Call (+1) 650-212-1212

Blog

Standard Deviation, Sporting Predictions and Investments

Sean Salleh 27 Aug 2013 Risk and uncertainty

Fancy a flutter? Actually, ‘flutters’, hunches’ and similar notions related to un-quantified intuition are avoided by many sports betting experts. They prefer a more analytical approach based on scrutiny of past sporting performance, comparative athletes,  team strengths, and some relevant math. While judgment also has a role to play, number-crunching is natural if you want to bet to win. After all, a bet is an investment and vice-versa, whether in sports, careers or finance. One of the basic measures frequently used to analyze performance, and incidentally to identify opportunities for improvement is standard deviation.

Input data for forecasting sports results Image source: gpb.org

Sports as an Investment Sandbox

One of the features of sports such as football is that you’ll know how a team did by the end of the season. Other entities, such as stock markets and share prices, have the added complexity of operating along a continuum, where today’s values can always increase or decrease tomorrow. With sports, when the season ends, you’re done (for a few months at least). This in-built cut-off and an inherent fun factor allows sports to be a sandbox or testing ground for trying out different modeling strategies and coming to grips with the real-world meaning of statistical constructs, including standard deviation.

Standard Deviation for a Simple Start

Feelings about using standard deviation to gauge performance and predict the future are mixed. Some people find that it rapidly runs out of steam for providing them with the information they really want. However, it has merit as a starting point to see what’s going on. The basic definition of standard deviation is that it shows the degree of variation or dispersion (this is stretching a more precise mathematical term here) of a collection of data. The function in Analytica for calculating standard deviation is ‘Sdeviation(x)’, where ‘x’ is the data file concerned (like ‘points_scored_per_game’).

Investment Noise or Performance Potential?

The smaller the standard deviation of sports (or investment) results, the more consistent the performance is, and therefore the easier to predict. In some cases, the factors influencing standard deviation of an individual performer will be more determined by that performer. In car racing, standard deviation of lap times for a given circuit will be determined by a driver’s individual attitude, strategy and skill. In football, the standard deviation of the performance of a player may also be determined by the type of football league and the position played (quarterback performance may vary more than others, for instance). On the other hand, a bigger standard deviation also suggests greater room for performance improvement. This brings in a further performance modeling factor, that of the trainer’s influence (or a CEO for investments in company stock).

Beyond Standard Deviation in Sports

A number of ‘investment’ experts (or skilled punters, if you prefer) use predictive analysis techniques that would leave many commercial enterprises astounded by their sophistication. From standard deviation as a starting point, higher league techniques include Monte Carlo simulation to map out individual and team performance forecasts.  Although we wouldn’t suggest that a management team spends all its time trying to maximize gains from betting on NFL, there may be something to be said for such an exercise as a way to include everybody in a structured modeling approach – especially those who have a built-in resistance to math or modeling, and who need a little extra encouragement.

If you’d like to know how Analytica, the modeling software from Lumina, can help you make better decisions about investments of any kind, then try a thirty day free evaluation of Analytica to see what it can do for you.

» Back

Sean Salleh

Sean Salleh is a data scientist with experience in guiding marketing strategy from building marketing mix models, forecasting models, scenario planning models, and algorithms. He is passionate about consumer technologies and resource management. He has master's degrees in Operations Research from University of California Irvine and Mathematics from Northeastern University.

Leave a Comment