Well, it's less than one week until the midterm elections are over. As I pointed out here, I have absolutely no idea how things are going to play out, but at least we have a better idea than we did a few months ago. In the House, where polling is scarce and largely inaccurate and politicians and pundits more or less only have bits of data and loads of gut instinct, it appears the GOP will take control. Personally, I still peg it at 50/50--they need 39 seats to flip to gain control, and most polls state that the Republicans will pick up between 30 and 50. (Gee, thanks.) The situation is slightly different than in 1994, when every single Republican incumbent safely defended their seat. This year, there are a handful (about five) seats that were in marginal districts to begin with or had a special election where there is no way they can survive a general election challenge. My feeling is that the GOP will gain control, but not by a significant amount.
In the Senate, it appears that the GOP will pick up between 6 and 8 seats, surprisingly close to what it was late in the summer. There was a brief moment in which it looked like they may pick up 51 seats, but that was in an era when the likes of Washington state were going to go red. Right now the sure things* are Arkansas, Indiana, North Dakota, and (surprisingly) Wisconsin. A probable pickup is Colorado. The polling in Pennsylvania, Illinois, and Nevada all slightly say Republican, but I find it hard to believe all three will. PA and IL are traditionally Democratic states, so it's hard to guess, and Harry Reid's position--for better or worse--will make him hard to dislodge. I'll say two out of three, which will bring (barring a huge disaster for the Republicans) the total to seven. In a no-holds-barred tsunami of GOP support, it would tip those three and probably Washington and West Virginia. That's ten. If they lose any one of these races, the Republicans would have to pick up California--something I highly doubt--to reach the vaunted magic number ten to gain control of the Senate.
Of course, that leads to another question: do the Republicans actually want control of both Houses? Officially, they have to say yes. But if the true prize is the Presidency, they would be better served to gain enough seats to effectively block any major legislation, but not take control so they wouldn't have to take any responsibility for the performance of Congress. In an odd way, the election of the Tea Party candidates helps them--now there are few moderate Republicans but plenty of scared moderate Democrats, making peeling off Senators and Representatives easier. (Of course, the presence of the Tea Party makes things much more uncomfortable in other ways, so I doubt it's a net benefit.) Once in power, though, having new moderate Republicans will become a benefit--and the next election cycle will have many more such opportunities.
The best example to look at is 1994. The Republicans took over, proposed a wide, huge swath of academic but unpopular programs that conservatives had left boiling in the laboratory for four decades, and virtually handed Bill Clinton a re-election bid. The exact same thing might happen this time around if the GOP doesn't keep tabs on its members; the unruly Tea Party will make this doubly hard this time around. Hopefully the Republican leadership will have better image consultants than Newt Gingrich did.
So the likely scenario is that the Republicans gain control of the House and the Democrats barely control the Senate. Obama won't be able to propose much complex or far-reaching legislation for two years. (Given the huge spending the Democrats engaged in for decades, and the huge amount of spending the Republicans then did for a decade after that, having both sides pull on both ends for a bit may be a boon to both parties--and, dare I say, us.) Obama, as an incumbent and with little recent controversy, will have an advantage for the presidency. (At this point, of course, a thousand unseen factors will have emerged. I'm generalizing here. And people who think that the health care bill and everything that's happened for the past two years will still be relevant in 2012 vastly overestimates the memory of the average voter.) Working for the Republicans is the fact that the next election cycle has an unusually large amount of Democratic senators in deep red states that got elected in the cranky year of 2006, and with a presidential election they will all be in tough battles.
But we're getting ahead of ourselves here. Polls say it will be about 230-205 Republican, and I'm happy with that number. (I'd trend a little higher since voters tend to break for the challenger, and I think that effect will be more pronounced this year. I'm willing to go 240, but I'm doing this not knowing how much the pollsters already took that into account.) The Senate is tricky, but I'll go with 52-48 Democrats. Arkansas, North Dakota, Colorado, Illinois, Pennsylvania, West Virginia, and Wisconsin will all flip. Nevada and Washington will be close but will remain Democratic. California isn't as close as it seems and will stay blue. And Kentucky becomes a much, much closer fight than anyone thinks but is still retained by the GOP.
The most important thing to note that in nearly all the polling, very few of the "winners" are polling above 50%. Which means that undecided voters still make up a chunk of the electorate. They will tend to go for the challenger (in this case, the Republicans) but they are still a question mark. We will find out next Tuesday.
*I am fully aware that a "sure thing" in politics is akin to "about 52% chance instead of 50% chance."