Vincent Marmara

The power of statistical analysis in Gaming

In statistical theory, we refer to ‘power’ as the strength, reliability and the level of confidence of a research study, but in business terms, power refers to the strength of your data to challenge your competitors – the strength of your data to take a competitive advantage. Realistically, both of them are directly related to each other.

Business Intelligence is nowadays commonly used in various industries, including gaming companies. As a result, a number of pertinent questions are often cited; “Why is there all this fuss about Business Intelligence and statistical analysis? Why is there so much focus on data? What is the main benefit of staying ahead of your data?”

Statistician Vincent Marmara

Statistician Vincent Marmara

I remember the days when I stared working in the gaming industry in 2006. This industry was relatively new, with few companies based in Malta. We compiled statistical reports to analyze several methodologies on customer acquisition. Acquisition campaigns were always on top of the agenda. As things evolved, it became harder to acquire new players. More efforts were required and, as a result, customer acquisition became more expensive. Sophisticated marketing campaigns were adopted to target the right niche customers. In most cases, these trends were the result of statistical analysis as well.

Fast forward nine years: things look different today. The focus has shifted on retention, not merely acquisition. Operators need to address two fronts: consolidate their customer database while simultaneously run campaigns to reactivate dormant customers. To do so, they must realize that with the data they already have, the need to make assumptions can be reduced once that data is properly analyzed. Furthermore, they need to realize that the faster that data is analyzed, swifter action can be taken to prevent losses.

I tend to compare this situation with an epidemic. Currently, I am reading a PhD in Mathematics (Statistics) in the area of Epidemiology and I describe such a case in the following way. In the field of epidemics, one of the major models is the ‘SIR’ model, where ‘S’ refers to the susceptible number of individuals acquiring an infection, ‘I’ is the number of infected individuals and ‘R’ is the number of recovered or dead individuals. The gaming industry works I a similar fashion. A population of active customers may be ‘susceptible’ to stop playing. Along their active gaming lifetime, customers might start to reduce their gaming activity due to several crucial reasons (‘Infected’) and hence, the gaming companies have two options in this scenario. The company either decides to ‘do nothing’ and hence, the customers move into the inactive group (they die out) or else they try to strengthen their gaming activity, thus saving (‘Recovering’) them from stopping to play.

This scenario is very similar to an epidemiological model. Statistical models, together with sophisticated techniques, which may include both a ‘Frequentist’ and ‘Bayesian’ approach, are the right solution in such cases. One needs to find the solution to cure the ‘Infected’ customers to recover them and save them from turning to the inactive group. The solution is to run sophisticated analysis to determine the main factors that are hindering players from strengthening their activity, or at least keeping their gaming activity consistent. Following the acquisition of such information, one will then develop the right statistical models to be able to predict the ‘infected’ group of customers.

This is why there is a need for live statistical analysis together with the need to analyze current datasets. This is the reason why statistics provides power and competitive advantage, based on good statistical and mathematical models, carrying a high power of accuracy.

To summarize it all, a significant number of gaming companies today do not realize that they have priceless information that is being overlooked. The answers to their biggest issues are likely hidden in their data already. Operators may not need to reinvent the wheel to remain competitive. They only need to look closer to home: to analyze and make best use of the data accessible to them.


Who is Vincent Marmarà?

Vincent Marmara is a Statistical Consultant and Researcher by profession. He obtained his first degree in Statistics, Operational Research and Mathematics at the University of Malta.  He later advanced his studies by obtaining Masters of Science in Statistics at Sheffield University, UK.  Currently he is reading for a PhD in Mathematics (Statistics) at the University of Stirling, Scotland, and also forms part of the research staff at the University of Stirling, UK.  He was entrusted with numerous research projects both at a national and international level.  He led research groups and analyzed data to a high level scientific extent.  One of his main research projects involved modeling missing data in surveys and in other research areas.  Mr. Marmarà has 9 years experience in the iGaming industry as a Business Intelligence Analyst. At present he is one of the partners and co-owner of the statistics research company, Sagalytics. Furthermore, he has been part-time lecturing Statistics and Mathematics at the University of Malta since October 2010. Moreover, he is the President of the Malta Statistics (MSTOR).

Vincent occupied several key important roles such as, member of the faculty board of science, Deputy CEO and Chief Regulatory Officer within the Malta Gaming Authority. During his term at the gaming authority, he was directly responsible for reducing the licensing process time by half.

Latest News