Battleground Gains For McCain
The Washington Post/WSJ/Quinnipiac U have a new survey out this morning showing that John McCain has tightened the margin against Barack Obama in four critical battleground states: CO, MN, MI and WI.
The Wash Post's Cillizza writes: "The first in the series of polls, conducted in the four states in mid-June, showed Obama comfortably ahead of McCain in Wisconsin and Minnesota while the races in Michigan and Colorado were closer although Obama still held the lead. The latest polling, showing a much tighter race, was conducted July 14 to 22, during Obama's high-profile trip to the Middle East."
CO
McCain 46%
Obama 44%
MN
Obama 46%
McCain 44%
MI
Obama 46%
McCain 42%
WI
Obama 50%
McCain 39%
Yeah, wonderful. Polls are fun:
NH-Pres
July 24 Rasmussen
Obama (D) 49%, McCain (R) 45%
FL-Pres
July 23 Rasmussen
Obama (D) 49%, McCain (R) 47%
And wow, so glad you listed Wisconsin. Look at the HUGE HUGE gains by McSame:
Wisconsin
Obama (D) 50% (-1)
McCain (R) 39% (+0)
Zero gain. And Obama lost 1 point. Wow. Such huge "battlefield gains".
Wrong sampling could be a major problem with this The Washington Post/WSJ/Quinnipiac U poll.
Take an example of Colorado: 40%Rep, 34% Dem, 26%Indep
LIKELY VOTERS.................................................
COLORADO Tot Rep Dem Ind Men Wom Wht His UnionHH
Obama 44% 7% 86% 47% 37% 50% 41% 57% 50%
McCain 46 87 7 39 55 39 51 29 40
----------------------------------------------------------------------
McCain and Obama each have almost identical leads among members of their own respective parties. And most importantly Obama leads 47-39 among Independents.
The only reason McCain is leading is because the poll had many more republicans than democrats. The poll has sampled 8% more R's than D's. Based on 2006 poll CO has about same percentage of D's and R's.
So the question goes to whether CO has really shifted toward Republican party ID while the rest of the country has shifted toward Democrats.
Also, in the U.S. Senate race, U.S. Rep. Mark Udall, the Democrat, and Republican Robert Schaffer are tied 44 - 44 percent in the same poll, compared to a 48 - 38 percent Udall lead June 26. No freaking way! All other polls show Udall to be way ahead of Schaffer.
Q poll is an outlier. Rasmussen also released a Minnesota poll on the same day with following results:
MN-Pres
July 23 Rasmussen
Obama (D) 52%
McCain (R) 39%
Now, the Q poll is:
July 23
Obama (D) 46%
McCain (R) 44%
Also, the difference in the senate poll:
Rasmussen: Coleman (R) 53%, Franken (D) 38%
Quinnipiac: Coleman (R) 46%, Franken (D) 49%
Sample, thanks for hitting that one out of the park. Sampling, sampling, sampling is the key. You're so right. These polls are flawed and definitely do not speak to the narrative of the race in those states.
Sampling of parties is rarely the major key to interpreting polls, though it is one factor. More often, it is wording which skews the results, as well as timing of outside and public events. Another factor, which usually is more predictive of later results, is whether the poll is of "registered," "likely" or "committed" voters. Don't know which of the reported polls took these into account. In the polls I have seen, the race has always been far closer among likely and committed voters than among those who are just "lookers."
The real issue is that no single poll means much. Trends over time are what counts, especially this early in the race.
Very true Pining. Look, we agree!
I think in this particular poll, however, CO and MN specifically are oversampled to Republicans, thus showing those races closer than they are.
The real important thing is, as you suggest, the trend line.
But it bothers me that in today's media spin room of "political analysis" all it takes is 2 oversampled polls and all of a sudden the narrative becomes "Battleground Gains For McCain".
This, sorry to say, speaks to the glaring inadequacies of the current crew at On Call. Glaring, maybe in the absence of a truly talented analyst, like say, Chuck Todd.