Exceptionally Low Turnout Can Account For Polling Errors
At The American Prospect: Tweet In the home stretch, I wrote that midterm polling is far less accurate than in Presidential years. Today, in The...
Gubernatorial final snapshots, 2014
Tweet
Here are final polling snapshots for gubernatorial races that are either close or likely to switch party control:
Put your own predictions in comments! Some more notes…
The calculations above are a benchmark for how well we can do with polls alone. No house-effects corrections or fundamentals-based modeling was done.
Technical notes: For each state, at least three polls were used. The number of polls was determined using the variance-minimization method. All data were taken from HuffPollster. The median (no more than one poll per pollster) and the estimated SEM are shown in the second column. The win probability was calculated assuming an additional possible 2.0% error in the home stretch to account for polling error or bias. This was converted to a probability using the t-distribution (3 d.f.). To see how many polls were used, see this spreadsheet.
I’ll incorporate this into the 2014 Geek’s Guide, hopefully by midday Tuesday. In the meantime, please give your own predictions in comments. What surprises do you predict?
Topics:
Just reading your polls. Yes I think it is going to be a good night for Democrats as far as the Gubernatorial races go. However as far as the Senate goes I think you’re wrong about Alaska. I think in the end it will stay Democrat. Too many polls all over the place. When it is like that, at least from what I have gathered in my time, usually the incumbent wins. So we will see how these elections go on Tuesday. Enjoy.
I think Crist wins in Florida, Scott barely won last time and Crist should pull in moderate Republicans.
I think LePage is done for as well and it won’t be close. Cutler recently all but endorsed Michaud, I can’t see Mainers making the same mistake twice.
I think (hope) Walker will go down and base it on his not getting above 50% in any polling average, if people didn’t decide for him by now I can’t see him winning in the end.
I have no idea about Alaska but Parnell seems to be mired in scandal at the worst time and the opposition is a fusion ticket which might attract support for its novelty so I will throw Alaska into the Republican loss category. The rest I go along with the polling.
US Senate/Michigan: Dem Gary Peters; +10.5%.
Bummed about the Massachusetts results, there is something bizarre about the way my fellow Baystaters vote in local (ie, state-wide elections) vs. national (ie, Congress, Senate and Presidential).
There hasn’t been a Republican elected for national office from Massachusetts in a regular election (Scott Brown being a special election) since Edward Brooke (senate) or Paul Blute (1997)…
There are several things going on (there’s been lots of discussion of this in various threads over at Balloon Juice).
Massachusetts Democrats are a mixture of culturally liberal folks (Cambridge, Provincetown, the Berkshires), minority voters, and old-fashioned city machine Democrats.
Massachusetts Republicans running for statewide office, and Baker is a prime example (as was Mitt Romney before he started trying to be President), generally try hard to sound moderate, culturally un-scary and independent from the national party. The NPR liberals sometimes don’t like the machine Democrats very much, and will vote for these nice-sounding Republicans over them.
And then there are the particular problems with Martha Coakley. I don’t particularly like Martha Coakley myself, but I voted for her over Baker because I believe that the Republican Party is, at this point in time, too destructive a force to let hold any office whatsoever; I don’t care how moderate Baker sounds, he’s going to carry water for the national party sooner or later, and I can’t let that stand without some sort of opposition.
But to a lot of even quite liberal people, that sounds like robotic partisan hackery, and they see crossing the line for some offices as an enlightened and reasonable thing to do.
I predict the Dems win Colorado, Maine and Florida. Alaska I have no clue and I think Snyder barely hangs on in MI. Aside from that I have to go with the polls – sadly here in Massachusetts.
if people didn’t decide for him by now I can’t see him winning in the end.
I think this is a common mistake. Incumbents often win races, despite not exceeding 50% in the polling average. Undecideds will either decide on election day, or stay home. That’s why the better measure is the margin between the candidates.
You forgot Minnesota.
🙁
Confident it will be incumbent Mark Dayton (D).
PA, NH, RI, KS, IL, AK, CT, CO, FL, ME- D
AR, MA, GA, MI, WI- R
No runoff in Georgia.
My Predictions:
Governors Race
Kansas DEMS
Alaska DEMS
Michigan DEMS
Georgia REP
Pennsylvania DEMS
Wisconsin DEMS
Maine DEMS
Florida DEMS
Illinois DEMS
Connecticut DEMS
Rhode Island DEMS
Massachusetts REP
Colorado DEMS
Arkansas REP
Based on what?
Given that Maine is experiencing widespread power outages from the recent snow storm and polling locations are being relocated, my guess is that voter turnout will be low and LePage (R) will take it. I really hope I’m wrong.
You left out CA. We’re deep blue out here thank the gods.
You left out Oregon. I think Gov. Kithaber will be re-elected.
Early exit polls in Naples FL, by 2 PM EST about 227 for Scott (D) vs. 101 for Crist (R). Given that in 2012 Naples went 50:50 for Obama / Romney this early exit poll is kind of interesting.
http://www.naplesnews.com/news/elections/election-2014-local-exit-poll-results
@Mike…not sure where you are getting your information…Florida 19th which includes Naples and surrounding areas is 65% GOP..and I dont believe it could have voted for Obama. Trey Redell used to represent the district.
Collier voted for Obama? think again…you must be deluded….
http://uselectionatlas.org/RESULTS/statesub.php?year=2012&fips=12021&f=0&off=0&elect=0
In Florida, Governor Scott [R] has angered a lot of people. Not only is he delaying marriage equality, but there’s a marijuana initiative on the ballot, that seems to be likely to bring out the dems. Also, social media has been trying to get the word out about the importance of this election. This all points to ousting Scott even though there’s a decided lack of enthusiasm for Crist. OTOH, #IStandSunday stumped for the right. OTOH, his name is Crist.
While following this blog, I was pretty appalled at the biased data and especially the commentary here. I was hoping for a real discussion to gauge how the night was going to go, but found some of the predictions borderline insane and clearly driven by pure emotion. Next time, how about real analysis without the optimistic delusions?