How to separate yourself in the Sports Analytics Field
Hi all! Today I wanted to recount some of what I’ve been doing lately as I try to distinguish myself in a field that’s growing quite fast-Sports analytics. It used to be easy to do Sports analytics. You just looked at the box scores and gave your ‘opinion’ which were very loosely backed by the data. Today, with interest in Machine Learning exploding, many people have begun to solve problems in Sports analytics using the ML framework. I myself have a few projects I’ve completed to this end which can be found here:
But with so many people doing this, it’s hard to distinguish yourself in the Sports Data Science community. To this end, I joined the Masters of Statistics program at Simon Fraser University in Burnaby. B.C, focusing on Sports Statistics. I’ve learned a lot already a month into my degree. To often we treat machine learning algorithms as a black box. We need to be more careful when selecting variables and we need to use the right tests to evaluate our models. After all, it’s easy for a well-trained model to succeed on predicting values in sample- it’s very hard for these models to be successful on out of sample data.
This is why I’ve taken a new approach for solving Sports analytics problems- one that is quite different from what’s out there today. Right now I’m working with CNG analytics- a company I helped co-found to bring Quantum algorithms to solve Sports problems. Our framework for Sports analytics is very novel-and has begun to gain the attention of well known teams and Sports analytics start ups. I encourage you to check us out, and reach out if you’re interested in learning more or working with us. Some of our articles:
Or, you can check us out on Linkedin:
I really look forward to hearing from you. Reach out to me @ eli_cavan@live.ca.
Reach out and let’s talk more!
