Unlikely Voter

Poll Analysis and Election Projection

Senate Projection for October 19

And it’s already time for the second installment of the new Tuesday version of my Senate projection. Republicans fell back a little after making new highs last week, but I can think of a few seats where Republicans might bounce back from that.

Then again, I can think of a race or two where the Democrats could regain more ground, so we shall see.

Situation

Senate right now: 59 D-41 R.

Seats up right now: 19 D-18 R.

Seats therefore not up: 40 D-23 R.

Minimum result needed for majority: -9 for Democrats, +10 for Republicans.

Method

I’m going to work off the polls when they’re available, working from Real Clear Politics, giving double weight now to Likely Voter polls, and make guesses otherwise. That’s usually easy because the seats that aren’t polled are typically safe. Changes from last time will be in boldface.

Seats

AL: Richard Shelby: Lead has grown to +32 in Rasmussen. 99.9% R victory.

AK: Lisa Murkowski (primaried): I still think polling prompting for Murkowski by name is vastly overestimating her support as a write-in, but I go to war with the polls I have. 60% R victory from 75%.

AZ: John McCain: This is getting lopsided. 99.9% R victory from 99%.

AR: Blanche Lincoln: John Boozman up 20+ in polling. 99.9% R victory.

CA: Barbara Boxer: The momentum has shifted here.. 45% R victory from 25%.

CO: Michael Bennet: Other polls follow PPP down, but not so far. 55% R victory from 75%.

CT: Chris Dodd (retiring): Polling still disagreeing on single or double digit gap. 10% R victory.

DE: Joe Biden (resigned): One less bad post-debate poll for O’Donnell not enough to move this. 1% R victory.

FL: Mel Martinez (resigned): Rubio racks up double digit leads. 95% R victory.

GA: Johnny Isakson: Isakson rebounds in polling from one scare. 99% R victory.

HI: Daniel Inouye: Two new polls, but with mixed results. 0.1% R victory.

ID: Mike Crapo: Crapo lead hovering around 40. 99.9% R victory.

IL: Barack Obama (resigned): Giannounlias and Kirk still battling hard. 50% R victory.

IN: Evan Bayh (retiring): Dan Coats lead growing. 99% R victory.

IA: Chuck Grassley: Grassley rebounds from one bad poll. 99% R victory.

KS: Sam Brownback (retiring): Republican Jerry Moran showing leads in the 40s. 99.9% R victory.

KY: Jim Bunning (retiring): The rebound for Paul fades, and the race tightens. 65% R victory from 80%.

LA: David Vitter: Vitter gets a bounce in Rasmussen. 95% R victory.

MD: Barbara Mikulski: Mikulski’s primary opponent’s internal even has her ahead. 0.1% R victory.

MO: Kit Bond (retiring): Blunt marching on from one bad poll I didn’t trust anyway. 85% R victory.

NV: Harry Reid: Angle keeps the momentum. 55% R victory.

NC: Richard Burr: PPP even has this at Burr +13 now. 99% R victory.

ND: Byron Dorgan (retiring): John Hoeven leads by so much he breaks the graphics in my analysis tool. 99.9% R victory.

NH: Judd Gregg (retiring): Hodes has never led this race. 85% R victory.

NY: Chuck Schumer: Schumer hovering around 60. 0.1% R victory.

NY: Kirsten Gillibrand: The party’s over for the NY GOP. 1% R victory.

OH: George Voinovich (retiring): Portman leads cross over 20. 99% R victory from 95%.

OK: Tom Coburn: Coburn lead approaching 40. 99.9% R victory.

OR: Ron Wyden: Recent polling very good to Wyden. My thoughts of this being a sleeper GOP victory seem to have been premature. 1% R victory.

PA: Arlen Specter (primaried): Sestak stuck in a band around 40… except in the new PPP. 75% R victory from 80%.

SC: Jim DeMint: A poll did come out. It’s not good for Alvin Greene and his felony indictment. 99.9% R victory.

SD: John Thune: Unopposed. 100% R victory.

UT: Bob Bennett (primaried): Deseret News confirms Rasmussen polling. 99.9% R victory.

VT: Patrick Leahy: Leahy lead down to 30. 0.1% R victory.

WA: Patty Murray: Murray now takes tiny lead. 45% R victory (from 60%).

WV: Robert Byrd (died): Manchin takes a lead solely on the Orion poll. 45% R victory from 60%.

WI: Russ Feingold: Ron Johnson has a bad poll.. in which he still leads. 80% R victory.

The simulation

100,000 trials using the probabilities above.

Results

The Republicans lose more than I thought. It’s still for the Republicans a 7 seat gain, but by a hair. R+7 seats got 22,634 trials, while R+6 seats took 22,082. The two next likely outcomes are 5 and 8, with that four seat range accounting for 77% (76,611) of the curve. The GOP also loses another 2% of their majority chances, going down to 3% (3,075). Exactly 2 trials out of the 100,000 gave Democrats a gain of one seat, up 100% from last week.

While the GOP does fall back this week, I think some of it is illusory, and caused by outliers having disproportionate impact. I will have to watch for the correctness and adjust my methodology for the next election, possibly adding some sort of hysteresis to the final per-seat ratings.

Possibly Related Posts

Comments

7 Responses to “Senate Projection for October 19”

  1. Yeah I think your numbers on Pennsylvania and West Virginia are low

  2. Hey Neil,

    You don’t, by chance, have a copy/version of the calculator that we could play around with, do you? I’m interested in trying out some outcomes, based on some different assumptions.

    (I could write my own, but man; I’d hate to reinvent that wheel!)

    In any case, thanks as always for the analysis. Unlikely Voter has definitely become one of my favorite resources this election cycle. (Gives me flashbacks to the work Jay Cost used to do.)

    • This calculator is more a process and I don’t have anything automated but the single poll tool which I’m not ready to release now.

      • I ended up just building my own Monte Carlo simulation using a humble spreadsheet. (Is it sad that I actually enjoyed doing it?) ;)

        I’d link to my related post, since I reference this one, but I don’t want to link-spam you. Bottom line: I project Republicans +9, with a 40% chance of them gaining 10 or more.

        • Oh feel free to link.

          And of course it’s fine to enjoy it. I’d enjoy mine a lot more if they weren’t SO MUCH WORK. Which I’m going to fix for next time around, definitely.

    • And thank you for the nice feedback!

Write a Comment

Comments are closed.