Polling Question for EA

[_Kevin Graham]

As I mentioned before Scott Pierson believes Romney is going to win this thing despite what the polls say, and believes he will win by a electoral margin of 100 votes. His reasons for believing this essentially boil down to the assertions by Dick Morris, in whom he has complete and absolute confidence, and he posted his website remarks as evidence

Now I remember sitting in my Mom and Step-Dad's living room with them almost four years ago, and we all watched FOX closely for hours before Obama was announced the new President (at which point my step-father screamed "nigger" at the top of his lungs), and I remember quite vividly Dick Morris being the guy who was constantly speaking throughout the day assuring FOX listeners that despite what it looks like, McCain was actually going to win that election. Even as the final tallies were coming in from battleground states, he maintained his position that the polls were misleading and that McCain was going to pull through.

Morris then disappeared for a few weeks never being asked to explain his failed predictions. FOur years later I see he is pretty much saying the same kinds of things about polls, but he only shows up when the Democrat is leading.

You're the polling expert I know here so I wanted to get your feedback on some of the things he says in this article: http://www.dickmorris.com/why-the-polls ... mney-vote/

Here is the final statement, just to give readers some idea:

So, when the published poll shows Obama ahead by, say, 48-45, he’s really probably losing by 52-48!

Add these two factors together and the polls that are out there are all misleading. Any professional pollster (those consultants hired by candidates not by media outlets) would publish two findings for each poll — one using 2004 turnout modeling and the other using 2008 modeling. This would indicate just how dependent on an unusually high turnout of his base the Obama camp really is.

Thanks in advance.

[_EAllusion]

Yes, Dick Morris has an absolutely remarkable history of being wrong in his prognostications. And they tend to flatter conservative desires. He recently predicted Romney would have a great convention and get a significant, lasting bounce from it. That did not happen. He famously predicted Condi Rice would defeat Hillary Clinton in the 2008 race. The idea that the Republican would win and that that Republican would be Condi Rice both were highly dubious predictions at the time.

He was predicting McCain to the bitter end when that was utterly ridiculous.

He's basically the world's worst political pundit when it comes to this sort of thing:

http://www.guardian.co.uk/commentisfree ... cal-pundit

Here's a particularly rich one:

http://newsbusters.org/blogs/noel-shepp ... z22FMEV7DQ

Anyway, regarding this:

Poll weighting methodology varies from poll to poll. Contrary to his comments, for instance, a decent % don't hit cell phones. Those that do tend to paint a much rosier picture for Obama than the general average suggests. The idea that all the polls are relying on 2008 turnout results is just false. Most either have separate likely voter surveys that guide current numbers or rely on internal "house recipes" for projecting electorate make-up that simply do not involve assuming a 2008 repeat.

The idea that all undecided voters are going to break for Romney is silly. It's not grounded in any electoral history. When undecideds break for one candidate 2/1 that's usually a big deal, and Romney right now is looking at needing something closer to 4/1 to make a go of it.

When Morris says this:

But the fact is that the undecided vote always goes against the incumbent. In 1980 (the last time an incumbent Democrat was beaten), for example, the Gallup Poll of October 27th had Carter ahead by 45-39. Their survey on November 2nd showed Reagan catching up and leading by three points. In the actual voting, the Republican won by nine. The undecided vote broke sharply — and unanimously — for the challenger.

it's just a misrepresentation of the state of that election.

This link should clear up what really was going on with Carter vs. Reagan polling:

http://www.outsidethebeltway.com/debunk ... er-debate/

Morris is cherry picking a single poll result to suggest something was going on that was not, in fact, going on.

[_Kevin Graham]

Much appreciated EA