For most of the year, the assumption among political observers has been that Republicans are likely to take control of the Senate in this November’s elections. Because Democrats are defending more seats, many of them in extremely conservative states, holding their position would be a surprising victory. Add in President Obama’s poor approval ratings, the tendency of the president’s party to lose seats in midterm elections, and the natural turnout advantage Republicans have in non-presidential years, and it seemed like almost a sure thing that Republicans would win enough seats to take control.
But in the last week or so, nearly all the well-respected predictive models are showing the Democrats with a better chance of keeping their majority than people thought. Republicans still have an advantage in most of the models, but in many cases it’s a smaller one than it was. What’s going on?
Let’s run down them quickly:
- New York Times: Republicans have a 67 percent chance of taking control
- FiveThirtyEight: Republicans have a 65 percent chance of taking control
- Washington Post Election Lab: Republicans have a 53 percent chance of taking control
- Huffpost Pollster: Democrats have a 56 percent chance of retaining control
- Princeton Election Consortium: Democrats have a 91 percent chance of retaining control
While two out of the five models show the GOP with an edge, it is not as large as it used to be in either — and three out of the five show it either very close to a toss-up or (as Princeton’s does) leaning Democratic.
What happened? The real difference is in the changes in weight each methodology gives to polls as compared to other factors that might influence the outcome. Yesterday, Chris Cillizza explained why the Post’s model has moved in the Democrats’ direction:
At the start of an election cycle, the model is based heavily on underlying fundamentals of past elections. It’s almost an entirely generic calculation that takes little account of candidates, polling etc. There’s a reason for that, of course. Early in election cycles there often aren’t candidates in races yet and, therefore, polling is limited. As the cycle gets into its latter stages — like now — the model tilts to rely more heavily on candidates and polling and less on fundamentals. (That doesn’t mean the underlying fundamentals don’t matter at all in the model. They just matter less and less as the election gets closer.)
The HuffPost and Princeton models use only poll results (with some adjustments for things like house effects), while the Times and 538 models incorporate more other factors. We won’t know until election day which is the most accurate, but it does look like the most changeable factors — the ups and downs of each race — are giving Democrats reasons for at least some optimism.
Republicans have been pleased this year that they’ve managed to avoid nominating the kind of buffoons who self-immolate before election day, as candidates like Todd Akin or Sharron Angle did in the past. On the other side, however, Democratic candidates have shown themselves to be tough contenders. Despite the challenge of holding seats in conservative states, Democrats like Mark Begich, Mark Pryor, and Kay Hagan have both avoided major mistakes and found effective ways to go after their opponents.
That isn’t to say they’re all going to win. Few smart analysts would say that Democrats now have the advantage. But they sure look to be in better shape than they did not too long ago.
Of course, that could simply reflect the fact that the models are increasingly showing what has been the reality all along — the battle for the Senate is, and has been, pretty close to a toss-up. Indeed, at the moment there are no fewer than nine races that could go either way.
There could also be unexpected political developments. Just yesterday, the Democratic nominee in Kansas dropped out of the race, leaving incumbent Sen. Pat Roberts to face an unusually strong independent candidate. Until then this was a safe Republican seat; now nobody’s sure what might happen. And the battle for the Senate is so tight that one seat shifting might be all it takes to determine control of the chamber.
Other changes are possible. President Obama could announce executive changes to the immigration system. We could have more positive job numbers (there will be another report tomorrow), leading to a raft of news stories about the improving economy. The administration could step up its military action against ISIS — which could influence elections either way depending on what happens. In short, we’ve reached a point where any small nudge could push the whole election in one party’s favor.