Month: November 2010

Okay, so, Maryland has the trifecta, and it’s not afraid to use it. Republicans get to carve up some states they control in particularly nasty, ugly ways (I’m not looking forward to seeing how Augusta determines to create a 1-1 split, probably ousting Rep. Michaud in the process; North Carolina will be a veritable bloodbath in redistricting, and New Jersey won’t be much better), so I think Democrats should get to fight back in states like Maryland, Washington, and hopefully New York.

So, without further ado, I present to you a positively wicked 7-1 redistricting proposal for my state of residence.

Let’s start with MD-01. I guess the app decided to plop the “CD01” right where D.C. is, so just to be clear, I’m not advocating Maryland annex the nation’s capital as its new 1st congressional district.

MD-01 is pretty straightforward. I chopped most of the population centers out of Rep. Bartlett’s MD-06, including the swing city of Frederick, and combined it with Rep.-elect Harris’s Eastern Shore district, now with 100% less Anne Arundel County. Back to that in a second.

The important thing about MD-01 is that it’s the Republican district. According to the app, this hypothetical new MD-01 gave Sen. McCain 58% of the vote in 2008, with only 40% for the skinny Illinois senator we now know as President Barack Obama. This MD-01 would not be the safest Republican district in the country, but it’d be pretty safe.

Next up, MD-02. It was roughly at this point that I realized how messed up the current congressional districts of Maryland are. It’s bad enough to include parts of Anne Arundel County in a district with the Eastern Shore, but the way in which Rep. Ruppersberger’s district is drawn would be punishable by death in some countries.

What I decided to do is simplify things without sacrificing, well, let’s call it functionality. I also needed a way to swallow up some populous parts of MD-01, namely most of Baltimore County and parts of Harford County. This MD-02 includes parts of Baltimore City, although less than is currently drawn into the district because it no longer slithers south into Anne Arundel County. I mean, come on, guys.

The tradeoff of consolidating MD-02 with an eye toward siphoning off conservative MD-01 constituents and balancing them with ultra-liberal North Baltimore (the parts of Baltimore City included are a mix of super-white and super-black precincts, because a Democrat is a Democrat and we don’t make bones about it in MD-05) is that the district isn’t quite as solidly Democratic, but it’s hardly a swing district in an average year. Obama won this hypothetical MD-02 with 57% of the vote; McCain scraped up 41%.

And now for MD-03. If the current MD-02’s blatant and bizarre gerrymandered shape merited the death penalty, the lawman would bring back hanging, drawing, and quartering for MD-03. Staunchly liberal Rep. Sarbanes holds court over this district, drawn for the benefit of white dudes just like him, right now.

Because the way MD-03 is drawn right now is just plain stupid and there was still some of Harford County that needed addressing, I decided to run the new MD-03 from a tiny portion of western Cecil County, technically part of the Eastern Shore, into southern Baltimore County, with some spillage into northern Anne Arundel County and a weird little tentacle extending briefly into Howard County. It also includes the lion’s share of Baltimore City. I’ll grant it’s not tons better-looking in my drawing than it is now, but at least I didn’t gerrymander it for the benefit of those poor besotted white Democrats.

My version of MD-03 is DemStrong, having gone 60% for Obama to 38% for McCain in 2008. It’s now 30% black, proportionally almost twice what it is now, but if Sarbanes or some other white Democrat can’t get African Americans to vote for him, that’s his problem.

On to MD-04. I don’t know why this district is drawn so weirdly right now, because I feel like the way I’ve drawn it makes a lot more sense. Oh, I’ve got it: it’s gerrymandered to be majority-black. Rep. Edwards, who is a damn good congresswoman, represents our current version of MD-04.

My version of MD-04 covers about half of Montgomery County, a few conservative precincts in Howard County I didn’t want going elsewhere, and most of northern Prince George’s County. It would become the new home of the University of Maryland, which doesn’t hurt a Democrat’s vote share.

Indeed, this version of MD-04 went for Obama 75-24. That’s dominating. I could probably feed more of Howard County into this district if I wanted to, but I decided to keep things as even as possible between the eight (2000 population for all is between 661,000 and 663,000, by the way).

MD-05 is my district. I switched around some parts of the eastern D.C. suburbs for compactness, although my incarnation of the district now represented by Rep. Hoyer, the House majority leader for not a whole lot longer, still has a weird finger extending into northern Prince George’s County, capturing Greenbelt and a few other mid-sized suburbs.

The majority-black eastern suburbs more than balance out swingy southern Maryland in my drawing of MD-05. This new district would actually be majority-black according to the 2000 Census, which says 51% of its hypothetical constituents are African American.

As you might expect, this MD-05 is heavily Democratic. It would have broken 77-23 for Obama if it had been a district back in 2008; either way, that’s what the voters who would be within its new boundaries did.

MD-06 is fun. It’s a completely different district in its present form, being the domain of arch-conservative Rep. Roscoe Barlett in northern and western Maryland. As mentioned before, I had MD-01 consume most of MD-06, so this new district is more fun.

This MD-06 is not the most fun, but it’s certainly the most all-new, because I basically carved it out of parts of the current MD-01, MD-03, and MD-04. It’s basically almost all of Anne Arundel County plus a good chunk of Prince George’s County, including Bowie, a midpoint city between D.C. and Baltimore.

