Talking Tech Featured Talking Tech: Applying LRMC Rankings to College Football, Part One While SRS can go a long way towards calculating a difficult problem like schedule strength, it has its pitfalls. Logistic-Regression/Markov Chain (LRMC) ratings provide an alternate approach.
Talking Tech Talking Tech: Applying LRMC Rankings to College Football, Part Two The previous post in this series covered how LRMC works and its mathematical implementation. Now it's time to take things a step further and apply this concept to ranking college football teams.
Talking Tech Featured Talking Tech: Calculating SRS (Pandemic Edition) You might wonder, why is it so difficult to calculate some of these metrics? We have games and data, right? While that may be true, there are several factors that serve to complicate things this season.
Analysis Featured Analyzing Variable Importance in CFB Machine Learning Models As always with machine learning models, a common question that comes up is “Why?”. Why did the model choose this team to win vs. the other? What variables are the most and least influential in a given predictive model? Time to take a deep dive.
Analysis Featured CFBD: Using machine learning to predict game outcomes and spreads Predictive models for college football are a great application of machine learning techniques. Today, we'll look at one technique called gradient boosted decision trees using the LightGBM and NGBoost libraries.
Programming Featured Making charts with Plotly and the CFBD Python library It's been awhile since I've done one of these. If you're familiar with my Talking Tech series, this entry will be much shorter. If you follow me on Twitter, you may have seen that the official CFBD Python client library dropped this past weekend. Introducing the official CFBD Python client
Analysis Featured The Four Quadrants of College Football College football can be confusing. It feels like a 13 game regular season can have so many twists and turns that the team you are watching at the end of the season hardly resembles the one in September. So how can we properly group teams to assess their performance? The
Talking Tech Featured Talking Tech: Predicting Play Calls Using a Random Forest Classifier Welcome back to Talking Tech! It's been awhile since our last post. To catch everyone up with where we've been thus far, we first went through setting up an environment for data science using Docker and Project Jupyter. When then went over creating a Simple Rating System for college football
By The Numbers By The Numbers: Previewing the National Championship Game Welcome to By The Numbers, a series in which I plan to preview upcoming games using numerical analysis. I hope to make this a regular feature starting next year. Since there are innumerable games to cover in a given year and only one of me, I am looking for people
Analysis Featured Talking Tech: Charting Data with Plotly In the last edition of Talking Tech, we created our own rating system using an SRS algorithm. We're going to build off of that work to demonstrate how to plot charts in Python using Jupyter notebooks. Before we get started, we need to find a charting library that will suit
Talking Tech Featured Talking Tech: Creating a Simple Rating System Welcome back! We didn't get to have much fun last time, as we were mostly concerned with setting up an environment for data analytics in Python. A few people even offered feedback on different tools and setups they use, which is fantastic. Just as there's more than one way to
Programming Featured Talking Tech: Building an environment for data analysis Welcome to the first edition of Talking Tech! This series is going to be a deeper dive into the weeds of predictive analysis. It will be geared more towards coders and less towards data scientists. In other words, expect to see more code in this series rather than mathematical jargon.
Introducing the CFBD Blog... I'm not quite sure how to begin this initial post. Creating an outlet like this has been on my radar for some time now. Well, it's been on my radar for the past year and a half at the very least. Since starting CollegeFootballData.com, I've always envisioned a place