Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/anowell/fantasy-football-forecasting
Exploring data-driven ideas for managing a FF roster
https://github.com/anowell/fantasy-football-forecasting
Last synced: about 1 month ago
JSON representation
Exploring data-driven ideas for managing a FF roster
- Host: GitHub
- URL: https://github.com/anowell/fantasy-football-forecasting
- Owner: anowell
- Created: 2024-08-11T15:16:37.000Z (5 months ago)
- Default Branch: main
- Last Pushed: 2024-08-13T13:07:33.000Z (4 months ago)
- Last Synced: 2024-08-13T15:54:02.811Z (4 months ago)
- Language: Rust
- Size: 37.1 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Fantasy Football Foresights (FFF)
[Pola.rs](https://pola.rs) exploration of NFL [play-by-play data](https://github.com/nflverse/nflverse-data)
This is mostly an attempt at having an unfair advantage in a FF league with my bro.
It's very exploratory/experimental currently. Some ideas that might get implemented:
- [x] Reliably calculate fantasy points using most widely-supported scoring systems
- [x] Explore fantasy scores for players over custom windows (e.g. last 5 weeks)
- [ ] Explore fantasy score trends for players against certain opponents
- [ ] Explore the impact certain defenses have on fantasy scoring
- [ ] Explore value-over-replacement [VOR](https://github.com/jjti/ff) calculation
- [ ] Explore heuristics that could be used to pick drafts, lineups, and waiver-wire additions.
## Getting started
- [Rust](rustup.rs)
- [Polars CLI](https://github.com/pola-rs/polars-cli) (also recommend [jq](https://jqlang.github.io/jq/))
- [Just](https://github.com/casey/just) (runner)
Download the play-by-play dataset: `just download-pbp 2023`
Running `fff`
```
just run --help
# or
cargo build --release
target/release/fff --help
```
```
Usage: fff [OPTIONS]
Options:
-v, --verbose...
-y, --year Loads data for a given year [default: 2023]
-t, --team Filter by team
-p, --pos Filter by position
-w, --week Filtering week number or range (e.g. 3 or 3-5)
-x, --exclude
--score Calculate fantasy score
--score-by Choose how to aggregate scores [default: player] [possible values: player-game, player, game]
-h, --help Print help
-V, --version Print version
````
You can also just use polars to explore the data:
```
$ just query "select qtr, time, desc from plays where week = 2 AND posteam = 'SEA' AND touchdown = 1.0" -o json
{"qtr":1.0,"time":"07:07","desc":"(7:07) (Shotgun) 9-K.Walker up the middle for 1 yard, TOUCHDOWN."}
{"qtr":3.0,"time":"14:22","desc":"(14:22) (No Huddle, Shotgun) 9-K.Walker left tackle for 3 yards, TOUCHDOWN."}
{"qtr":4.0,"time":"10:45","desc":"(10:45) 7-G.Smith pass short left to 16-T.Lockett for 3 yards, TOUCHDOWN."}
{"qtr":5.0,"time":"05:48","desc":"(5:48) (Shotgun) 7-G.Smith pass short right to 16-T.Lockett for 6 yards, TOUCHDOWN."}
````
## Why Rust?
~It's better than Python.~
I know Rust best. I like it. That's reason enough.
Or: My Python is pretty lousy. My Ruby has been getting Rusty.