Ultra-liberal Bowie and somewhat liberal Annapolis serve to balance out conservative Anne Arundel County. Voters in this hypothetical MD-06 broke 57-41 for Obama, which is not so bad when you consider the current version of MD-06 has a PVI of R+13.

Helping mop up is MD-07. Right now, MD-07 is a majority-black district represented by Rep. Cummings and covering most of the black neighborhoods of Baltimore City and Baltimore County, as well as most of Howard County. It had to take a hit, and I apologize for that.

There was a lot of the current MD-06 to split up, and a good portion of it was awarded to this proposed MD-07. It’s basically the current MD-07 with a few neighborhoods in Baltimore City split off and given to MD-02 and MD-03, then a huge swath of Carroll County and about a third of Frederick County absorbed for population balance. It’s a lot bigger, but it’s not abnormal population-wise (still falls within that range I mentioned earlier, at least under the 2000 Census).

Despite expanding to cover so large an area of rural, conservative Carroll County, this MD-07 would remain a Democratic stronghold. It broke 2:1 for Obama in 2008. It’s only 39% black with so much of mostly-white Carroll and Frederick counties added in, but a Democrat should be able to walk away with any election here regardless of race.

Last but not least, MD-08. Currently represented by plucky DCCC chair Rep. Van Hollen, the present version of MD-08 covers most of Montgomery County, in particular the eastern, whiter parts of Maryland’s richest county-level subdivision.

I decided I could put it to better use, so I gave parts of Montgomery County to MD-04 without radically changing the boundaries, and then I expanded up into Frederick County and parts of southern Washington County. This is where the city of Frederick, which broke fairly decisively for Obama in 2008, wound up.

The new MD-08 would have a 2008 voter breakdown almost identical to that of MD-07. 66% of its voters went for Obama; 32% of them went for McCain. It’s only 10% black, because of how predominantly white Frederick County and the parts of Montgomery County is covers are.

So there you have it. It can be done. Let’s hope Annapolis pulls the trigger, because we shouldn’t let the Republicans have all the fun in 2012 redistricting.

The lone spot of good news in all this for Dino Rossi? Now he won’t have all those Senatorial duties cluttering up his planning activities for his 2012 gubernatorial run.

Sen. Patty Murray pulled further ahead this afternoon , taking 68 percent of the nearly 70,000 votes counted in King County on Thursday.

The King County boost widened Murray’s statewide lead to more than two percentage points — nearly 50,000 votes — over Republican Dino Rossi statewide even as Rossi-leaning counties began to report new totals. The biggest Rossi counties, Spokane and Clark, both reported results today but his gains there were swamped by King County.

Also worth noting: Rick Larsen in WA-02 seems to be moving into healthier shape against John Koster. He’s now up more than 1,400 according to the SoS, suggesting that Dem-friendly precincts are late to report in the 2nd too.

• IL-Gov: The DGA is sending ’round a press release congratulating Pat Quinn on his who’da-thunk-it re-election victory. No official call yet, but the trendlines in the count (with a margin now of 20,000 votes) make it pretty clear what’s happening. Bill Brady says he isn’t ready to concede… yet. UPDATE: CNN has called it.

• CT-Gov: More vacillation in the Connecticut count, this time with Dan Malloy back on top by about 6,000 votes. This doesn’t seem to have anything to do with actual counting, but simply with the AP getting its facts straight (it seemed to be missing a large slab of urban and assumedly heavily Democratic votes, mostly from New Haven). Final tallies remain unknown because we’re still waiting on Bridgeport (another Dem stronghold), although SoS Susan Bysiewicz has promised results by the end of the day. (Doesn’t seem like she’ll make it, though.) Relatedly, Dan Debicella in CT-04 is “waiting and seeing” what the Bridgeport situation is before conceding, although that race has already been called against him.

• FL-Gov: This was pretty well decided yesterday, but Alex Sink officially conceded, finalizing the transaction on Rick Scott’s self-financed purchase of the gubernatorial election.

• NE-Sen: Ben Nelson says he won’t switch parties. (Why the heck would he want to, considering that the Dems still have a Lieberman-proof majority in the Senate, that he’s part of? He’d just get teabagged to death in the 2012 GOP primary a la poor Parker Griffith.) It’s worth wondering, though, if any House Dems are considering a switch… but, with the possible exception of Dan Boren, all of the most likely suspects have already lost, saving us the trouble of such speculation. UPDATE: Although here’s an interesting post-script to the Nelson story: GOP Gov. Dave Heineman, who’s led a few hypothetical polls against Nelson for 2012, has already announced that he won’t run for the Senate in two years.

• NV-Sen: The one major race where the polls really seemed off was Nevada, where Harry Reid’s 4-pt average deficit turned into a 5-point victory and in retrospect, that’s easy to explain: pollsters weren’t reaching Latinos. Exit polls suggest that Latinos did turn out well in Nevada (thanks in no doubt to the Reid-friendly hotel unions in Las Vegas). Nate Silver graphs the variance in other states with large Latino populations: it may also have played a role in Colorado, although the poll variance wasn’t as bad there, though it was enough to predict a Ken Buck victory… and Latino turnout didn’t seem to work in Texas, where Rick Perry exceeded polling predictions (and which explains the sudden and probably brief appearance of Rep. Blake Farenthold).

