We know 2010 is going to be a good Republican year. The polls all suggest they’re going to pick up seats in both the House and the Senate. But how many seats can we expect?

The Margin of Error Blog does a fairly sophisicated projection of the 2010 midterms based on past seat pick-ups relative to the generic party ballot (i.e. whether you would support a Republican or a Democrat, generally speaking, in the upcoming election). Here is a brief explanation of how the regression-based estimate works:

What you basically have is that the generic ballot this far from an election almost always overestimates the percentage of the vote the party who holds the White House… In your case, the 49.7% the Dems would get of the House popular vote would actually be more like 46 (and an 8 point edge for the Pubs) on election day considering historical trends…

Here is a figure illustrating the point he is making (namely that the generic ballot overestimates the vote in a midterm election for the party in control of the White House). The shaded areas are years the Republicans control the White House. Note that the dotted line (representing the % of Democratic seats) is below the average district vote during those years…while it is well above it during Democratic years. In short: Democrats do better than the generic when there is a Republican president and they do worse when there is a Democratic president.

As for the analysis itself, again, briefly: regression analysis merely means fitting a straight line to a set of data points (the line that ‘best’ fits those points…or the line where the distance between it and the points is the smallest). So here we see the line fitted for all elections since 1946 and, since lines extend out to infinity, that provides a point prediction for the 2010 elections. IOW, we have a number for what we expect the vote share of the Demcrats to be in 2010 (i.e. the point on the line). The two larger dots in the graph that MOE blog singles out are the best & worst ‘points’ on the fitted line that predict what the Democrats will win in 2010. Note how low those points are on the graph relative to the Y axis, which represents the % of seats for the Democrats. Being in the lower left-hand quadrant of the graph means your party is performing badly on the generic and your predicted vote total for a given election is low. And the Democratic points in 2010 are two of the worst points for Dems in the graph.

He estimates Republicans could pick up as many as 60 seats in the House. This total not only gives them the House (and the speakership), but it would blow 1994, the Republican Revolution, out of the water (54 seat pickup). I have one reaction to this: wow.

With current polling in conjunction with Bafumi et al.’s paper predicting a Republican national vote between 53.6% and 54.7%, the Republicans could easily gain 50-60 seats from their current 178. Gains of greater than 60 seats also look quite possible. Even in the best case scenario for the Democrats, it would seem that holding the House would be very, very difficult.

It looks like a red blizzard is going to sweep through Washington in November.

Wow.