Two days ago I published a diary noting that there were 41 House races within 5 points. At the time I said that the House was still very much in play. This diary greatly expands this analysis. What it shows is that while the GOP leads, the Democrats can still win the House.
This analysis uses the ratings of Charlie Cook as a yardstick. It is important to note that there is little public polling on House Races. In most CD's, there has been only one public poll. While Cook bases some of his analysis on campaign polling, we should not make the mistake of thinking that we know nearly as much about House races as we know for Senate. As a result, ratings for the House are based on less data, and as a consequence are subject to greater error. Nate Silver has noted this recently.
This table shows the polling averages for the different Cook ratings. The averages are based on all public polls taken since September 15th. On the surface, this appears depressing. The GOP leads in 26 of 42 of the tossup races (2 are dead even).
This table predicts the makeup of the House. Since we don't have polls for all of the CD's, this projection uses the current breakdown in each category to generalize the results. For example, their are 47 tossups. The Dems currently lead in 35% of the tossup seats, so I assumed they will win 35% of 45 of the tossup seats (2 tossups seats are tied, I assume the Dems win one). I want to focus your attention to the bottom number: the Seats That Need to shift for Dem Control. Because when you look at the tossup races, you find that it is more than possible that the Democrats could win the 15 seats they would need to hold the House.
This table summarizes the current state of play in the races Cook labels Tossup, Lean Democratic or Lean Republican. What becomes apparent is how close most of these races are. The GOP leads in 15 Seats that are within the margin of error (less than 3 points) and in another 6 they lead between 3 and with 5 points. If the Democrats were to win both seats that are tied, they would need to win 14 of 21 seats where the GOP leads less than 5. This is difficult, but by no means impossible. I can think of at least two elections (2006 Senate, 1982 Senate) where a party won most of the close races.
Again, this is not likely, but it is more than possible. The simple truth is that the house is more than winnable.
GOTV!!!
The data is here:
https://spreadsheets0.google.com/...