• KY-06: Here’s a race that’s definitely going to recount. Andy Barr, trailing Ben Chandler by about 600 votes, has formally requested a “routine recanvass.”

• NC-02: Here’s one more race where we’re not quite dead yet. Bob Etheridge hasn’t conceded, it turns out, nor should he: he’s down only 1,646 votes, after cutting his deficit against Renee Elmers by more than 500 in late counting yesterday.

• NC-11: When Heath Shuler announced that he’d challenge Nancy Pelosi for Speaker, the proper response was laughter, although my main question at the time was whether he planned to challenge her for minority leader. It sounds like, yes, he does still plan to go through with his kooky gambit (made all the kookier because there aren’t any core Blue Dogs left besides him who need to follow through on their promises to not vote for Pelosi). Anyway, the scuttlebutt seems to be leaning toward a voluntary sword-falling and a painless transition to Steny Hoyer, in which case the Shuler challenge would be even more pointless.

• WV-01: After maintaining for most of yesterday that he wasn’t dead yet (given the narrow margin, currently 1,357), Mike Oliverio did wind up conceding eventually late yesterday.

• Redistricting: As much as the mess in the state legislatures really messes up hopes of wresting an advantage out of the redistricting process, here’s some good news: Amendment 6 in Florida surprisingly passed, clearing the super-high 60% hurdle at 63%. This doesn’t create a commission, but it does impose requirements that redistricting be done without political consideration (but without violating federal law, namely the Voting Rights Act). However, two U.S. Reps… probably the two in Florida who have the most to lose from having less friendly districts drawn for them… Corrine Brown and Mario Diaz-Balart, are challenging the law in court. With the passage of Proposition 20 in California, though, which creates an independent redistricting commission there, it’s all kind of a wash, as a less-bad Florida map is balanced out by not being able to aggressively redistrict California. (Similarly, while it takes some sting out of Alex Sink’s narrow loss, it also deflates the impact of Jerry Brown’s victory.)

This is a diary by Stochastic Democracy, evaluating our final forecasts for the 2010 House of Representative Elections and comparing our predictions to those of other forecasters. Cross-posted at DKos, where you can find the pretty version of the post– SSP doesn’t buy all the HTML code. EDIT: Okay, DKos doesn’t buy it either, check out the StochasticDemocracy frontpage for the good style.DailyKos post.

While there are still a couple of uncalled districts out there, enough returns have come in to start looking at the results and how the model did:

Behold our new Congress, based on provisional returns, colored according to Democratic Vote-Share. Compare with map on side-bar that shows our projections. 

Swing in vote from 2008 to 2010. Red indicates districts where Republicans did better then last year, Blue indicates districts where Democratic standing improved, and White indicates no change. 

 Please click here to see a comparison of various predictions by John Sides of the Monkey Cage:

Stochastic Democracy predicted that Democrats would obtain 197 seats in the House of Representatives, 5 more than the currently projected outcome. This was the third most accurate forecast, 5 seats more accurate than FiveThirtyEight, and the most accurate one that also provided district-by-district vote estimates.

Red indicates the 10 districts that Democrats were predicted to win but did not, Blue indicates 5 districts Republicans were predicted to win but did not. White indicates the 418 correctly called districts, with 2 districts still uncalled (NY-25 (Maffei) and CA-20 (Costa).

Note that Dan Maffei currently looks likely to lose, which would add another miss to our tally, while Costa will almost certainly hang on, as projected by us.

While this method of grading makes our site look good, it’s not a very informative one due to two reasons:

1) Everyone could get a majority of seats right- not very hard to forecasts the outcome of elections in Manhattan or rural Nebraska. A map that gets 30 districts wrong, like Pollster.com seems to have done, would still  be mostly white, despite a pretty underwhelming performance.

2) The record number of competitive districts this year. If you have a bunch of 50-50 districts, it’s basically a matter of luck how many of them you get right. That’s why sites like 538.com and us are doing probabilistic forecasts: We’re giving both candidates credit for having a shot to win the district for their side. Therefore, a much better way to evaluate forecasters is to look at how well they managed to predict vote-share district by district.

District by district, our model seems to have over-estimated Republican vote-share by about half a point. This is because the regression model appears to have inappropriately applied a uniform national swing to heavily Democratic urban districts- in downtown Chicago or New York, Democrats always win about 80-20, no matter what happens in the rest of the country, really. If we restrict ourselves to the 357 districts where the result was between 25 and 75%,  the bias effectively disappears (4 hundredths of a percent toward the Republicans).

Mean absolute error was 3.2 points, while median absolute error was 2.6 points. This goes down to 2.8 and 2.5 respectively for the 25-75% districts.

But we provided full probability distributions as well as vote estimates for each candidate. It seems that our stated standard errors were quite accurate: 94.7% of results fell within our 95% confidence intervals. More generally:

5.2% of results came within our 5% confidence interval, 10.6% in our 10% interval, etc.

