Sabtu, 11 Februari 2012

Two contests, no prognostications.

 The results of the Maine caucuses are being reported tonight, but neither Nate Silver or I will be making any predictions because there were ZERO polls taken.  Since Maine is so close to Massachusetts, both Santorum and Gingrich put in zero effort, but Ron Paul has been wooing Maine voters like crazy.  There are people predicting an upset, but there is no real data, so these are just wild guesses. There are 24 delegates up for grabs, twice as many as New Hampshire this year because of penalties handed out to the Granite State.  But no one polled the good people of Maine, so we'll just have to wait until this evening to find out what happened.

The other numbers that will be released today are the results of the CPAC Straw Poll. There is no point in a polling company trying to poll such a small and meaningless gathering, but winning here at least gives a campaign a little boost in the 24 hour news cycle.  The organizers are miffed that Ron Paul won the straw poll in both 2011 and 2010, so they changed the way the contest is run to make it easier for all attendees to have their voice heard.  While the victories of Paul in the last two years are something of an embarrassment for the organization, Almost no one brings us that before Paul's winning streak the winner of the straw poll in 2007, 2008 and 2009 was... Mitt Romney.
This gives you some idea why the press covers CPAC. It's easy to get to (held in D.C.) and it's always good for a laugh. It had a "reputation" as the herald of Ronald Reagan back in the day, but like the rest of the conservative movement, the successes they can point to since then are few and far between, especially now that George W. Bush has been erased from history, even by his own team. The Bush whitewash is so effective, the ghost of Joseph Stalin let loose a low whistle and said with admiration, "Damn. You guys aren't playing around, are you?"


Rabu, 08 Februari 2012

Results from Prognostications #6 and #7: The non-binding caucuses in Colorado and Minnesota

There was only one company who tried to poll in Colorado and Minnesota prior to yesterday's caucuses, and given that lack of data Nate Silver declined to make a prediction.  While I consider myself a better mathematician than Silver is, I am not so foolish as to think there is nothing I can learn from him.  My prediction accuracy was stunningly bad, worse by far than anything I had done before.  There are several factors at play, but having so little data is certainly a major factor.  Without more fresh polling data, I will not make a prediction in future contests.
Let us first see the gory results. My updated predictions are better than my earlier stuff with less data, but none of them are even close to good. The actual numbers are first, my guess is second, the difference between them is in parentheses.

Colorado caucuses
Santorum 40.2% 28% (off by 12.2%)
Romney 34.9% 36% (off by 1.1%)
Gingrich 12.8% 22.5% (off by 9.7%)
Paul11.7% 13.5% (off 1.8%)
75.2% accurate, a C on the standard grading scale

===========
Minnesota caucuses
Santorum 45.0% 34% (off by 11%)
Romney 16.9% 23.5% (off by 6.6%)
Gingrich 10.8% 22%(off by 11.2%)
Paul 27.1% 20.5%(off by 6.6%)
64.6% accurate, a D on the standard grading scale

As Charles Barkley might say, I was turrrribull last night.  Let me list some of the reasons I sucked so bad.
  1.  Not enough data.  I brought this up already.
  2. Caucuses! We hates 'em!  There are so few voters at caucuses, they are hard to find by the methods pollsters use and all other things being equal, polls do a better job of estimating the votes in a primary, where the percentage of people casting votes is dramatically greater.
  3. The elusive Santorum caucus voter.  In the caucuses in Iowa, where there was a ree-donkulous amount of fresh data, no poll told us Santorum was a front runner. Even the polls that put him in second put him a distant second. When all the votes were counted, he beat Romney in a photo finish. In two caucuses last night, again Santorum over-achieved spectacularly over the polls' guesses.  Maybe people who vote for Santorum don't answer phones.  They must be at prayer meetings or something.
And so we have a caucus in Maine that ends this weekend, but so far nobody has any poll data. I won't make a prediction without three different companies giving their best guesses by this Friday.  if that doesn't come to pass, the next contests that count are February 28 in Arizona and Michigan.  I'm certain there will be boatloads of polls in those two states, so my next prognostication against Nate Silver will probably have to wait until then. I can boldly predict that I have no clue right now what will happen then.  Former British P.M. Harold Wilson said a week is a long time in politics, and he was not subjected to the 24 hour cable news cycle.  Three weeks might as well be an eternity now.







