THE PLANK JULY 15, 2008
-
Read Later
READ LATERAvailable only to subscribers. SUBSCRIBE TODAY
-
Listen
ARTICLE AUDIO
- Font Size
After a slow couple of weeks, the pollsters have been very busy in the past 48 hours:
So, what's going on here? A pretty decent set of results for Obama -- but also not ones out of line with our expectations.
South Carolina and South Dakota are relatively competitive? We knew
that based on how North Carolina and North Dakota had been polling.
Iowa looks out of reach for McCain? It's looked that way all year, with
the exception of one or two polls taken during the flooding. Obama's
numbers look pretty good in the CBS/NYT and ABC/WaPo national polls? Those pollsters have tended to show relatively favorable results for the Democrats all year.So
what looks like a pretty interesting set of polling is really more of
the same. Obama is polling a bit of his peak numbers, but he retains a meaningful lead in the
popular vote and some structural advantages in terms of the electoral
math.
--Nate Silver
1 comments
I'm interested in what the above list does not provide. See, in CO Obama leads with +4 and 10% is up for grabs. In IA, it is better as Obama leads by +10 and only 8% is still on the table. But does anyone believe the McCain's lead in SC is only +6 and the 16 points up in the air will land evenly between Obama and McCain? Even in WA where Obama has a +10, look at the unfinished business: 47 +37 = 84 (correct?) so are the 16 points are up for grabs or not? What good are the SD numbers (McCain has +4) when the +43 + 37 tell me sixteen points aren't in that result.
Is there a reason I should not wonder what these incomplete totals mean? I don't have confidence in the national polls for several reasons and I'm not sure why they're relevant but we have seen double digit swings nationally and that is both less than either candidates margin in many states and causes me to wonder if there is more volatility or potential for the 'don't know yet' data in the individual states.
I realize the undecided my be a less solid number than the totals suggest, depending on the state. But when I see eight, ten or sixteen points missing in the totals I don't know if I can draw a conclusion where either candidate is now let alone think these numbers have much predictive value for an election that is over three months away.
Nate can state the numbers are "more of the same" but to me that reveals a lead still means little until we see the totals reflect a decided electorate.
Sorry, a four point lead in a state where there are 16 points 'missing' doesn't persuade me that we know enough to draw any conclusions.
- michael
July 16, 2008 at 9:59am