In other words, our standard errors were well calibrated. Not only were we not wrong very often, but we were able to predict precisely how wrong we would be. That’s really amazing and can’t be stressed much enough: While most models really agree roughly on the mean of most predictions, they disagree heavily on how that should be converted to a percentage chance of winning. Nate Silver essentially had a very, very wide variance around his projections, so that he gave even candidates who were solidly ahead only a 70% or 80% chance of winning. We got the win percentage exactly right- about 95% of the candidates who we projected to have a 95% chance to win won, and the same is true for any other percentage.

We don’t have easily assessible data yet for the forecasts of FiveThirtyEight or Electoral-Vote, the only other forecasters we’re aware of who provided district level forecasts, and so we can’t yet replicate this analysis to see how our performance compares. On the other hand, Andrew Gelman, a renowned statistics professor at Columbia University, has looked at FiveThirtyEight’s forecasts and found that Nate’s confidence intervals were too wide:

So, yes, Nate’s forecasts do seem underconfident! There were 37 races where he gave the Republican candidate between 60% and 90% chance of winning, and the Republicans snagged all 37. Apparently he could’ve tightened up his predictions a lot.

We don’t think that looking at win-percentages solely is a fair measure, since most races are not close at all and race outcomes are not necessarily independent. Gelman agrees:

But . . . before jumping on Nate here, I’d suggest some caution. As we all know, district-by-district outcomes are highly correlated. And if you suppose that the national swing as forecast was off by 3 percentage points in either direction, you get something close to calibration.

Some other notes of interest:

Stochastic Democracy’s House forecasts were really a blend of two almost independent models: 1) A fundamentals-based regression model that took into account Incumbency, past election results, Income, Cook Ratings, and the Generic ballot adjusted for House Effects. And 2) A Bayesian state-space model that filtered polls, taking house effects, design effects, and potential industry bias into account.

The poll model shows that Democrats should have gotten 205 seats, while the regression model forecasted 186 seats. Paradoxically however, the poll model outperformed the Regression model in 52% of the individual races where polling was available.

Yet for races where polling and the fundamentals differed considerably, ID-1 (Minnick), AL-2 (Bright), MS-4 (Taylor), VA-9 (Boucher) etc (And CT-5 (Murphy) on the Dem side), the regression was considerably more accurate. There are a couple of theories as to why that could be, and it’s definitely something we’ll be looking at in the next couple of days.

Absolute Error of our forecast in a district vs the Number of Polls in the District

It’s still the week of Election Day, which means it’s time to take an incredibly premature look at the Senate races developing in 2012. As you may have heard, the map favors the GOP, suggesting that the party may have their second big cycle in a row. However, theoretically, the Democrats had a favorable map in 2010 and they ended that without a single pick-up.

Instead of making predictions that are sure to be outdated in two months, let alone two years, I’m embracing the uncertainty. For each race, this preview will look at:

The best-case and worst-case scenario for the incumbent party; a rating on which scenario is more likely to happen (1 for worst-case, 10 for best-case)

Likely candidates, mostly made up of this year’s “Tommy Thompsons”. If you’ll recall, for most of 2009 and early 2010, pundits and observers thought the only way Russ Feingold would be in trouble is if Tommy Thompson ran. In nearly every state, there’s the “obvious” candidate(s) that conventional wisdom cements as the “only viable” candidate(s)-Schwarzenegger in California, Huckabee in Arkansas, Napolitano in Arizona, and these projections are more often wrong than right.

Retirement shock: every cycle features retirements, some expected and some unexpected. This looks at where such a retirement would fall. (1 for expected (like Bunning), 10 for completely shocking (like Bayh))

And for kicks and giggles, a “cattle call” to be preserved for posterity, so we can look at this diary 23 months from now and look at all the things we didn’t know would happen.

Democrats (and those who caucus):

California, Diane Feinstein

Best-case scenario: Given the lean of the state and disinterest among strong contenders, Feinstein fails to attract a major nominee and runs against a Dick Mountjoy-esque candidate.

Worst-case scenario: California’s budget woes only get worse under Gov. Brown, resulting in high dissatisfaction across the state; the GOP nominates a Hispanic candidate for President or Vice-President while other Hispanic national stars become more prominent, stopping or even reversing the hold Democrats have on the Hispanic vote; Obama’s victory in California is secured while he is mired campaigning full-time in a dozen battleground states away from the Pacific West, so he spends little time getting out the vote. Feinstein loses in a squeaker to a moderate, governance-minded Republican.

Scenario rating: 6 out of 10; Feinstein will probably pick up a more serious challenger, but have little problem defending her seat.

Possible candidates: Tom Campbell, Steve Poizner, Steve Cooley

Retirement shock: 5 out of 10; At age 77, she’s far from “old” for the Senate, and she stated she’s running again at a recent Boxer rally.

Race rating: DEM Favored

Connecticut, Joe Lieberman

Best-case scenario: Lieberman retires or runs and is easily defeated in the primary.

Worst-case scenario: Lieberman defects and runs as a Republican, or Lieberman runs third-party and who knows what happens.

Scenario rating: ??? out of 10; who knows?

Possible candidates: Chris Murphy, Joe Courtney, Ned Lamont, Sam Caliguiri, Tom Foley, Rob Simmons

Retirement shock: 3 out of 10; it wouldn’t shock anyone, but it’s not expected.

