By Bethany Melitz, GPS Consultant.
We are approaching the best season of the year. This is the time for crisp air, apples, pumpkin spice lattes, and my favorite: football. Last season, I was able to see a couple college football games in person, but more than anything, I LOVE watching NFL football every Sunday.
When I saw that USA Today published an article on August 24th about the use of data mining in football, I was thrilled to see a collision between what I do for a job, and what I do in my spare time.
The article quotes several NCAA football coaches, and it is remarkable how so many of their goals and concerns echo exactly what we hear in private industry.
“One of the issues is the sense of being overwhelmed by reams and reams of data – inundated by potentially helpful information yet feeling unable to correctly process what’s valuable or what’s extraneous.
“You can go a thousand different directions,” Washington coach Chris Petersen said. “I want to make sure that we can weed through all this stuff, and tell us what we need to know. It can be paralysis by analysis…’”
I am currently working in a 100-year-old metal foundry for the automotive industry. They have an enormous amount of data at their fingertips and teams of people all reacting to different metrics. They regularly find themselves paralyzed by the data and different levels of analysis throughout all levels of the organization.
On this project, one of the first things we did was identify which metrics matter, when and how they need to react to them. One good foundry example relates to sand scrap—the top dollar loss in many metal foundries. There are at least 30 factors that contribute to creating this type of scrap. Using data mining and statistical modeling, we were able to recognize 5 factors that have the biggest impact, their best process operating ranges, and develop reaction rules and checks for those factors of interest. They are currently sustaining their best scrap levels in company’s history.
Last year I was in 3 Fantasy Football Leagues, 3 Survivor Pools, and 2 straight pick-em leagues. I may not be a long-term sports nut, or even know many of the old-time players, but I know data mining and statistical analysis strategies. I ended up winning almost every opportunity! This year, I already have multiple databases to help me answer strategic sports questions. It’s the analytics that will give me the edge. As long as I have worked in statistics, the data mining and analysis has been pretty profitable for me come kickoff. An unlikely candidate for top-of-the-league, people are blindsided by my prowess.
You might think you know football, or your business, but if you don’t know how to identify what’s important and react to it, you’ll be stuck running the same plays and getting nowhere… or rooting for the Dallas Cowboys. It might be time to call an expert.