Daily Polling Report 10/28
by CA Pol Junkie
Today was about the 43rd consecutive bad day for John McCain. While the national trackers have bumped his way the last two days, the state polls are showing status quo or movement toward Obama. Obama got an eye-popping 10 point lead in the Suffolk poll of Nevada and more strong polling in Colorado, Ohio, and Florida. Obama has had leads in 4 of 5 Indiana polls taken in the last couple weeks, making Zogby's telephone poll favoring McCain look like the outlier. Arizona social scientists and statisticians had their version of The Big Game today as two universities released Arizona polls. One had McCain ahead by 8, the other by 2, and we'll find out who takes home the trophy next week. In what passes for good news for McCain these days, polls have him ahead by 1 in Georgia and by 4 in Montana.
A Mason-Dixon poll came out showing North Carolina to be a dead heat, which makes me want to say to the pollser, "Are you suuure?" I don't know the internals of the poll, but any pollster's likely voter model is under alot of stress this time. I calculate a turnout ratio, which reflects how much higher a percentage of one group votes than another. It's calculated like this:
T = (A/Ar)/(B/Br)
where A and B are the percentages of groups A and B actually voting and Ar and Br are the percentages of each group among registered voters. If each group actually voted at a rate proportional to their percentage in the registered voters pool, T would equal 1. In North Carolina, the turnout ratio for Democrats relative to Republicans in early voting is running at 1.44. For African-Americans relative to white voters, the turnout ratio is about 1.50. That means that relative to the registered voter population, 3 African-Americans are voting for every two white voters. Mathematically, it is probably not possible for this to continue through election day, but that ratio has been maintaining itself throughout the early voting period. The number of early voters in North Carolina is now up to 40% of the total 2004 vote, so there is something very significant going on which I doubt the pollsters are capturing.
HigherPie at DailyKos is compiling data daily on Nevada early voting and figures Obama is ahead by around 18 points right now in Las Vegas and Reno, with the vote total approximately 40% of the 2004 total. Obama needs to win those two areas combined by about 6 points to make up for the rest of the state. In Las Vegas, Democrats are turning out at a rate 27% higher than Republicans while in Reno the Democratic turnout rate is 53% higher. That's alot of ground for McCain to make up on election day with less motivated supporters.
7 days left before we make history!
The current prediction is Obama 378, McCain 160. Colorado is the state which puts Obama over 270 electoral votes while Indiana is the closest state. Montana switches back over to McCain for now thanks to a favorable poll for him.
Obama Base (264 EV): California, Oregon, Washington, Hawaii, New Mexico, Iowa, Minnesota, Wisconsin, Illinois, Michigan, DC, Maryland, Delaware, Pennsylvania, New Jersey, New York, Connecticut, Rhode Island, Massachusetts, Vermont, New Hampshire, Maine
Competitive states, cumulative electoral votes, and new polls:
Colorado (Obama +6.7) 273 EV
Insider Advantage 10/27 (10/20):
Obama 53 (50)
McCain 45 (46)
Virginia (Obama +6.6) 286 EV
Ohio (Obama +4.9) 306 EV
Nevada (Obama +4.6) 311 EV
Florida (Obama +3.0) 338 EV
Missouri (Obama +1.7) 349 EV
North Dakota (Obama +1.4) 352 EV
North Carolina (Obama +1.1) 367 EV
Indiana (Obama +0.4) 378 EV
Research 2000 10/23-25 (9/29-10/3):
Obama 48 (46)
McCain 47 (46)
Montana (McCain +2.1) 160 EV
Georgia (McCain +4.1) 157 EV
Insider Advantage 10/27 (10/23):
McCain 48 (47)
Obama 47 (48)
Arizona (McCain +5.8) 142 EV
West Virginia (McCain +6.7) 132 EV
McCain Base (127 EV): Alaska, Utah, Idaho, Wyoming, South Dakota, Nebraska, Kansas, Oklahoma, Texas, Louisiana, Arkansas, Mississippi, Alabama, Tennessee, Kentucky, South Carolina
The poll averages are created by a magic spreadsheet. Self-selected (Internet and mail) polls are ignored; no favoritism is done among the remaining pollsters. Polls are adjusted to today's conditions by shifting them by the amount of change in the average of Gallup and Rasmussen tracking polls. The weight of polls in the averages decreases geometrically each day such that 7 day old polls have 1/2 weight and 14 day old polls have 1/4 weight. The weight of state tracking polls is divided by the number of days in the sample. This method is very responsive to recent changes in both state and national polling.