Race rating: Lean DEM

Delaware, Tom Carper

Best-case scenario: With no bench and Carper’s decision to run again, the GOP leaves the race unchallenged, or worse, nominates Christine O’Donnell.

ALTERNATIVE: Carper retires and Beau Biden decides to run for the seat.

Worst-case scenario: Carper decides to retire, and Mike Castle decides to run; having learned their lesson last time, he is unchallenged for the nomination.

Scenario rating: 8 out of 10; I think a Carper-Biden swap is likely, and the reason why Biden didn’t run in 2010.

Possible candidates: Beau Biden, Mike Castle, Christine O’Donnell, Michele Rollins

Retirement shock: 2 out of 10; Carper has repeatedly denied health issues but has had a long and successful political career. At age 63, he could easily run again but his retirement likely wouldn’t catch anyone off guard.

Race rating: Safe DEM

Florida, Bill Nelson

Best-case scenario: Rick Scott implodes as a Governor; Obama makes a major push to win Florida again; the GOP nomination is beset with infighting and a far-right and battle-fatigued candidate emerges from the scrum.

Worst-case scenario: The GOP nominates a solid candidate that keeps the party unified; the economy stays in similar shape; Marco Rubio is on the 2012 ticket or otherwise barnstorms around the state with his arm around the shoulder of the GOP nominee.

Scenario rating: 3 out of 10; Nelson will be a big target, and Crist’s defection left the state GOP remarkably solidified. There are a number of GOPers on the bench that could make this a marquee race.

Possible candidates: George LeMiuex, Jeb Bush, Connie Mack IV

Retirement shock: 8 out of 10

Race rating: Toss-up

Daniel Akaka, Hawaii

Best-case scenario: Akaka runs for re-election and Linda Lingle, Duke Aiona, and Charles Djou pass on the race.

Worst-case scenario: Akaka doesn’t run for re-election and Linda Lingle, Duke Aiona, or Charles Djou runs.

Scenario rating: 4 out of 10; Akaka says he’s “definitely running again”, but either way I think one of those three runs. Half of either scenario coming true would make this a 5 out of 10, but since I think Akaka’s retirement is moderately possible, I’ll downgrade it to 4.

Possible candidates: Linda Lingle, Duke Aiona, Charles Djou, Ed Case, Mufi Hannemann

Retirement shock: 4 out of 10

Race rating: DEM Favored

Ben Cardin, Maryland

Best-case scenario: Cardin is ignored in his bid for a second term.

Worst-case scenario: Some scandal tars the entire state party and Bob Ehrlich runs.

Scenario rating: 10 out of 10; Ehrlich has said he’s done with elections and Michael Steele is running for RNC Chairman again.

Possible candidates: Mary Kane

Retirement shock: 10 out of 10; Cardin is a freshman and will likely hold the seat as long as he wants.

Race rating: Safe DEM

Debbie Stabenow, Michigan

Best-case scenario: A brutal GOP primary made up of lackluster candidates produces a far-right but unqualified nominee.

Worst-case scenario: A pragmatic-minded GOP candidate gets the party unified behind him/her and runs a solid campaign against Stabenow.

Scenario rating: 3 out of 10; There’s always the potential for a sideshow, but Rick Snyder’s nomination and landslide victory provides a good template for the state GOP to use. If the party gets its act together, Stabenow will be in real trouble.

Possible candidates: John Engler, Terri Lynn Land, Candice Miller, David Camp, Mike Cox

Retirement shock: 6 out of 10; it’s not particularly likely, but given the turmoil in the Michigan Democratic Party, not that many people would be searching for a reason why.

Race rating: Toss-up

Minnesota, Amy Klobuchar

Best-case scenario: Most interested parties pass on the race, preferring to take on Franken in 2014 when Obama won’t be on the ticket, and Klobuchar is given a free pass against a minor nominee.

Worst-case scenario: She attracts a major challenger.

Scenario rating: 7 out of 10; we’re now looking at the personal whims of a handful of candidates who could theoretically make this a race, but it just seems unlikely that the GOP will be playing significant offense here.

Possible candidates: Norm Coleman, Tom Emmer, Jim Ramstad, Erik Paulsen, Paul Koering, Steve Sviggum

Retirement shock: 10 out of 10

Race rating: DEM Favored

Missouri, Claire McCaskill

Best-case scenario: A Tea Party candidate emerges from the primary on a far-right platform.

Worst-case scenario: The GOP primary unifies behind a solid nominee.

Scenario rating: 2 out of 10; while anything can happen in a primary, the Missouri GOP seems to be able to keep things under control. This is going to be one of the top targets for the GOP, and they’ll do anything they can to keep it that way.

Possible candidates: Jim Talent, Sarah Steelman, Jo Ann Emerson, Sam Graves

Retirement shock: 9 out of 10

Race rating: Toss-up

Montana, Jon Tester

Best-case scenario: The big name candidates decide to run for the open Governor’s seat instead of challenging him.

Worst-case scenario: The best-case scenario doesn’t happen.

