Today's Daily Kos Research 2000 tracking poll has Obama leading McCain 50-42. 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 EDT. LV=likely voter, RV=registered voter.
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: 50 (51) 42 (40) 3 LV
Reuters/Zogby: 50 (49) 45 (44) 2.9 LV
Rasmussen: 51 (52) 46 (44) 2 LV
Battleground: 49 (49) 46 (46) 3.1 LV Data in () from Friday
Gallup: 52 (51) 42 (42) 2 RV See also the LV I and LV II numbers
Diageo/Hotline: 50 (50) 42 (42) 3.4 LV
IBD/TIPP: 47 (47) 44 (43) 3.3 LV alternate link
ABC/WaPo: 52 (52) 45 (45) 3 LV
On successive individual days in the R2K poll (different than the topline, which is a combined three day sample), Obama was up +11 Friday, +9 Sat and +5 Sunday, with a +14 Thursday sample rolling off. Tomorrow's sample will repalce a +11, so further tightening is possible.
The R2K poll has tightened, but it remains to be seen if others will (no sign of it yesterday), or if it is sustained. The fundamentals of the race have not changed. Party support for McCain has ticked up 2 points, and indie support has moved a point, same as the last time we saw tightening.
Mark Blumenthal on the trackers:
The important point is that none of these changes is big enough to be considered statistically significant on any one poll. A consistent pattern of change might be suggestive of a real trend, but none is evident here either over the last 24 hours or the last week.
In the six days week since ABC News and the Washington Post started releasing results for their daily tracking poll, Obama's average lead on these seven has varied between a low of 6.7 (on 10/20) and a high of 8.1 (yesterday). Today's average is 7.4.
This is the trackers-only pollster.com graph from last evening, set from Oct through election day. At high sensitivity, it's easier to see the non-movement in the polls as of yesterday:
The above shows how the pollster.com average isn't varying all that much, even though there can be considerable variation between the polls. As Bernood Yost, the director of Franklin & Marshall College's Floyd Institute for Public Policy, notes in an email to Nate Silver:
Most pollsters have built their models through careful study by focusing on both intentions and past behaviors, which has worked well in the past. But in 2008 we are all concerned about how all of the new voters, who will likely be disproportionately young and minority voters favoring Senator Obama, will affect those models. No one knows for sure and this requires everyone to make untested assumptions about what is going to happen.
That means that the average of polls may or may not be predictive of the final number; if some of the pollsters are using models that are just wrong (and there's no way to know that now), the averages could be off.
Jay Cost walks us through the polling variations to illustrate that:
However, what if each pollster had a slightly different bag s/he was pulling from? In that situation, we should find more divergent results. That's basically what I'm suggesting here - that the bags the pollsters are pulling from are different. That's producing some of these larger-than-expected variations.
Now, I want to be clear: I am not making any claims about which pollster has the better sample of the electorate. I'm not singling anybody out for being right or wrong because frankly I do not know. I'm just pointing out that there seems to be disagreements among them that cannot be explained by random variation.
Importantly, there is one thing that the polls do not disagree on, the fact that Obama has a lead. All the polls show that. Also, we might begin to see convergence here soon. If pollsters have different methods for predicting what the electorate will look like, those methods might produce similar-looking "electorates" by the time we get to Election Day. At least for now, though, there is disagreement - not about who has the lead, but about how big that lead is.
Using Jay's graph (a graph can be made for the McCain side, as well), if the two polls to the left with low Obama numbers are wrong because of a faulty LV model, the Obama average would be higher if we exclude them. But the pollsters often change their numbers and positions on the graph from day to day, so the pollster off to the left today might be in the middle of the pack tomorrow. And while we can be suspicious of which group is the 'true' reflection of the public, and therefore try to pick which pollsters to rely on, or look for a pollster always off to the left (TIPP, for example, tends to consistently low-ball Obama but be in the middle of the pack with McCain), we really can't know right now which is the correct number to follow... and so most analysts include all the pollsters (I agree with Jay's "frankly I do not know", and also that 'Obama leads, however you slice it'). So, that brings us full circle back to trends, and those pollster.com trend lines that we find so useful.
Rasmussen adds some stability observations, with emphasis on how the concrete is setting:
Today is the 31st straight day that Obama’s voter support has stayed between 50% and 52%. During that period, the number voting for McCain has stayed in the 44% to 46% range every day and the gap between the candidates has ranged from four to eight percentage points.
However, while the overall levels of support have remained stable, voters have become more certain of their intent. Today, the percentage who say they could still change their mind is down to single digits. Forty-eight percent (48%) are now certain they will vote for Obama while 39% say the same about McCain. One month ago, while Obama enjoyed a five-point lead overall, just 41% of voters were certain they would vote for him. At that time, 39% said the same about McCain (see other recent demographic highlights and trends).
EJ Dionne on taxes:
The striking thing is that as the fall campaign has gone on – and despite McCain’s running attacks -- Obama’s advantage on the tax issue has actually increased. According to the Pew Research Center, 50 percent of registered voters questioned in mid-October thought Obama would do the best job in "dealing with taxes," compared with only 35 percent who said that of McCain. Back in September, Obama also led, but more modestly, 44 percent to 39 percent.
EJ makes the point that Obama's polling lead has grown, and whereas we can discuss why (in addition to framing it better, Obama has added that McCain health care tax credit gets taxed, which people do not like), the fact that the NBC/WSJ poll gave Obama a +14 on taxes as well means he is winning the argument. Even though taxes are a key part of McCain's closing argument
Mr. McCain will stick with the message he has embraced over the last week, presenting Mr. Obama as an advocate of big government and raising taxes.
the polls give us a clue as to how successful we can expect McCain to be with it.