TidyTuesday Week 33: Billboard Hot 100 Number Ones

This week we are exploring the Billboard Hot 100 Number Ones Database. This workbook contains substantial data about every song to ever top the Billboard Hot 100 between August 4, 1958 and January 11, 2025.

Graphs of the Billboard 100 :::

1. Python code

import pandas as pd
import numpy as np
from plotnine import *
import patchworklib as pl

billboard = pd.read_csv(
    "https://raw.githubusercontent.com/rfordatascience/tidytuesday/main/data/2025/2025-08-26/billboard.csv"
)
topics = pd.read_csv(
    "https://raw.githubusercontent.com/rfordatascience/tidytuesday/main/data/2025/2025-08-26/topics.csv"
)

rock = billboard[billboard["cdr_genre"] == "Rock"].copy()

rock["rounded_rating"] = np.rint(rock["overall_rating"])
rock["date"] = pd.to_datetime(rock["date"])
rock["decade"] = (rock["date"].dt.year // 10) * 10
rock["decade"] = rock["decade"].astype("category")

# Histogram of Ratings
hist = pl.load_ggplot(
    ggplot(data=rock)
    + geom_histogram(
        aes(x="rounded_rating"), color="darkblue", fill="lightblue", bins=13
    )
    + theme_bw()
    + theme(panel_grid=element_blank(), plot_title=element_text(margin={"b": 5}))
    + scale_x_continuous(breaks=range(0, 11))
    + labs(
        title="Distribuiton of Ratings for Rock Songs that Reached Number One",
        x="Rating \\n (rounded to nearest integer)",
        y="Count",
    )
)


# Box Plot
box = pl.load_ggplot(
    ggplot(rock)
    + geom_boxplot(aes(x="decade", y="rounded_rating", fill="decade"))
    + theme_bw()
    + theme(
        panel_grid=element_blank(),
        legend_position="bottom",
        legend_box_margin=0,
        plot_title=element_text(margin={"b": 5}),
    )
    + labs(
        title="Box plot of Rating of Rock Songs that Reached Number One",
        x="Decade",
        y="Rating \n (rounded to nearest integer)",
        fill="Decade",
    )
)


# Top 10 bands with multiple Number ones
artist_counts = rock["artist"].value_counts()
top10 = artist_counts.head(10)
top10_df = top10.reset_index()
top10_df.columns = ["artist", "count"]
top10_df = top10_df.sort_values(by="count", ascending=True)
top10_df["artist"] = pd.Categorical(
    top10_df["artist"], categories=top10_df["artist"], ordered=True
)
top10_chart = pl.load_ggplot(
    ggplot(top10_df)
    + geom_bar(aes(x="artist", y="count"), stat="identity", fill="steelblue")
    + geom_text(
        aes(x="artist", y="count + 0.5", label="count"),
        va="center",
        format_string="{:.0f}",
        position=position_dodge(width=0.9),
    )
    + coord_flip()
    + theme_bw()
    + theme(panel_grid=element_blank())
    + labs(
        title="Top 10 Rock Artists by Number of Number 1 Songs",
        x="Artist",
        y="Number of Number 1 Hits",
    )
)


# Proportion of Cow Bell Used in Songs
instruments = rock[["cdr_genre", "cowbell", "accordion", "banjo", "clarinet"]].copy()

instruments_long = instruments.melt(
    id_vars="cdr_genre", var_name="instrument", value_name="present"
)


cowbell_plot = pl.load_ggplot(
    ggplot(
        instruments_long.query("present == 1"), aes(x="cdr_genre", fill="instrument")
    )
    + geom_bar(position="fill", width=0.5)
    + scale_y_continuous(labels=lambda l: ["{:.0f}%".format(v * 100) for v in l])
    + labs(
        title="Proportion of Instruments Used in Songs by Genre",
        x="Genre",
        y="Proportion of Songs",
        fill="Instrument",
    )
    + theme_bw()
    + theme(panel_grid=element_blank(), axis_text_x=element_text(rotation=45, hjust=1))
)


combined = (hist | box) / (cowbell_plot | top10_chart)

<Figure size 672x480 with 0 Axes>

<Figure size 672x480 with 0 Axes>

<Figure size 672x480 with 0 Axes>

<Figure size 672x480 with 0 Axes>