Scenario rating: 2 out of 10

Possible candidates: Denny Rehberg, Marc Racicot

Retirement shock: 9 out of 10

Race rating: Toss-up

Ben Nelson, Nebraska

Best-case scenario: The big names pass on the race and Nelson faces another underwhelming or unknown opponent.

Worst-case scenario: Dave Heineman enters the race and clears the primary field.

Best-case scenario (for Nelson, not Democrats): Nelson agrees to a deal where he switches parties in exchange for not being primaried.

Scenario rating: 2 out of 10; even if Heineman decides not to run, there are still others who would make this a competitive race.

Possible candidates: Dave Heineman, Jon Bruning, Jeff Fortenberry, Lee Terry, Adrian Smith

Retirement shock: 6 out of 10; Nelson seems to be committed to the idea of running again, which is good because Democrats really don’t have that many other options here.

Race rating: Lean GOP

New Jersey, Bob Menendez

Best-case scenario: The state unifies behind a weak nominee or several big names pass on the race, leaving perhaps an underfunded and too conservative candidate.

Worst-case scenario: A strong and pragmatic nominee emerges with the support of Chris Christie, who receives a spike in approval ratings.

Scenario rating: 5 out of 10; this is a purely candidate-driven race. In a state like Jersey, the Democrats have the advantage.

Possible candidates: John Crowley, Tom Kean Jr., Steve Lonegan

Retirement shock: 10 out of 10

Race rating: Lean DEM

Jeff Bingaman, New Mexico

Best-case scenario: No real challenger emerges and Bingaman coasts to another election.

Worst-case scenario: Bingaman retires and Democrats are left unprepared.

Scenario rating: 9 out of 10.

Possible candidates: Heather Wilson, Pete Domenici Jr., Allen Weh

Retirement shock: 7 out of 10; Even though he’s a five-term Senator, Bingaman is only 67 years old.

Race rating: Safe DEM

New York, Kristen Gillibrand

Best-case scenario: Like in 2010, no big name rises to the occasion to challenge her.

Worst-case scenario: Gillibrand simultaneously receives a primary challenge that empties her bank account and leaves her scatched coming out of the September primary, while a strong GOP challenger awaits.

Scenario rating: 9 out of 10; 2010 was the time for either of those two things to happen.

Possible candidates: George Pataki, Rick Lazio

Retirement shock: 10 out of 10

Race rating: Safe DEM

North Dakota, Kent Conrad

Best-case scenario: Conrad runs again and with John Hoeven already in the Senate, there’s no one left to challenge him.

Worst-case scenario: Conrad retires.

Scenario rating: 7 out of 10; I don’t know what Conrad is going to do, but right now there’s no obvious reason why this would be a race.

Possible candidates: Rick Berg, Jack Dalrymple, Wayne Stenehjam

Retirement shock: 5 out of 10; after Dorgan’s retirement, I don’t think anyone would be shocked to death if this happened.

Race rating: Safe DEM

Ohio, Sherrod Brown

Best-case scenario: The GOP nominates a lackluster candidate while Obama’s ground-game massively turns out of the vote.

Worst-case scenario: The GOP nominates a solid candidate and Obama’s campaign team focuses on other battleground states.

Scenario rating: 5 out of 10; I think Obama will focus here, so it depends on who the nominee will be.

Possible candidates: Jim Jordan, Jon Husted, Mike DeWine, Mary Taylor

Retirement shock: 10 out of 10

Race rating: Toss-up

Pennsylvania, Bob Casey

Best-case scenario: Casey attracts only a minor candidate, or an otherwise roughed-up or far-right candidate.

Worst-case scenario: Casey attracts a solid candidate who runs with a unified party behind him/her.

Scenario rating: 5 out of 10; a lot could go wrong for either side here.

Possible candidates: Jim Gerlach, Rick Santorum

Retirement shock: 10 out of 10

Race rating: Toss-up

Rhode Island, Sheldon Whitehouse

Best-case scenario: Sheldon Whitehouse runs for re-election.

Worst-case scenario: Whitehouse is discovered to be a serial killer.

Scenario rating: 10 out of 10

Possible candidates: Don Carcieri, John Robitaille, John Loughlin

Retirement shock: 10 out of 10

Race rating: Safe DEM

Vermont, Bernie Sanders

Best-case scenario: Sanders runs for re-election.

Worst-case scenario: Vermonters suddenly decide they don’t like socialists.

Scenario rating: 10 out of 10

Possible candidates: Les Britton, Paul Beaudry, Brian Dubie, Jim Douglas

Retirement shock: 9 out of 10; Sanders is 69, but he’s also a freshman.

Race rating: Safe DEM

Virginia, Jim Webb

Best-case scenario: The Virginia GOP nominates someone from the far-right, like Del. Bob Marshall

Worst-case scenario: Webb declines to run again, catching the party off-guard, while the GOP nominates a solid candidate.

Scenario rating: 5 out of 10; a lot could go wrong here for both parties.

Possible candidates: George Allen, Bob Marshall, Corey Stewart, Ken Cuccinelli, Bill Bolling, Tom Davis, Rob Wittman

Retirement shock: 5 out of 10; Webb has been low-profile and all reports indicate he doesn’t enjoy life in the Senate.

