The Election, Hurricane Sandy, and Climate Change

There's a strong connection between this year's election, the deadly superstorm that ravaged our coast and the globally changing climate although it's not what the pundits would have you think.  The big takeaway of the last few weeks is not how a storm may or may not have influenced an election.  It's not that one storm is or can be an indicator of climate change.  It's not even about what either side of the political aisle says or does not say about climate change.  It's much bigger than that.  It's about the way we get our information.  It's what we choose to believe or not to believe. It's about methods that are tested and proven.  It's about science.

In the weeks prior to the election there was a bit of a 'to do' in the media over a gentleman named Nate Silver.  Mr. Silver has a now (in)famous blog, 538 Blog, in the New York Times, to which the paper purchased the rights in the wake of its surprisingly accurate predictions regarding the 2008 election.  It is not, however, simply the accuracy of those predictions that made the Times and others take heed.  There are folks all over the news making bold predictions, and if you pay attention to enough of them, one of them is bound to be right at least some of the time.  It is something of our nature to take notice when this happens. Nonetheless, what makes Mr. Silver somewhat exceptional is his thorough methodologies and his use of data--lots and lots of data.  It is also the conclusions that are reached through the use of that data and how they are presented that should be noted.

We live in what I would call a post-Information Age.  That is to say, we have entered a period in which we seek not just to have more information, but to have information about that information and its validity as well as to apply it in a way that makes accurate forecasts based on probabilities, statistics, and other fields of <gasp> mathematics.

After receiving criticism by Joe Scarborough of MSNBC, who was apparently annoyed at the level of confidence exuded in 538 Blog, Mr. Silver challenged him publicly to a bet of $1,000 which the loser would donate to charity.  To his credit, Mr. Scarborough did not bite.  The interesting thing is that while Mr. Silver displayed confidence (some interpreted it as arrogance), Mr. Scarborough projected certainty.  It was not certainty in the outcome of the election, rather it was in the belief that it was ultimately unpredictable--in his words: "a toss-up."  The wager was based on the odds that Mr. Silver's models had produced at the time, which I believe were around 78%.  Gambling is something that should be done only when: firstly, the gambler can afford to lose; and secondly, the odds are in his/her favor.  Both gentlemen's erudition in this regard is likely why one offered the wager and the other rejected it.  Of course, they both ended up donating to their charities of choice, demonstrating that it was really about the odds and not the money.  Mr. Silver was never certain that he was going to win.  He was, however, confident in the soundness of his methodologies and his odds, which he soon adjusted to 84% on the eve of election day.

As it turns out, 538 Blog (based on statistics produced as of Sunday evening) was correct in prognosticating every state won by the President but one (Florida--a virtual toss-up), and the outcome of every senatorial race except for two neck-and-neck races in Montana and North Dakota.  Being a savvy statistician, I believe even Mr. Silver himself would admit that there was some luck to that.  In fact, probability would have it that at least a few of those predictions would be wrong.  While the President ending up with 332 electoral votes was the result that came up most frequently in his models, it only did so 18% of the time. However, to say that he was just lucky, as many pundits worried about their job security are doing today, is widely missing the point.  He is using the best information available to him and using it to his advantage through the employment of modern computing and data collection.  Simply put, models perform consistently better than experts' predictions.

Jump back to the days just before Hurricane Sandy devastated the mid-Atlantic coast.  Meteorologists were confident in their models which showed the storm going up the coast and making a hard turn inland.  Years ago, having only historical norms to go on, this would have been deemed very unlikely.  Yet, we took these warnings seriously and governments were able to mitigate the resulting loss of life by ordering evacuations.  When it came down to it, we, by-and-large (and thankfully), chose to take these models seriously.  I shudder to think of what would have happened if we had failed to trust the science in this regard.

But that is exactly what has been happening when it comes to the matter of global climate change.  For years, scientists have formulated predictive models on the effect that a rise in greenhouse gases would have on our climate.  The models and methodologies varied as did some of the conclusions; however, they have in part led to an overwhelming consensus among scientists that global climate change is real, has anthropogenic influences and poses a serious threat to life on our planet.  It is true that there are some detractors, but they are few and far between, and they have the unfortunate advantage of a metaphorical megaphone bestowed upon them by powerful interests in the oil and gas industries as well as others who would benefit to pull the wool over our eyes.

The question going forward is this:  Are we going to let the guiding force that governs our positions and actions be one that is informed by science and reason or by 'gut feelings,' 'wishy' thinking and/or unverified claims? When the stakes are this high, I do not gamble, and neither should your government.