Team colors

palette and team_colors give you colors keyed by team, so a chart can be colored without hand-picking hex codes.

import sdvplot

import polars as pl

df = pl.DataFrame({"team": ["KC", "BUF", "PHI", "DET"], "epa": [0.12, 0.10, 0.09, 0.15]})
df

team	epa
KC	0.12
BUF	0.1
PHI	0.09
DET	0.15

A palette for seaborn

palette accepts a polars Series of abbreviations, so it can be passed straight to seaborn.

import matplotlib.pyplot as plt
import seaborn as sns

sns.barplot(
    df.to_pandas(),
    x="team",
    y="epa",
    hue="team",
    palette=sdvplot.palette("nfl", teams=df["team"]),
    legend=False,
)
plt.show()

One color per team

team_colors returns one color per team, in the order given. which="secondary" picks the secondary color.

sdvplot.team_colors(df["team"], "nfl", which="secondary")

shape: (4,)
Series: 'team' [str]
[
	"#ffb612"
	"#c60c30"
	"#a5acaf"
	"#b0b7bc"
]

Other libraries

The palette is a plain dict of team -> hex, so other libraries take it directly (not executed here).

Plotly:

import plotly.express as px

px.bar(df.to_pandas(), x="team", y="epa", color="team",
       color_discrete_map=sdvplot.palette("nfl", teams=df["team"]))

Altair:

import altair as alt

p = sdvplot.palette("nfl", teams=df["team"])
alt.Chart(df.to_pandas()).mark_bar().encode(
    x="team", y="epa", color=alt.Color("team", scale=alt.Scale(domain=list(p), range=list(p.values()))),
)

Fallback colors

Not every league has official colors. Where none exist, color_source is "fallback" and the color is one from a colorblind-safe categorical palette that only keeps teams distinguishable. Check color_source before treating a color as a team's own. Every OHL team below is a fallback; every NFL team has colors from nflverse.

sdvplot.teams("ohl").select("team_id", "name", "color_primary", "color_source").head()

team_id	name	color_primary	color_source
1	Brantford Bulldogs	#76b7b2	fallback
10	Kitchener Rangers	#b07aa1	fallback
11	Owen Sound Attack	#bab0ac	fallback
12	Sudbury Wolves	#76b7b2	fallback
13	Flint Firebirds	#f28e2b	fallback

Run it yourself

Download the notebook (outputs cleared) or open it on GitHub.