Selasa, 07 Februari 2012

Prognostications #6 and #7: The caucuses in Colorado and Minnesota - UPDATED WITH NEW DATA


February is slim pickings for primaries and caucuses that actually count.  Today, there are contests in Colorado and Minnesota, both caucus style, but no delegates go directly to the nomination.  Instead, a candidate who does well gets delegate to the state convention and the state convention makes decisions in the old-fashioned smoke-filled rooms. This differs from a place like Missouri, where the primary is strictly a beauty contest, not unlike the Iowa Straw Poll held in the fall, the one Michele Bachmann won.

My predictions are using three days of polling in both Colorado and Minnesota conducted by PPP. I give the more recent polling more weight than the older polling.  I remove the "None of the Above" votes from the sample and distribute the votes so that the sum of all the candidates is 100%.  In Nevada, I assumed the None Of The Above would be 0.5% and it was actually 0.3% of caucus goers opted not to vote. Since I don't know the in and outs of the caucus rules in these states, I'm going to assume it's difficult to vote for anyone but the four remaining guys.

Here are my numbers.  Nate Silver hasn't posted a prediction about either. Most of his tweets since the Super Bowl have been sports related.  No matter.  If he puts up predictions today, I'll update this post. If not, this was me keeping in practice.

Colorado caucuses
Romney 36% (was 41.5%)
Santorum 28% (was 27%)
Gingrich 22.5% (was 19%)
Paul 13.5% (was 12.5%)
(New data says Romney is being dragged closer to the pack, but still leads comfortably.)

===========
Minnesota caucuses
Santorum 34% (was 30%)
Romney 23.5% (was 28%)
Gingrich 22%(was 22.5%)
Paul 20.5%(was 19.5%)
(Santorum steals from Romney and Paul steals from Gingrich, more or less.)

If the polling data is even close, this should be a big news cycle win for Santorum.  The race for second place in Minnesota could produce yet another narrative.
When the final results are posted, I'll come back with a recap of how I did.



Senin, 06 Februari 2012

The results for Prognostication #5: The Nevada Caucuses

The Republicans finally finished tallying the results of the caucus they had on Saturday.  They tweeted the final results at 2:00 am Monday morning, about 27 hours after the last precinct closed up. (Most precincts had everything done by 7:00 pm Saturday, but some in Las Vegas finished at 11:00.)

27 hours to count about 33,000 votes.  This does not scream "competence" or "honest elections". Just to make matters worse, only about 1.3% of Nevadans were involved in this democratic process. By comparison, Sharron Angle, the ridiculous Tea Party candidate for Senate in 2010, got over 70,000 votes in the Republican primary.

As a person who has worked at a polling station and usually votes by absentee ballot, I can say without fear of contradiction that caucuses suck.

In any case, the final numbers are in and there are no big surprises. Mitt Romney won handily, Newt Gingrich was a distant second, Dr. Ron Paul (the only candidate who can save us from the Daleks) finished third and Rick Santorum was a distant fourth. The only drama was what the margins would be. Here are the final results.

Romney 50.0%(Both of us high, but I beat Silver by 0.8%)
 Gingrich 21.1%(Again, both of us high. Silver beats me by 0.4%)
Paul 18.7%(Both of us too low. Silver picks up 1.5%)
Santorum 9.9%(Both of us too low. I pick up 1.4%)
No vote 0.3%(Silver said 0%, I said 0.5%, I'm 0.1% closer)
Once again, my simpler method wins the prediction contest, 88.8% to 88.4%. This brings my record to 3-2.
There are two non-binding state caucuses held tomorrow in Colorado and Minnesota. Right now, the only fresh data is from PPP in both cases, so unless another company releases a poll between now and tomorrow morning, both Silver and I will be dealing with a single set of data in each case. If you follow this link, you'll see that Rick Santorum appears to be Romney's main challenger in both cases.  I'll publish my predictions and Silver's Tuesday morning and give the results on Wednesday, assuming people in Colorado and Minnesota are more competent at running elections than people in Nevada.

