With the inordinate amount of data available, and countless numbers of statistical models for use in analyzing this data, it might seem logical to assume the ability to predict performance in a wide variety of disciplines including financial markets, law enforcement, and nature. However, because of the inherent inability to predict the behavior of living things such as humans and animals, it is more logical to assume there can never be an infallibly accurate predictor of performance in these disciplines.
In the discipline of financial markets, the age-old truth of “past performance is not an indicator of future success” is more accurate than many would prefer to admit. According to Stibel (2009), the most important reason all statistical models have failed to predict the future behavior of financial markets is because humans always have been – and always will be – thoroughly unpredictable. As Stibel (2009) notes, even predicting one move in a chess game is nearly impossible: “There are an overwhelming 10 to the 120th power possible moves. That’s a 1 followed by 120 zeros! As James Hogan explains it in his book Mind Matters, that sum far exceeds the number of atoms in the universe.” Obviously, the amount of data and the accuracy of the statistical model mean nothing, when considering human behavior. In fact, when it comes to predicting outcomes, the best tool for the job is the human brain. Although a human brain cannot calculate a complex mathematical formula to determine where a baseball will land (after being hit by a player’s bat), it can make an educated guess and the result will be more accurate than any formula could hope for (Stibel, 2009).
Law enforcement faces a similar problem, for nearly identical reason. Although there is an overwhelming amount of law enforcement data available, and very accurate tools to analyze it (i.e. data mining and predictive analysis tools) the inability to predict performance remains the same. As noted by McCue (2003), the greatest challenge in law enforcement’s attempt to use these tools is “most, if not all, data encountered was never intended to be analyzed”. Further, the data itself creates challenges with respect to the form of data, its content, and whether or not it is reliable and valid.
In nature, the results are no better. According to the Australian Academy of Science (2009), the inability to predict natural events such as floods, earthquakes, and epidemics can lead to suffering and death for millions of people. Despite a confluence of data and analysis from various scientific fields, there is still no accurate way to predict natural events or disasters, as nature itself is unpredictable.
Because of the inability to predict the behavior of living things (including nature itself), there are numerous areas where the use of data and statistics will be of little use in the attempt to predict future performance. Some of these areas include law enforcement, psychology, sociology, medicine, financial markets, weather, and sports. Because all of these areas include the highly unpredictable behavior of humans and nature, the data and analysis tools will be useful for analyzing what has already happened, but nearly useless in predicting what could happen.
- Australian Academy of Science (2009). Predicting natural events. Retrieved from http://science.org.au/nova/092/092key.html.
- McCue, C., Ph.D. (2003). Connecting the dots: Data mining and predictive analytics in law enforcement and intelligence analysis. The Police Chief, 70 (10). Retrieved from http://www.policechiefmagazine.org/magazine/index.cfm?fuseaction=display_arch&article_id=121&issue_id=102003.
- Stibel, J. (2009, January 22). Why we can’t predict financial markets. [Web log]. Harvard Business Review, HBR Blog Network. Retrieved from http://blogs.hbr.org/cs/2009/01/why_we_cant_predict_financial.html.
