Numbers Don’t Lie, but People Do
To anyone who followed US elections over the last few years, Nate Silver’s reputation precedes him. A renowned statistician and journalist, Silver has been lauded for his predictive acumen, which he has applied to fields ranging from baseball to political elections. He received a bachelor of arts in economics at the University of Chicago, graduating with honors, and spent a year studying at the London School of Economics. Following his undergraduate degree, Silver spent four years working at KPMG as an economic consultant while simultaneously developing PECOTA (Player Empirical Comparison and Optimization Test Algorithm), a method of statistical analysis that predicts the performance of baseball players. After quitting his consulting job, he made a living as an online poker player. Utilizing probability and statistical algorithms, he earned nearly $400,000 over the course of several years (Biography in Context). More recently Silver has dedicated his time to forecasting presidential, senatorial, and congressional elections with intense accuracy, correctly predicting the presidential winner in 49 and 50 states in 2008 and 2012 respectively.
In 2012, Silver published his first book, The Signal and the Noise, which evaluates the successes and failures of forecasting in recent decades. In spite of his notoriety for using empiricism in predicting baseball and political elections, Nate Silver’s writing professes a healthy degree of skepticism for empirical knowledge. Each chapter of his book is dedicated to forecasting in a certain field. One could argue his work discredits a number of esteemed public intellectuals’ works with his contention that, with few exceptions, empirical knowledge has tremendous limits. Silver argues that, instead of enhancing prediction, the recent surge in data availability has led to an increase in inaccurate predictions (The Signal and the Noise). Nate Silver makes a very compelling argument that many pundits, public intellectuals, and so-called experts often fall victim to the bias of their own beliefs. While such intellectuals often utilize data and hard analysis to support their point of view, Silver argues that such predictions or analyses are simply an extension of their belief (The Signal and the Noise). Regarded in statistics as confirmation bias, this tendency involves one extrapolating from arbitrary sets of data an outcome that is in line with their a priori beliefs (Pindyck and Rubinfield).
Nate Silver’s desire for greater skepticism and less reliance on biased experts has tremendous implications for modern day public intellectuals. While one would hope for a public intellectual to be one who changes his or her mind in accordance with new facts, Silver claims that numerous public intellectuals stubbornly promote a priori beliefs in the way they analyze data. Inaccurate predictions are in no way a new phenomenon. However, with the surge of data available in recent years there are more and more opportunities for the public intellectual to support his or her a priori beliefs with a sea of improperly analyzed data. This has led to more faulty predictions and a climate in which such predictions are granted greater legitimacy.
This trend of public intellectuals falsely extrapolating from data conclusions that support their a priori beliefs diminishes the value of their intellectual work. The unacknowledged bias that clouds many intellectuals’ work goes against their duty to criticize, probe, and adhere to facts. Stephen Mack, a professor of writing at the University of Southern California discusses this briefly in his article, The Decline of the Public Intellectual. Mack, argues that, “learning the processes of criticism and practicing them with some regularity are requisites for intellectual employment.” (The Decline of the Public Intellectual (?)). However, too few public intellectuals practice this criticism inwardly and towards their public intellectual colleagues. In Mack’s article he cites Jean Bethke Elshtain who reiterates Stephen’s point arguing, “a public intellectual is not a paid publicist, not a spinner, not in the pocket of a narrowly defined purpose… [they] should be party poopers.” (Elshtain). Nate Silver would very much agree with this point, particularly in the context of statistical analysis. He argues that too few intellectuals admit how little they know with certainty. Rather they overextend their knowledge, almost always imposing the bias of their beliefs. While it is hard to blame public intellectuals for such tendencies given the pressure they have to offer opinions and provide answers, it is inappropriate of them to promote dubious facts in their work.
What makes a Nobel-prize winning economist’s opinion about a matter any better than a layman’s if the economist’s opinion is clouded by his beliefs? Just because one’s misinformation is backed up by more data and greater intellect does not necessarily translate into legitimacy. In Silver’s survey of Nobel Prize winning macroeconomists he found a majority of them make predictions and analyses that prove incorrect the majority of the time (The Signal and the Noise).
This surge of inaccurate analyses that are accompanying intellectual works point to a lack of quality, which Mack would argue is something we must strive for. As those who continue to push quality through in their intellectual works become a growing minority one might question what the future of discourse is in the coming years. According to Mack, the measure of intellectualism is not so much a reflection of their influence or the scope of their audience as much as it is whether or not their work is of substance. He concludes his article on this note, positing that, “The measure of public intellectual work is not whether the people are listening, but whether they’re hearing things worth talking about.” We can rest assured that there will always be public intellectuals who strive for quality, however, as those in the public sphere who possess the most influence increasingly produce work that lacks quality in predictions it would appear that this minority of public intellectuals could find their excellent intellectual work drowned out by the noise of inaccurate claims.
This noise that Silver refers to is of great concern to those who value the truth. It is often said that there are three types of lies: lies, damned lies, and statistics. However this aphorism’s truth is lost upon a society that has become so dependent upon forecasting and analysis. We increasingly rely on empirical data and statistics to make generalizations and assumptions about our world with little concern for their validity. Macroeconomists and pundits use models backed by statistical analysis that supposedly will show us the ideal method for alleviating a depression, predicting the risk of future instability, and analyzing other causes and effects in society. It is unfortunate that such respected, intellectual figures fail to retain any skeptical tendencies in their works. In his work, Silver contends that even if such figures wanted to make more cautious assertions, they would quickly find their influence diminish as other reporters of misinformation replace them and supply the public with the answers and predictions that they crave (The Signal and the Noise). Such misinformation, Silver argues, is detrimental to society. Our dependence upon intellectuals to provide insights coupled with our acceptance of correlation indicating causation has become our Achilles’ heel. As a society, we increasingly put stock in the legitimacy of predictions and analyses that do not pan out and that often spur political decisions that are costly, ineffective, or even deleterious.
This problem is not something that can be solved in the foreseeable future. Silver remarks that the inaccuracies of most analysis will continue indefinitely due to the fact that most of our desired answers surround dynamic systems, or realms with infinite variables that cannot be accurately modeled (The Signal and the Noise). Silver is not proposing that we attempt to enhance our analytical skills to the point where we are more accurate; rather he is arguing that for the time being we should concede the limitations of our knowledge and must resolve not to spout misinformation. Until we maintain greater skepticism about the scope of our knowledge we will continue to be plagued by the failure of predictions and the misanalysis of data. However temping it may be for us to seek definitive answers and accurate forecasts it can prove as futile as attempting to measure how a butterfly flapping its wings in Brazil will affect weather patterns in New York.
“Nate Silver.” Gale Biography in Context. Detroit: Gale, 2009. Biography In Context. Web. 14 Sept. 2013.
Silver, Nate. The Signal and the Noise: Why So Many Predictions Fail–But Some Don’t. New York: Penguin Press, 2012. Print.
Mack, Stephen . “The New Democratic Review: The “Decline” of the Public Intellectual (?).” Stephen Mack. Web. 15 Sept. 2013. http://www.stephenmack.com/blog/archives/2013/08/the_decline_of_9.html#more
Pindyck, Robert S., and Daniel L. Rubinfeld. Microeconomics. 6th ed. Upper Saddle River, N.J.: Pearson Prentice Hall, 2005. Print.
Elshtain, Jean Bethke. “IAV | About Jean Bethke Elshtain.” AmericanValues.org. Web. 15 Sept. 2013.