It’s a great day in sports analytics! Today, we will look at another method of analytics that is widely used in sports—and that is uplift and persuasion modeling. Uplift and persuasion modeling is an aspect of data science which uses data aggregation to create predictive models for performance analysis.
Sports analysts build predictive models, and test and refine them based on factors such as demographics, geography, leagues, teams, player skills, and so forth in an effort to perform statistical analysis, which help identify the characteristics of the data in other to draw certain inferences and decisions.
So, what is uplift and persuasion modeling? Uplift modeling deals with the increase in likelihood of the outcomes of an event with the treatment when compared to the outcomes of an event without the treatment. That is why uplift modeling is also called “treatment effects modeling”. There is no way you can tell the different between treated and untreated events in a direct way or using a direct technique, but rather by inferring answers from an experiment.
Now, let’s briefly discuss persuasion modeling. There is only a slight difference between uplift modeling and persuasion modeling. Suppose you have business products you want to sell, so you create and launch advertisements. Once you’ve advertised your product, some consumers have already decided to buy from you, while some others have already made up their minds not to buy. By focusing future advertising on those who have decided not to buy, you are likely to anger these consumers and not get a return on your investment.
Then there is the set of potential customers. We’ll call them potential, because they don’t know if they want to buy your product or not. This group of people need to be convinced. But the challenge is isolating this group of potential customers, while avoiding the first two groups who have made firm decisions. This is where persuasion modeling is effective.
How is this applicable to sport analytics? It’s mostly applicable to the business side of sports. It helps club owners and managers make persuasive decisions about the business that surrounds their sport.
Now, after the persuasion modeling has been used to identify targets. The next question is how do you turn this identified group into customers? This is achieved by uplift modeling. Uplift, simply put, means to raise something higher. That means you want to increase your chances of winning those potential customers to your side.
There are two main areas of uplift modeling. They are (1) Predictive models, and (2) Prescriptive models
While predictive models focus on the use of statistical tools and models to provide insights into future events with the aim of making predictions, prescriptive models employ the use of algorithms to optimize and simulate data and/or events with the goal of giving possible outcomes of an event.
In conclusion, coaches and analysts utilize uplift and persuasion modeling to analyze subjects, categorize them to determine their profitable aspects, and then treat those aspects with the aim of optimizing the performance of their team.
Find out how Sports Analytics Expert Victor Holman can give your team the competitive advantage.
How mature is your team’s analytics program? Take the Sports Analytics Maturity Assessment.
Learn about the Groundbreaking Sports Analytics Model coaches and sports analysts are talking about!
Learn all about Sports Analytics here.