It’s not about hope. It’s about numbers.
Here’s what is going on.
1) Pollsters and the media focused very early on the big numbers appearing in the Dem primaries. It essentially frightened them into weighting almost everything.
2) As best I know there seems to be NO pollster just reporting the raw, random sample. Everyone is weighting by party ID this cycle. If someone knows of a pure random sample getting reported, speak up.
3) The supposed Democrat registration machine is, based on preliminary data, NOT producing what was intended. The early voting data from various places indicate clearly that there’s no enormous shift being seen in the votes cast as regards party ID.
4) All the pollsters DO HAVE the pure random samples available. They weight them after the fact. This is a bit bad. If the random samples differed hugely with the post weighting result, they would get scared and stop weighting.
5) The no response result of calls is at a historical high. This is bizarre. What is more bizarre is the large number of undecideds that remain, even given the no response total. Why do samples answer the phone and go through the survey in order to say undecided? Why not refuse to respond? Weird.
6) The early vote results being seen make it pretty clear at this point that the polls showing huge leads for Obama are way off. The bad news is . . . McCain seems to lead in none of the polls. He’ll need a few more undecideds to break his way to sway the IBD and other close polls and walk into Tuesday with a few showing a lead. And no sooner, actually. Unlikely voters remain unlikely if they think they have won.
"Show me just what Mohammed brought that was new, and there you will find things only evil and inhuman, such as his command to spread by the sword the faith he preached." - Manuel II Palelologus
5) The no response result of calls is at a historical high. This is bizarre. What is more bizarre is the large number of undecideds that remain, even given the no response total. Why do samples answer the phone and go through the survey in order to say undecided? Why not refuse to respond? Weird.
_________________
It’s the Politically Correct Chickens coming home to roost, the high numbers of undecideds are voting for McCain as are a portion of the democrats who don’t want to be bothered with any questions about why they’re not voting for the black guy.
The Politically Correct Gestapo has made accurate polling impossible this election.
That's kind of what I'd like to see. Averaging polls might be a better approach with the raw samples than the weighted ones.
Sounds like a tough polling job.
Randomly choosing names from a set is the easy part. Compiling a set of voters to randomly select from would probably be much harder. And then you would have to take a very large sampling...and make absolutely sure you get the opinion of everyone selected...including those nearly impossible to reach.