Offseason Chart Dump: Excitement, Heisman's, and Chaotic Webs
Hello everyone! I hope you’re all doing well. These past couple weeks are a busy time at my day job, so I haven’t had as much time to dive into the world of college football data as I would have liked.
To fill the time, I am going to mash up some one-off graphs that I posted on twitter into a smorgasbord of CFB data analysis. If you follow me on twitter, these will mostly likely be familiar. If you aren’t endless scrolling like the rest of us, this is for you!
Digging A Little Deeper Into Excitement
A familiar metric of the newsletter, the Excitement Index is essentially just the sum of all of the changes in win probability. The idea being that lead changes, big game changing plays, and “momentum” swings equal excitement. Of course, if a team comes into a game as a heavy Vegas favorite, it is naturally harder for them to accumulate heavy doses of win probability. In an effort to combat this, we can create a mixed effect model that controls for the strength of both teams. This is an attempt to answer the question: “Regardless of the teams playing, which home venue has the best chance for an exciting game?”.
The result is the top 20 statistically significant venues in terms of the excitement index. Really, when you think about it, no other team is better suited for the top spot than the Purdue Spoilermakers. They return a very dependable Aidan O’Connell, which means ranked B1G teams will be officially put on notice this fall.
Two former powerhouses in Texas and Miami make the list, both looking to return to their former glory. Texas is hoping OSU transfer and one time No.1 recruit Quinn Ewers lives up to his hype and takes Texas back to old heights. Miami on the other hand is ushering in the Mario Cristobal era. They already appear to be one of the trailblazers of the new NIL era, so it will be interesting to see how high Mario can take them.
The Heisman: Is it an efficiency trophy?
One of the projects on my docket is to dive deeper into the Heisman trophy votes and see what sort of metrics drive voters selections. The main vote getters are overwhelmingly QB’s, so that is the best place to start. As a quick peak, I decided to start with ESPN’s QBR, which is their per play efficiency metric. As you can see, the Heisman is not really an efficiency trophy. If it was, Tua Tagovailoa, Mac Jones and Justin Fields could stake a very worthwhile claim to the trophy. Heisman winners have typically been very efficient, but it isn’t a great indicator of whether someone will hoist the trophy (If you want to know what is a good indicator, here is a hint…. its a volume stat!).
The Chaos of Out of Conference Matchups
“What on earth am I looking at?” Im glad you asked, this is what is known as a chord diagram. Teams that are connected via a line have played each other in an out of conference matchup in the CFP Era (2014-2021). The thicker the line, the more matchups they have played against each other. So, for example, Notre Dame and USC have a thick line because they are rivals who play each other every year. The teams are also grouped by their respective conferences, which really brings out some of the scheduled conference vs. conference matchups. The ACC and SEC always play near the end of the year, and you can really see those matchups come to light with this graph.
Here is the same thing for the group of 5. It should be noted that some G5 teams shuffled conferences in the CFP era, but they are still grouped with the conference they were in to finish last season. Side note: if you’re looking for a game to rewatch this summer, I highly suggest 2020 Coastal Carolina vs. BYU. A very exciting game between two potent offenses (including #2 overall draft pick Zach Wilson). The game also came together in basically a week, which makes each teams efforts all the more impressive.
Projects On The Docket
I have a document full of project ideas, but for now I will share some of the ones that I have at least started to explore:
The aforementioned Heisman data dive. Obviously QB’s are at the top of the list, but I do also want to potentially look at what a non QB would need to do to win the trophy.
A look into returning production and how it relates to a team improving their win total from the previous season.
“Home field advantage” but with recruiting data! Essentially: “Team X has a ____ point recruiting advantage over Team Y”.
Hopefully I can share one of these with you soon. In the meantime, I hope this chart dump was something you found interesting.
If you want to dive in to the data like I do, check out @CFB_Data and @cfbfastR on Twitter, where you can learn how to get started in the world of College Football data analysis!
If you want to see more charts and one off analysis, follow my twitter page, @CFBNumbers