Saturday, February 6, 2010

Latest 2010 Rundown

We are 269 days from general election night in the 2010 mid-terms. Here is the latest run-down.

Since my last full update, I'd already documented 3 individual race changes: Indiana and Wisconsin moving from Likely DEM Hold to Lean DEM Hold with Evan Bayh in a surprisingly close polling race against potential GOP challengers and Feingold actually slightly trailing potential challenger Tommy Thompson and Delaware moving from Toss-up to Lean GOP Pick-up with Beau Biden not running on the Democratic side, leaving the party no strong rival for popular At Large Rep. Mike Castle (R).

We have 2 new updates to add to that list, 1 favorable to the GOP and 1 favorable to the DEMs.

Illinois moves from Lean DEM Hold to Toss-up following Mark Kirk's win in the GOP primary and a subsequent Rasmussen poll showing him leading the race by 6%. One close poll is not enough to swing this race to the GOP, but this one is clearly up for grabs.

Nevada moves from Likely GOP Pick-up to Lean GOP Pick-up - Harry Reid has experienced a quiet, partial resurgence and is trailing the two potential GOP challengers by a range of 6 to 8% in recent polling.

We have also seen a number of polls confirming our current race ratings in states including Colorado, Arkansas, North Dakota, New Hampshire and Kentucky.

This leaves us with the following state of the race:
Safe Democratic Hold - 6
Connecticut, Maryland, New York (Schumer), Oregon, Vermont, Washington

Likely Democratic Hold - 1

Lean Democratic Hold - 3
California, Indiana, Wisconsin

Toss-up - Dem Controlled - 3
Colorado, New York (Gillebrand), Illinois

Toss-up - GOP Controlled - 1

Lean GOP Pick-Up - 4
Arkansas, Nevada, Pennsylvania, Delaware

Lean GOP Hold - 5
New Hampshire, Kentucky, Ohio, North Carolina, Arizona

Likely GOP Pick-Up - 1
North Dakota

Likely GOP Hold - 5
Georgia, Alaska, Kansas, Louisiana, Florida

Safe GOP Hold - 7
Iowa, South Dakota, Alabama, Idaho, Oklahoma, South Carolina, Utah

Net Projection: GOP Pick-up of 4 tp 8 Seats (Central Projection +6)
Best Case GOP (all leaners): GOP +11 Seats
Best Case DEM (all leaners): DEM +5 Seats

For the first time in this election cycle, I can actually envision a scenario where the GOP could retake control of the Senate. It's still a very long shot -- it would require them winning all the toss-ups and picking off at least 1 of the 3 Lean DEM holds - Indiana, California or Wisconsin. But it is within the realm of possibility.

Of course, with the momentum behind the GOP, everyone is ignoring the other possibility -- that the race could break back to the left before November. The GOP stands to make big inroads in the Senate today, but there are many, many competitive races and as you can see from our "Best Case" scenarios, the difference between the GOP having a 52 to 46 majority in the Senate and them being relegated to a 62 to 36 minority is just a few percentage points nationally.

If the Democrats did not have an inherently favorable map (they would argue with this, but I'll explain), this might be a far worse situation for them. The way the election cycles have worked however, with their big gains coming in 2006 and 2008 and the 2010 race essentially a replay of the races from 2004, which was a more favorable year to the GOP, the GOP has more turf to defend and less targets to go after than they will in the next two Senate cycles.

Of course, if Democrats can lose Massachusetts, be trailing in Delaware, be even in New York and only slightly leading in California, no map can solve that problem. They are going to have to rediscover at least a little mojo or we could see a real shift on election night.

Our latest average of averages in the generic ballot question shows the GOP at +3.9%. Based on my statistical modeling this implies:

GOP Pick-up of 44 Seats

So at a high-level, I continue to project the GOP to narrowly retake the House in November, although the issue is certainly far from decided.

Since it is very difficult with 435 House races happening simultaneously, I have never attempted to do a race-by-race rundown, but it is worth looking at the reports of those who do. The Cook Political Report is one of the best sties at tracking race-by-race dynamics in House races. Cook currently lists 113 Hosue races as competitive, 24 controlled by GOP members and 89 currently controlled by Democrats. Of those 113 races, he has them rated as follows:
Likely DEM - 37
Lean DEM - 27
Toss-Up - 22
Lean GOP - 10
Likely GOP - 17

So if those ratings were to hold, it would imply a GOP Pick-up of 3 to 25 seats in November.

Of course, the challenge in this is that it is extremely hard to get recent, accurate polling in ANY House race, let alone 113. The "Lean" races could easily shift in either direction and even the Likely races could actually be a lot closer. Cook does a great job trying to document everything, but to a large extent, there is a heavy lag in these numbers. Cook is also well known to be very conservative in his calls - not ideologically mind you, but rather slow to shift races away from the incumbent party -- for instance, he still had the Brown/Coakley race listed as a "Toss-Up" going into election night, when we were solidly projecting a Brown win.

The high end of Cook's range is a feasible scenario -- I could certainly see the GOP "only" picking up 25 seats if the elections were held today. The low end, seems unfathomable in the current environment.

At any rate, the GOP stands to gain as things stand now, the debate is simply over how much.

Double-Witching 2012
Every 4 years, we elect a President in this country. Every 10 years we take a census, which reapportions House seats to the states, thereby leading to all the legislative districts being redrawn, as well as shifting the electoral vote count between the states (since electoral votes are determined by number of Representatives plus Senators.) Therefore every 20 years we have the "double-witching" effect of a Presidential race coupled with the newly redrawn map form the census.

The last time this happened was 1992 and it will happen again in 2012. These races can lead to a high-degree of change as you have generally high voter turnout for a Presidential election coupled with new faces in Congress due to the combination of existing districts and the creation of new districts.

2012 is still a long way away, but it's just interesting to note that although 2010 is shaping up to be a very significant election in terms of political change, 2012 has all the elements to change the game even more.

Next Up -- Assessing the newest unemployment report and what it means, as well as our regular update on Presidential approval

No comments: