What is the likely outcome of the House races?
OnTheIssues predicts the Democrats will gain 7 House seats, leaving control of the United States House of Representatives in Republican hands, 235-200. Our evidence is laid out below.
So why do the Democratic pundits claim that they can gain the 25 seats necessary for their party to gain control of the House? For example, the Kansas City Star on Oct. 24, 2012, cites a Democratic Congressional Campaign Committee (DCCC) spokesperson saying "The Republican majority is in jeopardy," and expressing that he is "confident that his party can buck the odds and pick up the 25 seats needed to regain control of the House of Representatives." Let's look at the sort of evidence they present, starting with the current party split -- a large Republican majority -- of 242R-193D:
OnTheIssues conducted a "vulnerability analysis" of all incumbent House members. We define an incumbent as "vulnerable" if their district supported the opposing party's nominee in both previous presidential elections. A Republican House member is vulnerable if his district voted for Obama in 2008 and Kerry in 2004, and vice-versa for Democrats. We estimate the degree of vulnerability by the percentage of the opposing party's presidential victories. For example, Ben Chandler, a Democrat in Kentucky's 6th district, has a vulnerability score of -29: McCain beat Obama 58%-41%; and Bush beat Kerry 55%-43%.
A vulnerability score of -29 -- the most vulnerable incumbent in the country -- is VERY vulnerable. For the Republicans, there are 26 incumbents who fit our vulnerability criteria. Here are the top 5:
(The rest of the 26 can be seen at our vulnerability analysis page). The Democratic pundits at the DCCC could, according to our OnTheIssues analysis, claim that there are 26 vulnerable Republican incumbents, and if all of them lose, the Democrats would gain the House majority. And that is true -- it would result in a House split with a slight Democratic majority, 217R-218D.
This year also marks the first electoral results of the 2010 census -- called "redistricting." We analyze redistricting in detail below, but it has already had a substantial effect in the House primaries, because many pairs of incumbents had to run against each other. That knocked out 5 incumbent Republicans and only one incumbent Democrat. That provides a cushion of an additional 4 seats above the 26 gains due to vulnerability -- so it looks possible for Democratic majority -- but it's all only half-truths.
Of course there are some Democrats who are vulnerable also. The DCCC has a point that there are more vulnerably Republicans (26) than vulnerable Democrats (6), but the DCCC's desired results would only occur if all of these Democrats win re-election also:
Those 5 Democrats are very vulnerable; but the DCCC could (and will) defend their seats. However, it gets worse for the Democrats. Five vulnerable Democrats retired; another one changed districts; and two lost their primary races. That's 8 additional seats which no longer have Democratic incumbents running for re-election, but which are likely to change party to Republican -- and in fact, are more likely because there's no incumbent to defend the Democratic seat. Counting all vulnerable Democratic seats, not just the Republican ones, means the Democrats will gain only 12 seats, not 26 -- that means a Republican House majority of 230R-205D.
But it's even worse than that, due to redistricting. The census reflects population movements by redistricting House seats every 10 years. In 2010, compared to 2000, millions of people moved to "red states" from "blue states." OnTheIssues defines "red state" as "having a Republican majority in their House delegation" and defines "blue state" as "having a Democratic majority in their House delegation." That is a relevant definition for House elections because the House delegation will work hard to get their own party into the new seat, or to force the opposing party to lose a seat if the state is losing representation. Here are the relevant states:
| Blue states gaining seats | Blue states losing seats | Red states gaining seats | Red states losing seats |
| WA (+1) | IA (-1) | AZ (+1) | LA (-1)* |
| | IL (-1) | FL (+2) | MI (-1) |
| | MA (-1)* | GA (+1) | NJ (-1) |
| | MO (-1) | NV (+1) | NY (-2) |
| | OH (-2) | SC (+1) | |
| | PA (-1) | TX (+4) | |
| | | UT (+1) | |
Due to redistricting, the Democrats will net lose 6 seats and the Republicans will net gain 6 seats. That makes our running total 236R-199D.
There are a few exceptions based on individual races, indicated by asterisks above. In Massachusetts, the Republicans cannot lose a seat because they don't have any House seats in that state. And in Louisiana, redistricting resulted in two incumbent Republicans running against each other rather than trying to oust a Democrat. That's a net effect of two more Democratic seats. Also, four Democrats switched districts, while only one Republican did -- they are partly vulnerable with a net effect of one more seat for the GOP. Hence our overall prediction: 235R-200D.
OnTheIssues does not claim to be prognosticators -- but we do claim to have real substantiation for our prediction, especially when compared to the hocus-pocus of other pundits. Our vulnerability analysis has worked to identify incumbents in our local State House who have been ousted; the redistricting analysis is complicated but accurate. We would be very shocked if the Democrats do better than 230R-205D or if the Republicans do any better than 240R-195D. More extreme results than that would indicate a "landslide mandate" for one party over the other.
