Good question. I'll add the MIT license.

I do have an example github repository. Here you go

https://github.com/bischrob/CFBAnimation

Made in R using these packages: cfbfastR, ggimage, tidyverse, gganimate, ggthemes.

Explanation of FPI efficiency: "Team efficiencies are based on the point contributions of each unit to the team's scoring margin, on a per-play basis. The values are adjusted for strength of schedule and down-weighted for "garbage time" (based on win probability). The scale goes from 0 to 100; higher numbers are better and the average is roughly 50 for all categories. Efficiencies update daily during the season." It's basically an advanced metric designed to measure offense and defense.

Good question, it is cumulative

Week one has outliers past 0.4. It makes a big difference.

I just used the dataset from nflfastR. Here's an explanation of it https://www.nfeloapp.com/analysis/expected-points-added-epa-nfl/

Without them everyone just looks bunched up in the middle at the end

I agree with your top picks. No one else jumps out. As much as I hate to say it, I wouldn't count the Bucs out yet.

A helpful guide

It's been a weird start to the season.

Glad I'm not a Packers fan

Top right is good, bottom left is bad. Made in R using these packages: nflfastR, nflplotR, tidyverse, gganimate, ggthemes.

Explanation of EPA https://www.nfeloapp.com/analysis/expected-points-added-epa-nfl/

They're hidden at first by Utah, but they end up close to the 75, 75 mark

I don't know, but I think there are fewer extreme cases as the season goes on.

It's percent efficiency as rated by ESPN'S formula. 100 is best, zero is worst.

Made in R using gganimate with data from ESPN.com