Rick from SoCal e-mails:
I recently opened up an Intrade account to put money on Romney. With D oversampling in polls both nationally and state-by-state (I'm sure you are aware of this, but if you want I can grab the links), independents heavily favoring Romney, and what I see to be a personal known in that I know people who voted for Obama who are voting for Romney, and I know of people (authors, newspaper endorsements) that voted for or endorsed Obama and are now voting for Romney, the pulse I take is that Romney is going to win this outright and early enough for me to call it from my command center so that my parents will be able to go to sleep early next Tuesday. I tend to be very frugal with my money, but when you see the above scenario played out and a 1:1.5 return on your money, I see that as a smart play. If it were 50/50, I likely wouldn't risk it, but the reward is just too good. If this bears out, I will definitely have to thank Nate Silver for keeping this as lopsided as it is.
All that being said. I have another metric which has proven good to live by for 30 years or so. As a young teenage buck I delivered the Strib. I read the newspaper early one January morning to be the first to know that the Iranian hostages had been released. It was the beginning of my love of reading and being news current. I noticed even back then that the Minnesota Poll had the effect of being the most demoralizing poll to people of my slant. Election after election would happen and any chance of hope was dashed the weekend prior to it by a poll that had R's losing by wide margins. Then, on Wednesday morning the results would not be as extreme as the poll stated, but they'd still have called it right and the pollster would say that they were just off a bit but within the far part of the margin of error.
I believe the only year that the aforementioned scenario didn't play out was in 2004 when there was a huge demonstration by bloggers who took to the streets to call them out prior to the election. They were off by less that 1% that year, if memory serves correctly. With that one exception, to my current sense it seemed like they were typically off by 7-10%. I spent some time trying to find the Minnesota Poll historic information this weekend, but had no such luck. I'm wondering if you know where I can find it?
Seeing that they had it a 3 point race last week, warmed my heart. My vote has never been part of a winning slate of electors, and unless (and until) we hit rock-bottom here in CA, I've got a long time to wait. My hope is that for the first time since you and I were toddlers and for your first time as a 'legal' voter (I know you used to use a fake ID to vote a la 16 candles) may your vote COUNT!
Scott Johnson at Power Line has always been the go to guy for analyzing the problems with the Strib’s Minnesota Poll. Here are a few of his pieces on the poll:
Nov 8th, 2002 The Trouble With the Star-Trib Poll
Sept 28th, 2010 The trouble with the Star Trib poll, cont’d
Nov 4th, 2010 The Minnesota Poll strikes again
As Scott notes, the Strib MN Poll is no longer conducted by the paper itself as that work was has been handed over to an outside firm. While their performance in the 2010 election was as poor as previous Strib MN Polls, there is at least some hope that their latest poll showing Romney within three points of Obama in Minnesota was more accurate. A recent Rasmussen poll had the gap at five which is pretty close to what the Strib MN Poll showed and Rasmussen is usually a comparison poll when it comes to accuracy.
I still believe that the only way Romney wins Minnesota is if it’s a national landslide election, but to think that it could be this close this late makes it a little more interesting for those of us in a state usually taken for granted when it comes to presidential elections. Rick and other conservatives in the Golden State will have to continue to experience this through other less solidly blue states.