Race rating: Toss-up

Washington, Maria Cantwell

Best-case scenario: Deflated from losing in 2010, the Washington GOP gives this race a pass while focusing on the Governor’s race instead.

Worst-case scenario: Cantwell attracts a strong challenger and is caught flat-footed in campaigning.

Scenario rating: 7 out of 10; I can see a decent challenger emerging, but Cantwell will probably be fine.

Possible candidates: Clint Didier, Don Benton, Dave Reichart, Cathy McMorris-Rodgers

Retirement shock: 8 out of 10

Race rating: Safe DEM

West Virginia, Joe Manchin

Best-case scenario: Manchin threads the needle of being a Washington D.C. Democratic incumbent with maintaining his popularity back home.

Worst-case scenario: Manchin is viewed as “going Washington” at the same time he’s down-ballot from Obama.

Scenario rating: 5 out of 10; it could really go either way here, however the GOP still needs a strong challenger.

Possible candidates: Shelley Moore Capito, Betty Ireland, David McKinley

Retirement shock: 10 out of 10

Race rating: Lean DEM

Wisconsin, Herb Kohl

Best-case scenario: Kohl runs for re-election.

Worst-case scenario: Kohl retires.

Scenario rating: 5 out of 10; I’ve heard both, but I’ve also heard of people preparing for Kohl to retire, so it looks like no one would be caught off guard.

Possible candidates: Ron Kind, Tom Barrett, Paul Ryan

Retirement shock: 2 out of 10

Race rating: Lean DEM

Republicans:

Arizona, Jon Kyl

Best-case scenario: No serious candidate emerges against Kyl

Worst-case scenario: A serious candidate emerges against Kyl

Scenario rating: 3 out of 10; how serious remains to be seen, but this looks to be near the top of a short list for Democrats.

Possible candidates: Gabrielle Giffords, Terry Goddard, Felicia Rotellini

Retirement shock: 8 out of 10

Race rating: Lean GOP

Indiana, Richard Lugar

Best-case scenario: Lugar runs for re-election and isn’t seriously challenged

Worst-case scenario: Lugar retires or is primaried by a far-right hardliner

Scenario rating: 6 out of 10; Lugar has announced for another term, but he’s near the top of the Tea Party hitlist. We’ll see who they come up with and if they follow through.

Possible candidates: Marlin Stutzman, Brad Ellsworth

Retirement shock: 4 out of 10; Lugar will be 80 in 2012.

Race rating: GOP Favored

Maine, Olympia Snowe

Best-case scenario: Snowe isn’t primaried.

Worst-case scenario: Snowe is primaried.

Scenario rating: 5 out of 10; if Snowe is the nominee, she’ll coast to victory as she always has and always will. If the Tea Party is stupid enough to challenge her and wins, they’ll have given Democrats another seat for free.

Possible candidates: Chandler Woodcock, Elias Cutler, Tom Allen

Retirement shock: 8 out of 10

Race rating: GOP Favored

Massachusetts, Scott Brown

Best-case scenario: Martha Coakley runs again.

Worst-case scenario: Democrats put forward a strong nominee while Scott Brown has to fight to get out of the primary, if he’s even able to.

Scenario rating: 3 out of 10; I see Brown getting re-nominated pretty easily, but he’ll have a tough challenger regardless.

Possible candidates: Mike Capuano, Joe Kennedy, Ted Kennedy Jr., Stephen Lynch, Ed Markey, Marty Meehan, John Tierney, Elizabeth Warren

Retirement shock: 10 out of 10

Race rating: Lean DEM

Mississippi, Roger Wicker

Best-case scenario: Wicker runs again.

Worst-case scenario: Some major skeleton is found in Wicker’s closet.

Scenario rating: 9 out of 10; there’s some talk of a primary challenge to Wicker, but it’s not clear who or if it would even matter.

Possible candidates: Travis Childers, Gene Taylor

Retirement shock: 10 out of 10

Race rating: Safe GOP

Nevada, John Ensign

Best-case scenario: Ensign retires or resigns quietly, the GOP nominates a solid alternative, and Democrats fail to capitalize on the situation.

Worst-case scenario: Ensign somehow wins the nomination or a far-right candidate wins; meanwhile more embarrassing stuff about Ensign comes out throughout the cycle.

Scenario rating: 5 out of 10; this state has been a constant headache for the GOP.

Possible candidates: Brian Krolicki, Dean Heller, Dana Titus, Rory Reid, Shelley Berkley, Oscar Goodman

Retirement shock: 2 out of 10; Ensign says he’s running but that may not hold up.

Race rating: Toss-up

Tennessee, Bob Corker

Best-case scenario: Corker is unchallenged in the primary and Democrats, still reeling from 2010, decline to pursue.

Worst-case scenario: Corker gets a primary challenge and Democrats nominate a solid candidate.

Scenario rating: 8 out of 10

Possible candidates: Steve Cohen, Phil Bredesen

Retirement shock: 10 out of 10

Race rating: Safe GOP

Texas, Kay Bailey Hutchison

Best-case scenario: Hutchison quietly retires and a strong GOP emerges from the primary, while Democrats pass on the race.

Worst-case scenario: GOP infighting produces a roughed-up nominee while the Democrats coalesce behind an up-and-comer.

