproportionately toward the wave. The exception was 1982, when the anti-Republican wave that hit the House missed the Senate as the closest races broke in the GOP's favor.

The odds still favor Democrats holding their majority, but that is no longer given.