Well - last night surprised me!

I was not a lot wrong. But I was wrong enough. Why did this happen? In retrospect, I see now that I made two analytical mistakes - one theoretical and one methodological.

(1) I overestimated the extent to which our electoral institutions would mitigate GOP loses. I never doubted that (a) the public was in a foul mood, (b) they blamed Bush and (c) this would induce GOP loses.

However, my intuition at the time was that, at least in the House, this would reduce the extent to which the GOP would suffer loses. It did. But not as much as I thought it would. They lost about 10 or so more seats than I thought they would, and about 2 seats more than my 95% range of possibilities. The error here was my overestimation of the change in our electoral structure that 1994 produced.

As it turns out, Charlie Cook did not jump the shark. I did! I let my "institutional bias" take me right over a damned shark! Sorry, Charlie!

(2) I recently put together an estimate of the House playing field based upon challenger financing and party activity. Going into yesterday, I was using this as my "crib" sheet. However, and much to my chagrin, the list was not complete. It missed several seats that switched last night - IA 02, KS 02, MN 01, NH 02, NY 19, NC 08 (almost!), and PA 04. The divergence between the range in my final estimate and the actual result is entirely explicable by the seats my list missed. Where did I go wrong?

I did not include a measure for incumbent financing/activity. If I had, I think I would have picked up on many of these races. The GOP seems to me to have lost all of these because the Republican incumbents were not as active/effective as they could have been. They did not accurately assess the threat that they faced and/or did not take enough steps to mitigate the threat. Others, like Jim Gerlach, Chris Shays and Heather Wilson did - and they survived. Theoretically, the mistake I made here was to presume that the incumbency advantage that obviously exists (this year's incumbency reelection rate is still about 95.2%) is automatic. Incumbents are in a good position to insulate themselves. But they are not automatically insulated. They must actually do the insulating.

From my scan of the seats that flipped, I think that this election supports the theory of Gary Jacobson and Samuel Kernell, which I have discussed at many points in time during the campaign season. Our House elections are not referenda strictly speaking. They are qualified referenda - the qualifications are (a) good recruitment, (b) good fundraising and (c) good campaigns. If they were true referenda, the GOP would have lost many more seats than they actually did. Fortunately for them, most voters did not get a true choice last night because their Democratic challengers were under-funded and under-qualified relative to their incumbents. The incumbents who lost were the incumbents who either faced strong challenges or who themselves ran very weak campaigns.

Indeed, by my count -- there were only 3 Republican incumbents who ran essentially flawless campaigns and nevertheless lost: Nancy Johnson, Mike Sodrel, and Clay Shaw. Mike Fitzpatrick and Rob Simmons both appear headed for loses, so I would add them to the list. The rest of the Democratic pickups, 83.33% in all, were pickups in either (a) open seats, (b) seats held by scandal-ridden incumbents or (c) seats held by ineffective campaigners.

Thus, Republican mistakes, specifically campaign-related mistakes, very clearly were a major factor in the loss of the House. However, my feeling is that the mistakes that were made were the kind of mistakes that are inevitably made in our type of politics. The political parties really have much less power than people think. So, when people blast Tom Reynolds for not forcing Don Sherwood to step down -- my response is, what could Reynolds possibly have done? All you could ultimately do is appeal to Sherwood to bail. You cannot force the guy out. Ditto with Republican incumbents like Leach, Hostettler and Bass. None of them raised nearly enough money to survive this kind of environment. But what was Reynolds to do? Force them to go to fundraisers? These guys are really responsible to and for themselves. They are not like children. Candidates are largely independent of parties today.

I think the reason that the GOP lost so many seats that they "should not" have lost is that many of these incumbents have not faced real challenges since they were elected. Some of them have never faced real challenges. Accordingly, they just were not ready.

Call it evolutionary electoral politics. The strong survive when conditions turn against them. The weak do not. Last night, almost all of GOP loses were their weak seats.