House Election Update - the Number Just Keeps Getting Bigger

In Are the Experts Underestimating the Republican Tide, I predicted that Republicans would pick up 50-60 seats. Looking closely at the data, I now think this may be low.

There are now 47 Democrat held seats where the Republican is leading in the polls. My assumption is that these races are basically over. For an incumbent to be behind at this point is the kiss of death. Still, assume 10% of the Republicans find a way to lose. That's a pick up 42 seats (rounding down).

There are another 34 seats where the incumbent Democrat is up by some margin, but are below 50%. If you're a stats/polling geek like I, you know what this means. It means huge problems for the Dem.

Let's look at an example, at North Carolina's 8th District. A poll just came out today showing the Democrat, Larry Kissell, ahead of the Republican challenger, Harold Johnson by one point, 45-44. If Kissell hasn't convinced 50% of his district's voters to re-elect them by now, he isn't going to in the next four weeks. The 11% who are undecided have already decided to vote against Kissell; they just haven't figured out who's running against him yet. That's how these things always work. Undecideds this late in the game almost always break against the incumbent. Look up the NC 8th on November 3rd. You'll see.


But let's be conservative and say that Republicans only manage to pick up half of these districts. that's another 17 seats. We're up to 59.

There are another 9 seats where the incumbent Dem is over 50%, but not by much. I'll give the Repubs one of those. That's 61.

Then there are 28 Democrat-held seats that are "of interest," but where there's no available polling data. Some of these include races that are rated as "toss ups" by Cook Report. I think roughly 7 of these might be picked up by the Republicans. That's 68.

Lastly, there are maybe 3 seats that Dems may pick up from the other side.

Final number: 65.

I'll update this as the election grows closer.