This blog is still alive, just in semi-hibernation.
When I want to write something longer than a tweet about something other than math or sci-fi, here is where I'll write it.

Sunday, January 22, 2012

Results for Prognostication #3:
The South Carolina primary

A bad week for my way of predicting the future. The loss of Huntsman as a candidate didn't hurt much, but losing Perry so close to the election date made a lot of polls fairly useless. Nate Silver's numbers were closer to the truth of Gingrich's big win over Romney, but both of us had Ron Paul in third place when Rick Santorum was actually five percentage points better.  Here are the final numbers.

Gingrich 40.4%(Silver beats me by 1.7%)
Romney 27.8%(Silver beats by by 2.2%)
Santorum 17%(I gain 0.1%)
Paul 13%(I lose 0.4%)
Other 1.8%(I gain 0.4%)
 
Final score: Silver 90.4%, Matty Boy 86.6%.  This is the first time either of us has climbed over the 90% mark.

The next test is Florida, a week from Tuesday.  Current polls have Romney ahead by double digits, but that means nothing.  Romney had a comfortable lead in South Carolina a week ago before the debates and the Bain Capital attacks.
 
Former British Prime Minister Harold Wilson gets credit for the line "A week is a very long time in politics", and these ten days will certainly be tumultuous.  The hatred the Republican establishment has for Newt Gingrich will once again rain down, but I can't tell if it will counterbalance Romney's negatives with the rank and file, who don't like his regal manner, religion or record in office. Bill Clinton's aphorism "Democratic voters fall in love, Republican voters fall in line" doesn't sound quite as clever as it did after New Hampshire. 
 


2 comments:

Anonymous said...

Don't feel too bad, Matt - the polling machines in SC are more than a little suspect!

Who knows? Maybe you hit the numbers right on the head?! Have you noticed a trend as we move from paper, hand-counted ballots (Iowa) to unverifiable, touch-screen balloting (SC and FLA)?

Just sayin' ...

Best regards!

Matty Boy said...

I don't expect the problem was the polling machines, but instead the speed at which the Republicans are changing their minds or perhaps O should say how much they are flailing about. The numbers for None Of The Above are still very high and except for Ron Paul voters, the commitment to candidates is not very deep.