Can AI Create The Perfect SEC Football Schedule
Despite a strong personal and professional skepticism towards the current AI boom, I decided to see if it could be useful for one thing: generating complete nonsense for a fun blog post. While I find most AI applications to be generally useless and a detriment to the quality of the internet, I put it to the test on a complex sports problem.
My goal was to have an AI formulate new, permanent conference rivalries for the expanded SEC. No AI was used to write this post, only to generate the matchup data.
The AI Experiment A False Start
Initially, I asked the AI to create three permanent rivals for every SEC team, considering historical games, proximity, and competitiveness. The first attempt looked good on the surface, but it failed a key logistical test. The AI didn't understand that the rivalries had to be reciprocal and couldn't overlap.
The result was a list where teams like Auburn and Alabama were assigned five or six rivals each. That’s simply not how scheduling works. It was a classic example of needing to be painstakingly specific with an AI that lacks real-world context.
Round Two The AI's SEC Rivalry Solution
I refined my prompt: “Create three permanent rivals for each SEC team based upon historical games played, proximity, and closeness of games. Also, each team can only have three rivals and those rivals cannot overlap.”
This time, the AI delivered a viable solution:
- Alabama: Auburn, Tennessee, LSU
- Arkansas: Texas, Texas A&M, Oklahoma
- Auburn: Alabama, Georgia, Mississippi State
- Florida: Georgia, Tennessee, South Carolina
- Georgia: Auburn, Florida, South Carolina
- Kentucky: Mississippi State, Vanderbilt, South Carolina
- LSU: Alabama, Ole Miss, Texas A&M
- Mississippi State: Ole Miss, Auburn, Kentucky
- Missouri: Vanderbilt, Oklahoma, Ole Miss
- Ole Miss: LSU, Mississippi State, Missouri
- Oklahoma: Texas, Missouri, Arkansas
- South Carolina: Georgia, Florida, Kentucky
- Tennessee: Alabama, Florida, Vanderbilt
- Texas: Oklahoma, Texas A&M, Arkansas
- Texas A&M: Texas, LSU, Arkansas
- Vanderbilt: Tennessee, Kentucky, Missouri
How the AI Justified Its Choices
The bot also provided an explanation for its logic:
Why these fit the brief
- Historic, named rivalries are protected
- Competitive/close series with stakes
- Proximity to minimize travel and fuel local intensity
- Missouri’s balance
Some fanbases might argue for alternates (e.g., Arkansas–Missouri or LSU–Florida) every year. In a strict 3‑rival, non‑overlapping construction, including those would force drops elsewhere; the set above retains the largest number of named/historic matchups while keeping geography reasonable and making the network solvable.
A Fan's Analysis of the Results
Frankly, the result is not bad for a machine. I was particularly struck by the AI singling out Missouri in its justification, despite me using a VPN and never mentioning my affiliation. It seems the bot understands that Missouri and Oklahoma are the geographical outliers in the new SEC, though Oklahoma's historical brand power makes it an easier fit.
Looking at Missouri's proposed rivals—Vanderbilt, Oklahoma, and Ole Miss—I'm not opposed. The Ole Miss matchup is new, but the proximity is good, and the programs share a similar history of recent highs and notable past success. As I explored in a previous analysis of potential SEC rivalries, securing the historical game against Oklahoma is the most critical piece for Mizzou. The AI correctly prioritized that.
While some might argue for an Arkansas-Missouri matchup, the AI's decision to keep Arkansas with its former Southwest Conference rivals Texas and Texas A&M makes perfect historical sense. The Hogs also have deeper roots with LSU and Ole Miss. Overall, for a quick and fun project, the AI chatbot assembled a surprisingly coherent and defensible schedule.