I should thank my wonderful wife Elizabeth for reminding me – since I could have forgotten with all the championship games happening today. But it is Pi Day!!!! So, go have your favorite type of pie (whether it be pizza, dessert, or simply a mathematical constant) to celebrate the second best mathematical constant date according to the Lunatic. Since Elizabeth and I by good fortune were married on Mole Day, that will always be my favorite day.
The long awaited box scores / schedule / net rankings Excel file has finally been loaded to the website. If you go to the Research link on the menu, you will see that the 2026 Schedule File is now available. The typical disclaimer is there. My process checks to see that the records (by site and by quad) match what is on the NCAA NET Rankings page.
For those of you who wonder why I do this each year, here is the brief story. Each year, we get several people who have demonstrated the power of statistics by building models in order to predict the games – some with tons of success. Bill Kahn had been near the top of the standings multiple years by building Bradley-Terry models. And our 2006 champion, David Shaddick, won by using a model he created with this data for a stats training course at work for analysts. At the end of the day, good statistical techniques can be used to do a great job of forecasting the game – even if it doesn’t capture all the strange possibilities that can happen.
I will apologize in advance. Since I had to change to using the ESPN data and I have so many other things to do before Selection Sunday, there are format differences in variable names and team names from past years. Hopefully, that will not mess up the modelers too much, but you will have to do some extra research into the data.
But considering how behind I am, it seemed like such an appropriate way of celebrating Pi Day – by providing everyone the gift of data!!!!
As always, if you notice any issues with the data, please let me know!!! Enjoy the data!!!