I do agree with the article’s premise (though IMHO the drunk driving revelation hurt Bush far more than the cocaine rumors in the final days of the 2000 race).
A further reason to not trust the polls is the inaccuracy of the polls themselves. As we’ve seen in recent elections here and overseas, there’s a much larger margin of error than in the past. Looks to me like a real margin for error in good polling is about +/-8. IMHO, polls can show a very broad picture, but fail at the fine detail.
Also, even in the past, pollsters often got it wrong. For example, according to the records (such as newspaper articles) the Reagan-Carter battle in 1980 was, even on election eve, way too close to call and predicted to be a cliffhanger. Instead, it was a landslide.
For many in the media polls are a way to make news and influence people, not get an accurate read. So they want to bias the outcome.
The national polls are particularly useless. We know from the past that once the early states vote those numbers move around wildly. Giuliani and Hillary Clinton had big leads in national polls in 2007, but when both failed to win Iowa their numbers tumbled. Ditto Howard Dean in 2003-4.
I’m amazed that Trump has been so dominant in early state polls for so long, but I still don’t think he will win both Iowa and New Hampshire. Voters in those states often make up their minds just hours before they go to the polls, and I think many will ultimately balk at the idea of Trump in the White House. If someone else can beat him in one of those states that candidate will reap a windfall of positive press and donations. If multiple candidates beat Trump in Iowa and New Hampshire the story will be about how far he’s fallen, which throws a wrench into his self-positioning as a winner.
A few months ago I predicted that Trump would be out of the GOP nomination race by Super Tuesday. I’m less sure of that now; having been a frontrunner for so long he may well decide to stay in the race even after losing some early states. I do predict that he won’t win both Iowa and New Hampshire, the highest delegate count on Super Tuesday, or the nomination.
I do agree with the article’s premise (though IMHO the drunk driving revelation hurt Bush far more than the cocaine rumors in the final days of the 2000 race).
A further reason to not trust the polls is the inaccuracy of the polls themselves. As we’ve seen in recent elections here and overseas, there’s a much larger margin of error than in the past. Looks to me like a real margin for error in good polling is about +/-8. IMHO, polls can show a very broad picture, but fail at the fine detail.
Also, even in the past, pollsters often got it wrong. For example, according to the records (such as newspaper articles) the Reagan-Carter battle in 1980 was, even on election eve, way too close to call and predicted to be a cliffhanger. Instead, it was a landslide.
For many in the media polls are a way to make news and influence people, not get an accurate read. So they want to bias the outcome.
The national polls are particularly useless. We know from the past that once the early states vote those numbers move around wildly. Giuliani and Hillary Clinton had big leads in national polls in 2007, but when both failed to win Iowa their numbers tumbled. Ditto Howard Dean in 2003-4.
I’m amazed that Trump has been so dominant in early state polls for so long, but I still don’t think he will win both Iowa and New Hampshire. Voters in those states often make up their minds just hours before they go to the polls, and I think many will ultimately balk at the idea of Trump in the White House. If someone else can beat him in one of those states that candidate will reap a windfall of positive press and donations. If multiple candidates beat Trump in Iowa and New Hampshire the story will be about how far he’s fallen, which throws a wrench into his self-positioning as a winner.
A few months ago I predicted that Trump would be out of the GOP nomination race by Super Tuesday. I’m less sure of that now; having been a frontrunner for so long he may well decide to stay in the race even after losing some early states. I do predict that he won’t win both Iowa and New Hampshire, the highest delegate count on Super Tuesday, or the nomination.