Political Predictions a Tricky Game

In his book, The Signal and the Noise, Nate Silver, of FiveThirtyEight.com fame, notes that political pundits have a very low success rate for accurate political predictions. He proves that some, like Dick Morris, formerly of FOX News, are absolutely horrible predictors of politics. Random chance more often has better odds. Silver’s study of panelists on the McLaughlin Group resulted in 39 percent of their predications being completely correct and 37 percent being completely false.

Silver found 7 percent of the predictions to be mostly true, 10 percent to be mostly false and 8 percent to be partly true and partly false. Measured in statistical accuracy, 46 percent could be claimed as true, and 47 percent could be claimed as false—with an argument over the accuracy of the other 8 percent.

So, one could basically flip a coin on pundit predictions and have a slightly better chance of being correct than in choosing one of the “informed” opinions of the McLaughlin Group. But the “noise” does make for great entertainment.

Silver’s method of prediction is much more accurate. He gathers information and objectively puts the information into a statistical model that produces predictive percentages. Because nothing in life is a “sure thing,” some level of uncertainty is always present. An example is the story of the mafia guy who paid off all the other jockeys in a horse race so his long shot could win. As the race began, his horse darted out of the gate and broke its leg. So, while one could argue that a “fixed” race is a sure thing, even that has some level of uncertainty.

And sometimes, as anyone who has gone to a casino knows, percentages don’t always work out. But they will more times than not, which is the key to predictive models and profits at the casino.

So, for entertainment purposes, here are some predictions for 2013 that have more or less a coin flip’s chance based on Mr. Silver’s model. But, let’s throw in some percentages anyway, thereby giving the author an excuse if they do not come to pass.

Measure B will be tossed out by the courts — 95.7 percent. This has a high rate of probability based on precedent.

▪ The San Jose Sharks will win the Stanley Cup — 33 percent. A week ago this had no chance of happening. Now that the lockout is over we give it shot; we know they’re better than the Kings.

▪ At least one Santa Clara County supervisor will not serve a full four years — 80 percent. At least three are looking at a potential higher office, and one may be looking at jail time.

Crime will increase in the city of San Jose — 92.6 percent. Look at current trends.

▪ The A’s will announce relocation plans to San Jose, and MLB will sanction the move — 3 percent.  Even this prediction may be construed as wishful thinking.

▪ The NFL will announce the 49ers’ new stadium Santa Clara as the home of the 2016 Super Bowl — 73.9 percent. The windfall to the surrounding area would be enormous.

▪ The minimum wage in cities surrounding San Jose will increase — 83.3 percent. Wages will go up either by local fiat or through state action.

Gay marriage will be legal for everybody in California — 50 percent. Considering the Supreme Court make up, with some uncertainty given previous decisions by Justices Roberts and Kennedy, this one’s a toss-up. Should the court base its decision on precedent, logic and reason, this becomes 100 percent certain.

▪ The possibility that the Mayan Calendar was off by one year and that the end of the earth will be 2013 — .0000000000000000000000001 percent. But, hey, somebody out there is buying a bunker and 10,000 cans of creamed corn.

▪ Odds on winning the San Jose Mayor’s race in 2014 as of today: Sam Liccardo - 25.4 percent; Dave Cortese - 25.4 percent; Michael Mulcahy - 21.3 percent; Pierluigi Oliverio - 7.9 percent; Madison Nguyen - 4.3 percent; Pete Constant - 1.3 percent; other - 14.4 percent. These percentages will change significantly over time. If a popular woman and county law enforcement leader decides to run, she would instantly become the prohibitive favorite. But would Laurie Smith even consider the job?

Now, remember, the key to becoming an expert in predicting the future is being able to recall only those predictions which actually occur, and if 40 percent are correct a call from the McLaughlin Group could follow.

Rich Robinson is a political consultant in Silicon Valley.

Rich Robinson is an attorney and political consultant in Silicon Valley. Opinions are the author’s own and do not necessarily reflect those of San Jose Inside.

3 Comments

  1. I agree with the prediction that Measure B will almost certainly be thrown out in court and I will make my own prediction:  The chances that Chuck Reed will refuse to admit his huge and expensive mistake and instead blame the greedy police and firefighters for thwarting the “will of the voters”….100%

  2. “Silver’s study of panelists on the McLaughlin Group resulted in 39 percent of their predications being completely correct and 37 percent being completely false.

    Silver found 7 percent of the predictions to be mostly true, 10 percent to be mostly false and 8 percent to be partly true and partly false. Measured in statistical accuracy, 46 percent could be claimed as true, and 47 percent could be claimed as false—with an argument over the accuracy of the other 8 percent.

    So, one could basically flip a coin on pundit predictions and have a slightly better chance of being correct than in choosing one of the “informed” opinions of the McLaughlin Group. But the “noise” does make for great entertainment.”

    Yeah, well, I guess in a world where we’ll all mindless muppets who are expected to digest media content uncritically, that would be problematic.  But as someone who has seen probably over 4/5ths percent of all episodes of “The McLaughlin Group,” among those broadcast since he first began watching it (in the summer of 1984), I can assure you that if you filter those predictions through the mind of an informed viewer, you can easily dismiss the obviously-not-correct predictions, and bring the accuracy rate up to somewhat over 75 percent.  I mean, yeah, if you’re just some mouth-breather who likes to watch John flirt with Eleanor, the accuracy rate is pretty low.  If you’re capable of making a distinction between serious predictions, and goofy ones, then the accuracy rate skyrockets.

  3. This essay features a lot of disordered categories of discourse.  For example, Khamis is used as a sign post saying San Jose’s leadership “shifted further to the right.”  But his recent background is his several years on the county’s Human Relations Commission without ever speaking up about anything rightward to my knowledge.

    In addition, Khamis and others elected or re-elected must have noticed already that Mayor Reed has only the next three to six months during which he can punish defectors or reward supporters.  That’s about 24 hours in political time.  Why the underlying assumption that Reed’s DINO stands will carry the City Council to the right under any combination of City Council members?

    Even though Madison Nguyen was Vice Mayor, what possible good could Reed do for her in the face of campaigns by Liccardo and Cortese?  Her first job will be to draw clear lines between these campaigns by March or April, not do the bidding of lame duck Reed. 

    Then the essay moves into labor versus business interests.  But in San Jose it is labor that is the conservative force…labor isn’t about to dabble in charter schools or any other liberal experiment that touches on perquisites or benefits already won.  This is obvious when one asks, which party is willing to join in grand bargains in which some benefit is left aside?  Labor is the interest group in San Jose that cannot say yes…it is the conservative force in San Jose politics.