McCain By Four Among Likely Voters?
I don’t believe it for a minute…but here you go. The more broad-based RealClearPolitics average has Obama by 3.2. I’m more inclined to go with the InTrade crowd, who have Obama leading by nearly 2-1…
|
||||||
McCain By Four Among Likely Voters?I don’t believe it for a minute…but here you go. The more broad-based RealClearPolitics average has Obama by 3.2. I’m more inclined to go with the InTrade crowd, who have Obama leading by nearly 2-1… 9 comments to McCain By Four Among Likely Voters?Leave a Reply |
||||||
|
Copyright © 2012 Informed Speculation - All Rights Reserved |
||||||
Have you ever read Fivethirtyeight, Mark? Just put a dot-com at the end of it. It’s a really fascinating site run by a sports statistician applying some really interesting and in-depth statistical models to the Presidential and Senate races. The proprietor of the site is pretty skeptical of these numbers, too.
I’m of the mind that polls this year are even less reliable than usual.
All (good) polling is based upon statistics. We know from exit polls that x% of the electorate call themselves Democrats, y% Independent and z% Republicans. These number vary from state-to-state, and vary (slightly) from election to election, but nationwide they’re relatively stable. When polling is done, party affiliation is one of the first questions asked. If the results of the poll don’t follow these “known” %ages, then the pollsters weight them appropriately. Most do. Some don’t which is why you sometimes get these odd polls that talk show hosts them scream about having poor methodology.
However, this year, the Republicans have a depressed base. The Democrats have an energized base and have registered tons of new voters. Does that mean that the “known” %ages are off and that x is going to be higher and z lower this time around? If so, the weightings are all wrong, and Obama may have a bigger lead than any of the polls have suggested.
Also, the term “likely voters” comes from asking the voter if they voted in the last few elections. With these new voters, the answer will be no. But some of them are likely to vote. Again, advantage Obama.
On the other hand, when you look at the details of the polls, there is encouraging news for McCain. I’ve seen some where he leads over Obama in every single age group except 18-35. The significance of that is that the 18-35 year olds are the least reliable voters. And most of these new registered Democrats fall into this category. Will they actually come out and vote or will they decide to go out and watch Harry Potter instead (I know..HP doesn’t come out until 2 weeks after the election)?
Add to that, Obama’s history in the primaries of being a poor closer, and that only adds to the confusion. He has very bad numbers on all the character issues (hugely bad…on some of them, McCain holds 40% point leads). Will people decide that his “hope for change” message is more important than their doubts about his character?
The pollsters don’t know the answers to these questions. So, they don’t know how to weight the numbers this year.
Take every poll you see with an even bigger grain of salt than usual.
Wow, Chris and I agree almost completely. Huh.
Hey, finally a post we can all agree on! That means I screwed up somewhere, though…thanks for the tip, Fargus, I’ll check it out…
Fargus is correct. I’ve been following 538 for a while now. I have some minor disagreements with his methodologies and occasionally posters on his site become cheerleaders for Obama so I did stop reading it for a while. I started back up again late last week though.
BTW, I believe he had a post last week saying that the odds of Obama winning dropped below 60% for the first time since “securing the nomination”.
Oh, and thanks, Ryan. I was thinking of adding more details to that and adding it as a post on my own blog. After hearing that you and I are in agreement, I will almost certainly do that.
The polls are near meaningless at this point, so long as one candidate isn’t behind by double digits. The debates will really decide this thing.
I wouldn’t say meaningless. Historically, the Republican candidate picks up about 6-7 points nationally between July and Election Day. For a good part of July, McCain was down by about 8-9 according to RCP averages. McCain’s camp had to be very nervous about that. And if Obama’s lead is understated due to using the wrong percentages, it’s worse.
RCP averages now have McCain down about 2.5 nationally. If that is accurate, it’s very very good news for McCain.
Though, I don’t necessarily agree with your name
, I compeltely agree with your post. Polls give the candidates and public a rough glimpse at what MAY happen, but the debates are key for people developing opinions. Especially those that don’t follow politics too closely. I personally feel a lot of those people will watch the debates to not only get a rouch idea on different candidate’s platforms, but also to see how the candidates handle themselves when they’re on the spot. Either way, I really feel it could go either way right now. It will be very interesting to see the outcome as well as how many people actually vote.