How Machine Learning Is Changing The Way Sports Analytics Are Tracked

How Machine Learning Is Changing The Way Sports Analytics Are Tracked

​First off a quick explanation, Machine Learning is the process of using data and advanced methodologies to teach a computer how to learn and improve on its own. Think advanced AI or Terminator light. (The mental part only)
There are two ways sports are using Machine Learning and its advanced AI. Injury management and ball trajectory.
​
In basketball a computer with the right data set can almost always predict if a shot will be a make or miss based on trajectory and the mathematical formula used. This is very useful in helping with analyzing shot selection and offensive efficiency. Allowing for only offensive plays with a higher probability of success.
 
Predictive Modeling is the defined process of using a data set to prevent and or predict specific outcomes. Injury management is one of the key focuses of this model.
For example a study in The American Journal of Medicine and Science in Sports and Exercise looked at hamstring injury occurrences and the probability of reinjury.
The primary factors were:

• Seven genetic variants
• Previous hamstring injury
• Age (with players over 24 more likely to become injured)

The model was excellent as explaining why an injury occurred but only right about 50% of the time at predicting future injuries so improvement is needed.
A separate study placed a great deal of weight on genetic factors and injury risk. Certain genetic markers and inherited traits made injury risk more prevalent.

This field is growing and shows promise, it will be interesting to see where Machine Learning takes us in the future.