More Election News

• FL-24: Money Fuels Race to Unseat Kosmas
• CA: Brown Rips Whitman's Wealth but Has a Bit of His Own
• CO: Is an Upset Brewing in Democratic Primary?
• CA: Barbara Boxer Faces Her Toughest Race Yet
• NM: Race Gives Big Donors Last Splash Before New Laws Kick In
• TN-8: GOP Seeks Unity for General Election
• NY: Key Unions May Pass on Andrew Cuomo
• NV: 'None of the Above' Vote Could Play a Role
• CO: Big Money Pours Into Swing Races
• MI: Candidates on Prowl for Undecideds
• KS: Wichita Could Sway GOP Primary
• WA-8: Familiar Themes in Race for Politically Ambivalent District
• OK: Fallin, Askins Race Takes Shape

Democrats are trying to nip in the bud the notion that their majorities are vulnerable.  Charlie Cook has written about their arguments twice this week, and Greg Sargent has an article up at the Washington Post.  The gist of the argument is as follows:

-- Four Republican seats -- DE-AL (open), LA-02 (Cao), HI-01 (Djou) and IL-10 (Open) are in the bag, raising the number of seats that Republicans need to win to 43.

-- Additional Republicans will probably lose, such as Dan Lungren.

-- Given the DCCC's strong performance in special elections, the Democrats' worst-case scenario in their open seats is eight of the following seats: LA-03, TN-06, TN-08, AR-01, AR-02, IN-08, KS-03, NH-02, MI-01, PA-07, WV-01, WA-03, WI-07, MA-10, RI-01, and NY-29.

-- Republicans can't knock off the 35 incumbents they would then need to defeat to win back the House, because they'd have to put 70 seats in play to do so.

I understand the DCCC's desire to make it seem like holding the House is likely -- it tamps down GOP excitement and avoids a discouraging meme that might tamp down Democratic excitement.  And to be clear, I think it is perfectly plausible that the Democrats might limit their losses to twenty or thirty seats, and plan to write on this in the next week.  But the idea that this is somehow their worst case scenario is simply false.  Let's look at these arguments individually.

(1) The Democrats definitely have a great shot at taking most of the four seats listed in the first bullet.  But it is by no means a done deal, and I wouldn't be at all surprised if the GOP wins a couple of those seats.  The most likely seat they keep is IL-10, an open seat in the North Chicago suburbs.  It is pretty Democratic at the national level -- though not overwhelmingly so -- but it still has some Republican strength at the local level.  Moreover, the GOP nominee, Bob Dold, is running against a Democrat who couldn't get it done in two much more favorable years (albeit against a strong Republican incumbent). In the other open seat, Lt. Gov. Carney is a statewide officeholder who is having trouble getting to 50% against unknown GOP opponents.

In Hawaii-01, which has voted Republican for Governor in four straight elections, Charles Djou is betting that he can pick up most of the supporters of moderate Rep. Ed Case against the more liberal Colleen Hanabusa.  And in LA-02, Anh Cao has released an internal poll showing him above 50%.  I'm not sure how much I believe the poll, but I also wouldn't have thought that even the craziest poll weighting could have gotten Cao above the 50% mark (and the internals on this poll look reasonable).  In addition, there may be some third party African American candidate who siphons off the anti-Cao vote, allowing him to win with a plurality.

RealClearPolitics lists Cao's seat as Likely Democrat, DE-AL and Djou as Leans Democrat, and IL-10 as a tossup.  We're actually less forgiving than Cook, who has three of the four as tossups (though Cook rarely puts an incumbent in the takeover slot until the end of the cycle), or Stu Rothenberg, who has Djou and IL-10 in the tossup category.

The point is, none of these seats are in the bag for the Democrats, and I think the odds of the Democrats taking all four are smaller than you'd think.

(2)  I can be brief here.  I think there's a very good chance that, aside from the two incumbents just discussed, none of the Republican incumbents will lose.  We saw a similar effect after 1994; for the rest of the decade only a handful of Democrats lost, because the Democrats who survived 1994 were generally strong incumbents who fit their districts well.  And aside from the two open seats above, the remaining GOP open seats at least lean toward the Republicans; in this environment they should be pretty solid for the GOP.  RCP lists only three Republican seats in the "Leans GOP" category, Rothenberg lists three, and Cook lists three (interestingly, no seat is listed as vulnerable by all three).

(3) The DCCC's argument gets a bit disingenuous here.  Let's look at these open seats.  We might expect the Democrats to win half or so of these seats if they were all tossups, and we might cap their losses at half if we really believed that the DCCC was far superior to the NRCC.

But these races aren't all tossups.  RCP has two of them (LA-03, TN-06) as Likely Republican, and seven (AR-02, IN-08, KS-03, NH-02, NY-29, PA-7 and TN-08) as Leans Republican.  The only one I might question is TN-08, where the GOP has a divided primary and there's a long history of local support for Democrats.  Except for RI-01, the rest are tossups.  Rothenberg and Cook are largely in agrement with us, if a bit more generous to Republicans.

We might expect the tossup races here to all break toward the Democrats if we believed that the DCCC was somehow inherently superior to Republicans in open seats.  They have a great track record over the past few cycles.  But this track record comes with some caveats.  From 2005-2010, they only ran three special elections in a negative environment (Obama was still popular for the NY-20 election).  One they lost, and one (NY-23) they won because of a bizarrely-divided GOP, where the Republican candidate dropped out and endorsed the Democrat on the eve of the election.  That leaves PA-12, which I've discussed at length.  In short, I'm not convinced it's as surprising a win as many make it out to be, but regardless, one good win does not a trend make.

In other words, losing eight open seats is probably the floor for Democrats, not the ceiling.  Eleven or twelve is probably more likely.

(4) Finally, even if the GOP really does need to defeat 35 incumbents, I'm not sure why people think this can't be done.  If the GOP wins the seats RCP currently has as "Likely/Leans GOP," loses the "Likely/Leans Dem" seats, and splits the tossups, it would pick up 39 seats.  And it isn't at all clear what 70 Democratic seats couldn't be put in play:  Cook sees 66 Democratic seats currently in play, Rothenberg sees 76, and RCP sees 82 (like Cook, our "Likely Democrat/Republican" designation can be thought of as a "watch list").  Granted, it's rare to lose that many incumbents, but as I've written before, the factors at work this year are also rare. 

The bottom line is that unlike 1994, when the GOP had to break new ground to take back the House, all the GOP has to do to win back the House is win 2/3 of the seats that it held four years ago.  In this environment, that's not a particularly tall order.