Sabtu, 04 Februari 2012

The difference between The New York Times and The Huffington Post

Ben Gazzara has died at the age of 81.  Long career, lots of movies and TV shows. His big early break is in Anatomy of a Murder with Jimmy Stewart and Lee Remick.  Baby boomers might remember his TV show Run For Your Life.

The teaser sentence from the New York Times obituary says he was Brick in the original Broadway cast of Cat on a Hot Tin Roof. This I did not know. They also mention his roles in the influential movies of John Cassavetes in the 1970s, which film buffs will know are a very important part of his career.

The Huffington Post has his picture on the front page with the headline "Big Lebowski actor dies".  Gazzara was in The Big Lebowski, but it was a tiny role. I swear, if the HuffPo had been around when Laurence Olivier died, the headline would have read "Polaroid spokesman passes away".

I shouldn't let stuff like this get under my skin, but it feels like we are handing the keys to our culture over to young people who can't find their well toned butts with both hands and a flashlight.

Jumat, 03 Februari 2012

Prognostication #5: The Nevada Caucuses

Tomorrow, the Republicans of Nevada will have their caucuses. Compared to earlier contests, the amount of polling is woefully small. There are some other primaries and caucuses on Saturday and Tuesday, but this is the only one will any fresh polling data at all, so this is the only one where my system will make a prediction so far.

Nate Silver is happy to use data I do not use in most cases, polls I consider too old to be useful.  He tries to predict into the future using this stale data, a method I do not use or endorse.  But this time, we are using the exact same data, one poll from the Las Vegas Journal-Review taken in the last few days of January and another from PPP taken yesterday and the day before.  In some ways, this is the cleanest test of our different methodologies.  May the best algorithm win.

Here are the two prediction set. As always, my numbers come first and Silver's are second in parentheses.

Romney 50.5%(51.3%)
Gingrich 26.0%(25.6%)
Paul 13.5%(15.0%)
Santorum 9.5%(8.1%)
Other 0.5% (0%) 
I made my prediction without seeing Silver's. I expect he barely knows I exist.  This time, Silver is bullish on Romney and Paul, while my system expects better results from Gingrich, Santorum and Other. I decided to give Other half a percent instead of none because even in the Iowa caucuses with several more choices, the number of people expressing a preference not in the main group was closer to half a percent than to zero.
Because Nevada is farther west than any other contest so far, the polls will close later and the results may not be complete until Sunday morning.  In any case, when 100% of precincts report, I will publish the results and pick over the bones.

For me, this is a lot more fun than reading the tabloids and trying to figure out if some female celebrity really is pregnant or not.  It's a hobby that better suits my talents.



An interesting number from a meaningless race.

Public Policy Polling is a Democratic company and they work cheap, relying on robo-polls.  For all that, their track record for accuracy is good.  Late last month, they published a poll about next Tuesday's Missouri primary, a "beauty contest" that produces no delegates. Newt Gingrich's name does not appear on the ballot, so he wasn't a choice in the poll PPP conducted. In that sampling, Santorum beat Romney 45% to 34%. Ron Paul was at 13%, about what he gets in most polls in a four way race.

Eventually, Santorum or Gingrich will drop out, and right now it's much more likely to be Santorum, given the beatings he took in both South Carolina and Florida. When it gets down to Romney, NotRomney and Dr. Paul (the only one who can save us from the Daleks), Romney will be in second place for a while.  In the long run, I still think Romney wins the nomination just because Republicans will finally be resigned to the fact that Santorum or Gingrich will be crushed by Obama.
Back to PPP for a second.  They published their results for the Nevada caucuses this morning, so unless there is another poll no one has mentioned yet, I have my numbers ready for the Nevada race, which is held tomorrow. Nate Silver hasn't factored PPP in yet, so I'll wait until he does and publish our predictions together. We both have the same data and there is so little to use, I doubt we will disagree by much. 
Maine has a caucus as well, but no polls have been produced so I have no way to predict anything and I'm going to take a pass.
Playing around with numbers!  So much fun for a nerd like Matty Boy.