Tuesday, September 18, AD 2012
Datech Guy explains why most presidential horserace polls are, to put it politely, worse than useless:
For the last two weeks we have been treated to the narrative that Barack Obama is surging at the polls, Mitt Romney is in trouble and unless there is a massive change in direction it is all over.
Simply put this is a lie.
Of all the polls you have seen, there is one poll that has gotten no attention, it is a poll that has been taken monthly, it is a poll that Doug Ross spotted and promoted on his site. It is the Rasmussen Poll of party identification.
They have party identification results online dating back to 2004. Here are the 2012 numbers through August
That shows a GOP advantage in registration this year but you might say: “Hey, Datechguy, you’ve been hitting polls all year, why can’t THIS poll be wrong?”
That’s a good question, we can answer it by asking another question: Does this poll of party identification correspond with the results of national elections?
Lets take a look:
Go here to read the brilliant rest. Most presidential horserace polls are assuming a Democrat advantage at the polls this fall of D-7, which is the advantage the Democrats had in 2008. That simply isn’t going to happen after the failed Obama Presidency. In 2010 the parties had roughly equal amounts turn out at the polls. I imagine that the election result this year will range from parity to a D-2 advantage. Assuming Romney wins the independents, as most polls show him doing, a D-2 electorate would lead to a Romney victory. Parity would be a Reagan 1980 landslide.