Scenario rating: 5 out of 10; it could go either way.

Possible candidates: Michael Williams, Roger Williams, Elizabeth Jones, David Dewhurst, Tom Leppert, Bill White, John Sharp, Chet Edwards

Retirement shock: 1 out of 10; it’ll be a shock if she runs again.

Race rating: Lean GOP

Utah, Orrin Hatch

Best-case scenario: Whether Hatch is primaried or not, the GOP puts up a candidate for the race.

Worst-case scenario: Republicans nominate someone extremely radical while Jim Matheson runs for the Democrats.

Scenario rating: 1 out of 10

Possible candidates: Jason Chaffetz

Retirement shock: 4 out of 10; whether a retirement is forced or not, Hatch’s departure wouldn’t shock anyone.

Race rating: Safe GOP

Wyoming, John Barrasso

Best-case scenario: Barrasso runs again, or doesn’t.

Worst-case scenario: In the greatest paperwork snafu of history, Republicans accidentally forget to file and are without a nominee.

Scenario rating: 1 out of 10

Possible candidates: Gary Trauner

Retirement shock: 10 out of 10

Race rating: Safe GOP

Summary:

Anything could happen. Everyone learned that lesson coming into the 2010 cycle. However, it strains the imagination to come up with a scenario where Mississippi, Tennessee, Utah, or Wyoming become winnable for Democrats, and states like Indiana and Maine would require a series of events to come onto the map. The only targets otherwise are Arizona, Massachusetts, Nevada, and Texas; as far as maps go, that’s pretty favorable for Republicans.

Meanwhile, it all depends on candidate recruitment, but assuming no real failures (like FL-2006 or DE-2010), Florida, Michigan, Missouri, Montana, Nebraska, Ohio, Pennsylvania, and Virginia are top targets immediately. Developments could put Minnesota, New Jersey, North Dakota, West Virginia, and Wisconsin on the map, while California, Hawaii, and Washington remain long-shots.

Putting even the fringiest of races on the table, Republicans have 6 targets to defend to the Democrats 12-15. Looking at likely targets, its 4 to 8. It’s possible, like 2010 for the GOP, those 8 races all break for the incumbents while the 4 races stay as targets and others come onto the map. However, even if the 2012 cycle was a neutral cycle, this would still favor Republicans.

Now we just have to sit back, wait for retirements to come in, see if any unexpected appointments or tragic passing put another race on the map, and watch as our best combined conventional wisdom falls apart in front of our eyes.

Here’s my initial 2012 cattle call. What’s yours?

1. Nebraska (D)

2. Massachusetts (R)

3. Montana (D)

4. Virginia (D)

5. Missouri (D)

6. Nevada (R)

7. Florida (D)

8. Michigan (D)

9. Ohio (D)

10. Pennsylvania (D)

11. Arizona (R)

12. West Virginia (D)

13. Texas (R)

14. Wisconsin (D)

15. New Jersey (D)

16. Indiana (R)

17. Connecticut (D/I)

18. North Dakota (D)

19. Minnesota (D)

20. Washington (D)

21. Maine(R)

22. Hawaii (D)

23. California (D)

24. Tennessee (R)

25. Delaware (D)

26. New Mexico (D)

27. New York (D)

28. Mississippi (R)

29. Maryland (D)

30. Utah (R)

31. Vermont (D/I)

32. Rhode Island (D)

33. Wyoming (R)

The list, so far:

Here’s a full list of the state legislative chambers flipped by Republicans on Tuesday: Alabama House and Senate, Colorado Senate, Indiana House, Iowa House, Maine House and Senate, Minnesota House and Senate, Michigan House, Montana House, New Hampshire House and Senate, North Carolina House and Senate, Ohio House, Pennsylvania House and Wisconsin House and Senate.

Additionally, the New York state Senate is looking bad.

One story, from August, sums up an entire election:

Leslie Pearce frowned as Congresswoman Suzanne Kosmas entered the near-vacant Bandshell on Saturday morning.

“Here comes Suzanne. I’m going to hide under the table,” said Pearce, president of the Democratic Women’s Club of Volusia County. Pearce had organized the 10 a.m. rally for local Democrats, but didn’t get the turnout she wanted.

The crowd maxed out at about 30 when Kosmas spoke just before 11.

Kosmas, who is fighting to keep the District 24 seat she won in 2008, joked as she took the stage about the “throngs of supporters” she was about to address before railing against Republicans. […]

“We know that this is going to be a difficult time for us. That’s why we need you there,” Kosmas said before turning the stage over to Dan Gelber, a candidate for Florida Attorney General.

As Gelber finished his speech, screams of “Kraut dog! Somebody order a kraut dog?” reverberated through the Bandshell. The shouts came from a hot dog stand adjacent to the Bandshell. No one came forward to claim the hot dog. […]

Donna Garcia, who runs DJ’s Hotdogs and Sausages with her husband, John, said she wished someone had told her about the rally, but not for political reasons.

“I wish I knew. I would have opened earlier,” Garcia said, before looking over at a hot dog smothered in sauerkraut. “I also wish I could find whoever ordered that kraut dog.”

In the end, we were all left holding the kraut dog.