American Football Insights

One of the fun things about working on this project is, I'm watching an NFL game, I see something exciting, and I think 'How can I highlight that?'.

In week 1 of the 2024 season, the most exciting thing I saw was Anthony Richardson's incredible 60-yard touchdown pass to Alec Pierce:

I had it in my mind to generate a weekly-review, which would highlight notable things that happened in that gameweek. One of those things is definitely big plays, so I said I'd start with that.

Interface

I spent a decent bit of time on the interface to this. Initially, it was just a table of stats, but it seemed too much like reading an Excel spreadsheet. Then I changed it to text, but it started getting too long-winded. So I hit on a hybrid of table and text, which I'm quite happy with.

Up next

I want to do a bit more with the weekly-review. I think it will be text-based, highlighting notable things in the week. e.g.

• The 3-0 Chiefs won in LA, leaving them the only unbeaten team this season.
• The rookie Jayden Daniels had his best game of the season, throwing for 324 yards and 2 touchdowns on Monday Night Football.

That kind of thing, anything notable. I'll play around with it, and see what happens.

The problem: there are about 16 games played per-week, I don't have time to watch them all. I wanted to see which were the most exciting ones, in a spoiler-free way, and then watch selected games.

Here are my initial results from week 1 of the 2024 season:

So, per my graph, the three most exciting games were:

• Green Bay at Philadelphia, on Friday night.
• Houston at Indianapolis, on Sunday afternoon.
• LA at Detroit, on Sunday Night Football.

You can click on the bars to see the details, but what you'll tend to see is high-scoring close games.

How did I generate the excitement-value?

There are a few factors:

• I started off with the idea that exciting games tend to be close games. So I calculate average score-difference per-play.
• But not all close-games are that exciting, you can have a 9-6 game that is technically close, but will not be exciting. So I made number of touchdowns scored a factor.
• Closeness in the 4th quarter is more significant than other quarters, so I made that a factor.

And so on.It's kind of a basic calculation so far, I will refine it as I go on.

How can I improve it?

• Refine the excitement calculation. Add more factors, lead-changes, overtime, turnovers, big-plays.
• Explain the excitement-calculation interactively, show why the game was exciting, instead of having a black-box excitement-factor.
• Break down the excitement per-quarter. Some games may have been unexciting as a whole, but had one quarter that was; I'd like to highlight this.

Conclusions

• Despite it being a fairly basic calculation, it's already useful.
• I used it particularly on Sunday Night Football and Monday Night Football. These games are on at about 1am Irish Time, where I am, and I can't watch them live. I can watch them the next day with NFL Gamepass, but I want to know in advance if it'll be worth my while.
• The stats told me LA-Detroit was an exciting game, so I watched it. It was an exciting game; the Lions jumped out to a 17-3 lead, before LA staged a big comeback to lead 20-17. The Lions ultimately won, 26-20, with a 4th-quarter touchdown.
• The stats also told me the Monday night game between San Francisco and the New York Jets was not an exciting game, so I skipped it. Looking at the report, San Francisco won easily, and the Jets benched their starting quarterback, Aaron Rodgers, because the game was out-of-reach.
• I've got a useful insight here. It needs refining, but it's encouraging that it's useful in this basic form.