Today's Daily Kos Research 2000 tracking poll has Obama leading McCain 51-46. All trackers are data from three days to five days prior to posting, with the R2K numbers from today (yesterday's numbers in parentheses) and the other trackers from yesterday (previous day's data). Data is updated as new information becomes available. Daily posting is approximately 7:30 am EST (I hope you set your clocks back.) LV=likely voter, RV=registered voter.
The last R2K poll of the election is today!
Nate Silver (fivethirtyeight.com) wrote up a tracking poll primer covering the eight available trackers. It includes sample size, time of publication and quirks, as well as Nate's opinion of the trackers. Recommended.
Obama McCain MoE +/- RV/LV
Today
Research 2000: 51 (51) 46 (45) 3 LV Final
Reuters/Zogby: 54 (51) 43 (44) 2.9 LV Final
IBD/TIPP: 52 (48) 44 (43) 3.3 LV alternate link
Rasmussen: 52 (52) 46 (46) 2 LV Final today
Battleground: 50 (50) 48 (44) 3.1 LV Final projected
Yesterday
USAToday/Gallup: 53 (52) 42 (45) 2 LV Final
Gallup: 53 (52) 42 (43) 2 LV Final all variations
NBC/WSJ: 51 (52) 43 (42) 3.1 LV Final
Diageo/Hotline: 50 (50) 45 (45) 3.4 LV Final
DCorps(D): 51 (52) 44 (43) 3 LV Final
Marist: 53 (50) 44 (43) 3.5 LV Final
CBS: 51 (54) 42 (41) 3 LV Final?
Fox: 50 (47) 43 (44) 3 LV Final
ABC/WaPo: 53 (54) 44 (43) 3 LV Final?
Ipsos: 53 (48) 46 (42) 3.6 LV Final
On successive individual days in the R2K poll (different than the topline, which is a combined three day sample), Obama was up +4 Sat, +4 Sun, and +7 Monday (52-45) with a +9 Fri sample rolling off (rounding can take place.)
Added:
- Simple average (Battleground excluded) of the above polls: Obama 52 McCain 44
- Look at the Obama number (a tight 50-54). The McCain numbers are 42-46. Mix and match and you get any poll you want.
- Further, all the polls picked up late tightening, then even later widening.
- If you want to criticize the R2K final number, you'll have to criticize Hotline and Rasmussen (similar numbers) as well. Wait until we see the numbers tonight before you do.
Be sure to check the internals and the trendlines for more details. That includes a Congressional generic preference of +9 for the D's and Sarah Palin's final fav/unfavorables (at - 11 she is disliked more than any of the other four – Obama finishes at +21.)
This is already 1980 redux (close race->debates->change->comfort level->not close race.) When people want change, Rev. Wright commercials seem small-minded and off-putting. But the real question is whether 2008 is another election like 1932.
For more prognostication from the bloggers, see this National Journal poll. 100% on the left and 59% on the right think Obama wins. We'll come back to that 41% tomorrow.
The final pollster.com high sensitivity graph is here, all polls except internet polls. The graph dates from "the fundamentals of the economy are strong":
BTW, you get the same result with just the 15 most recent polls listed above:
Are the polls right? Jon Cohen from the WaPo thinks so.
Despite my list of worries, a few things remain clear to me: Not all polls are created equal. We've been bombarded with polls that fell far short of the methodological rigor required for a good survey. If you mix in bad polls with the good ones, as happens all over the Web, you just may get dodgy results.
I also remind myself that humility is built into my field's DNA. The mathematics of the "random sample" on which all polling is based says that five times out of 100, we will be badly off the mark. Call it the pollster's law of averages.
But these seem to be topics for another day. The polls appear to be in general agreement that Obama is ahead; the only question is by how much. And this time, the pollsters' findings are being reinforced by the work of two other groups of campaign obsessives: the political scientists who use predictive models drawn from past election results to predict the next one (the one professor whose forecast had McCain headed for victory has "adjusted" his model), and the reporters out there knocking on doors and interviewing voters.
Nate Silver hopes so, too, in this very good interview on CNET:
[CNET]: How would you respond to criticisms that it's impossible to build a good poll-based predictive model for elections?
Silver: If the polls are wrong, then we're going to be wrong too. We can massage the data, but if there is some kind of a Bradley effect, for example, then our numbers are going to be wrong.
But we do emphasize that the actual errors in presidential polls are a lot larger than the reported (margin of error). In the primaries, the average miss was six or seven points. When the polls miss, they all miss in the same direction, so to that extent, averaging them can only do you so much good.
Nate also talks about the future of fivethirtyeight.com after the election, making this a great read.
Here's a screen cap from Countdown last evening, highlighting the poll ranges:
Tonight and tomorrow we get to check and see how we all did. Do we do as well as Pat Buchanan?
Thanks for your patience and your comments, folks.
Go Vote.