A-Rod, PED’s, and SAP: How Big Data Analytics could prevent the next doping scandal


By Jeremy Stierwalt
Service Line Director, Analytics
Optimal Solutions an NTT Data Company

As I left SAP FKOM 2014, Bill McDermott’s use of the words “Simplify” and “Innovate” bounce around in my head, and my thoughts turn towards the world of professional sports once again.

In my last post, I shed some light on how Big Data Analytics are improving the sports industry in new and exciting ways: maximizing profits, enhancing and predicting athlete performance, and creating the best experience for fans.  But what about the darker side of the sports industry?  From Lance Armstrong to Marion Jones to Alex Rodriguez, doping allegations and admissions have become ubiquitous with the dark side of athletics.  In fact, 60 Minutes aired a piece on the A-Rod/Biogensis scandal last week which shed light on the inner workings of this darker industry. Ultimately, these headline making revelations have irrevocably stained the names of former highly esteemed sports stars our youth once viewed as role models and leaders.

It’s a sad fact that Performance Enhancing Drugs (PEDs) were a major catalyst for propelling these athletes into the limelight and led to their evolution to “superstar” and “role model” status.   These events have cast a pall of scandal and doubt over the entire sports industry.   If fact, doping has become such a rampant problem that it has become nearly impossible for regulatory agencies and associations to keep up with the science behind testing, much less prevent the use of banned substances.  (See National Geographic: http://news.nationalgeographic.com/news/2013/08/130808-sports-doping-testing-science)

In the case of Rodriguez, it is alleged that detailed tracking and painstaking tactics were utilized to game the system. Tactics included, taking the contraband at specific times to avoid it showing up in tests, and capturing urine samples ‘mid-stream’; all in order to avoid a positive PED test result from MLB’s governing body.  While PED testing at all levels of athletics has been, and will continue to be, a game of cat and mouse, Big Data Analytics can potentially play a role in predicting the likelihood of PED use.  Once realized, this could lead to quicker interventions and achieving the ultimate goal of prevention altogether.

But how do we even begin to accomplish this lofty goal?

A 2010 study published in the Journal of Sports and Exercise Psychology suggests that it may be possible to predict which athletes are more susceptible to engage in the use of PED’s by measuring certain factors.  The key determinates in identifying at-risk athletes included ‘attitudes’ towards doping, and ‘social norms’, the perception of society’s attitude towards doping and the perceived prevalence of it. ‘Situational temptation’, or the actual accessibility of PED’s was also considered, as well as any past doping behavior.

The study employed a hierarchical regression analysis to pinpoint which factors were significant in predicting doping behaviors. The results indicated that attitudes towards doping, perceived behavioral control and situational temptation could predict doping behavior with some statistical degree of accuracy.

In the same fashion, insurance actuaries utilize Big Data Analytics to evaluate the probability of events to mitigate undesirable financial outcomes; similar methodologies could be applied to address the propensity of PED use at all levels of athletics.  With the use of next generation Big Data Analytics solutions  (and associated predictive modeling tools: see SAP InfiniteInsight), capturing and correlating the key data points in order to predict the next doping scandal before it even happens are quite possible.

Let’s face it, thwarting the use of PED’s throughout all levels of athletics is not a simple task, and the existing testing methodologies aren’t working.  Remember, A-Rod and the others involved in the Biogenesis scandal never failed a test.  Although a turnkey, data driven solution is not yet available, the foundation for creating one has been established with existing Big Data Analytics solutions.  With a little more effort, perhaps we will never have to remove those A-Rod posters from a child’s bedroom, let alone explain why they are being taken down.

Leave a Reply