Happy Halloween, to those of you outside the Sandy disaster area who are celebrating. I've got nothing but time, seeing how a concert I was scheduled to perform last Friday was cancelled due to the stoppage of subway service, my neighborhood is without power until further notice, subways still aren't running, and I am out of work at least until Monday.
So all that said (and I'm very happy things weren't worse for me, shocked and saddened by the extent of the devastation in the area, compassionate toward everyone victimized by this storm from the Caribbean to the Northeast, and glad that I can stay with my girlfriend, who has power), here are what may be my final Senate ratings. All changes from my 10/23/2012 ratings are indicated in parentheses.
Short version: Polling has been generally good for the Democrats. My ratings and race changes, though cautiously, reflect these trends.
Ratings key:
Safe = No path to victory for the other candidate, absent some unforeseen disaster to the leading candidate.
Likely = Perhaps a 10-20% chance of victory for the trailing candidate.
Lean = Slight edge to one candidate, with the other having about a 25-35% chance of winning.
Tossup/Tilt = A very tenuous edge to one candidate
Arizona: Tossup/Tilt D - Whether to rate this Tilt D or Tilt R is guesswork, and the guess has most to do with how much the writer expects Hispanics to turn out. It will be interesting to see what the results are.
California: Safe D.
Connecticut: Likely D. While staying at my girlfriend's place, I've had the unusual experience of watching a lot of TV (I do not own a TV). Lots of ads from McMahon are being aired - or, to be precise, the same ad is being re-aired repeatedly. But I would consider a McMahon win a pretty big upset, in spite of some polls showing a close race.
Delaware: Safe D.
Florida: Likely D. Like Connecticut, I would consider a Republican win in this race a pretty big upset. Connie Mack IV is a bad candidate and Bill Nelson is a very good one.
Hawaii: Safe D.
Indiana: Lean D (was Tossup/Tilt R). When the best a Republican can do is give a tied internal poll result, he's not leading. Mourdock sunk his chances when he said that when a woman is raped and impregnated, that's what God wanted.
Maine: Likely I (caucusing with the Democrats).
Maryland: Safe D.
Massachusetts: Lean D. This race is really verging on Likely D now. I'm keeping it at Lean D because Scott Brown may still have about a 25% chance of winning, possibly. But not more than that.
Michigan: Likely D.
Minnesota: Safe D.
Mississippi: Safe R.
Missouri: Likely D (was Lean D). I think McCaskill has pretty much put this one away. Akin winning would be a bigger upset than a Republican win in CT or FL.
Montana: Tossup/Tilt D. This, like AZ-Sen, is a real tossup, and I really don't know what will happen. I think what this probably comes down to is whether the overall Republican lean of this often ticket-splitting seat is sufficient to defeat an incumbent who is authentically Montanan and replace him with a somewhat scandal-scarred incumbent Representative (I refer to the drunk boating incident) who's made sure to show signs of moderation recently. The difference between this race and ND-Sen is, if anything, the number of polls showing narrow leads for Tester and, perhaps, the fact that he's an incumbent. But I could easily see both races going for either the Republicans or the Democrats, or the races splitting the other way from my predictions. I think Tester's chance at victory is less than 55%.
Nebraska: Lean R (was Likely R). Kerrey's comeback in this election perhaps shouldn't have taken everyone (OK, almost everyone) by surprise, but it did, and I am certainly among the surprised ones. But a series of polls showing an increasingly close race cannot be ignored. This race has moved a substantial way toward tossup territory, and a win by Kerrey would be less of a shock now than a win by Mack in Florida. I think Kerrey has at least a 30% chance of victory in this election.
Nevada: Tossup/Tilt D. My reasoning on this election hasn't changed, and a recent Mellman poll has strengthened it: Public opinion polling consistently undercounts Democrats in Nevada, particularly Latinos. The number of polls that don't conduct interviews in Spanish should be considered a disgrace and a farce. I believe Mellman.
New Jersey: Safe D. The biggest problem New Jersey has right now is relief and reconstruction. It's quite unclear to me how loads of New Jerseyans without power, in largely destroyed communities, will be able to vote next Tuesday. But Senator Menendez has nothing to worry about in terms of reelection.
New Mexico: Likely D. This election is verging toward Safe D. I give Wilson no more than a 20% chance at an upset, if that.
New York: Safe D.
North Dakota: Tossup/Tilt R. My reasoning on this race hasn't changed. It pits a great, popular Democratic moderate against an unpopular Republican incumbent Representative. Heitkamp has perhaps a 48% chance at a victory, but I'm not confident enough to actually call the race for her, given the possibility that the undecideds will vote Republican. This could be the closest Senatorial election this year.
Ohio: Likely D. I'm tempted to call this Safe D. I would be truly shocked if Brown were defeated, and I give Mandel maybe a 10% chance of an upset.
Pennsylvania: Likely D. Similar to Ohio-Sen. I said in my last ratings diary that Casey has a 90% chance of victory, and I see no reason to revise that estimate.
Rhode Island: Safe D.
Tennessee: Safe R.
Texas: Safe R.
Utah: Safe R.
Vermont: Safe I (caucusing with the Democrats).
Virginia: Lean D (was Tossup/Tilt D). I give Kaine at least a 65% chance of victory. Mostly a gut feeling here, so feel free to disagree, but I think there will be a big vote from NoVa, and Allen isn't getting it.
Washington: Safe D.
West Virginia: Safe D.
Wisconsin: Lean D.
Wyoming: Safe R.
To summarize:
Safe D: CA, DE, HI, MD, MN, NJ, NY, RI, VT (I), WA, WV
Likely D: CT, FL, ME (I), MI, MO, NM, OH, PA
Lean D: MA, IN, WI, VA
Tossup/Tilt D: AZ, MT, NV
Tossup/Tilt R: ND
Lean R: NE
Likely R:
Safe R: MS, TN, TX, UT, WY
Race changes in favor of the Democrat (4): IN (Lean D from Tossup/Tilt R), (Likely D from Lean D), VA (Lean D from Tossup/Tilt D), MA (Lean D from Tossup/Tilt D), NE (Lean R from Likely R).
Race changes in favor of the Republican: None.
Bottom line: My prediction for the most likely outcome of the Senate races is a 56-44 majority in favor of the Democrats and the two Independents who will caucus with them. A realistic upper limit for Democratic representation (again, including the two Independents) is 58. If the Republicans swept the tossups, the Democrats would still have a 53-47 majority. In order for the Republicans to gain a 51-49 majority, they would have to sweep all the tossups and all the Lean D races - a very tall order, as all the Lean D races, with the possible exception of WI-Sen, are trending against them. Even earning a 50-50 split is a very unlikely result for the Republicans, and a 58-42 Democratic majority is more likely - an outcome few would have even dreamed at at the start of campaigning for this cycle.
Kudos to all who are contributing, canvassing, phone banking, and otherwise participating in these campaigns! We will soon know the results.
And a special shout out to those of you who believed in Kerrey when polls were showing him down by lopsided margins. He has a real chance to win, and I was quite wrong to laugh at you a few weeks ago, when his race looked hopeless.