He makes his case here. There is of course a lot of guesswork involved, but I think this part of his analysis is solid:

[O]ne could argue that if one simply returned to the dismal, scandal-ridden 2006 environment with that same electorate, we’d be 10 to 20 seats better off than we are now. Now, start factoring in stuff like Republicans tied or leading in the generic ballot, which they hardly ever were even in years the successfully held the House, like 2002 and 2004. And more tellingly, the bumper crop of good candidates that’s stepped forward after the drought of 2006 and 2008. 

I’ve argued thus far that political whiplash may be greater this year. But in truth, it may not be that much worse than the utter Republican collapse from 2004 to 2006. That collapse produced a loss of 30 House seats. But the starting point was a stable equilibrium established over 5 successive election cycles without a double digit gain in seats by either party. The starting point in 2010 is a very unstable one where Democrats have accumulated more than 50 new seats in four years, over 20 of them somewhat artificially because of the Obama coalition.

I think he is also right that under the circumstances the right move for Republicans is to expand the map. But what’s even more important to my mind is that the candidates stand for a program of practical conservative reforms.