Pitchfork is one of the most influential sources of music criticism, especially in the indie, alternative, and experimental genres. Known for its in-depth reviews and rigorous scoring system, Pitchfork reviews can significantly shape public perception and an artist’s career.
In this report, I explore trends and patterns within Pitchfork’s review data, focusing on how scores are distributed, how they’ve changed over time, and how artists and genres are represented.
Data ranging from 1999-2017
library(tidyverse)
library(RSQLite)
library(reactable)
pf <- "database.sqlite"
sqlite.driver <- dbDriver("SQLite")
db <- dbConnect(sqlite.driver,
dbname = pf)
genre <- dbReadTable(db,"genres")
reviews <- dbReadTable(db,"reviews")
artist <- dbReadTable(db,"artists")
reviews |>
inner_join(artist, relationship = "many-to-many") -> mergedPitchfork review scores range from 0.0 to 10.0, but I wanted to know how they are distributed. Are most reviews around the middle? Do perfect scores happen often?
ggplot(merged, aes(x = score)) +
geom_histogram(binwidth = 0.2, fill = "steelblue", color = "white") +
scale_x_continuous(breaks = 0:10, limits = c(0, 10)) +
scale_y_continuous(expand = c(0, 0)) +
labs(title = "Distribution of Pitchfork Scores",
x = "Score",
y = "Count")
This histogram shows that most scores fall between high 6s and low 8s, with very few perfect 10s or extremely low scores. The distribution suggests a tendency toward moderate-to-positive reviews, though there’s a visible skew toward the higher end of the scale.
I was curious whether Pitchfork’s average scores have changed over time — are they more generous or more critical than in the past?
merged |>
mutate(year = lubridate::year(pub_date)) |>
group_by(year) |>
summarise(avg_score = mean(score, na.rm = TRUE)) |>
ggplot(aes(x = year, y = avg_score)) +
geom_line() +
geom_point() +
labs(title = "Average Pitchfork Score by Year", x = "Year", y = "Average Score")
This chart shows some year-to-year variation, but the average scores have remained fairly consistent, mostly hovering between high 6s and low 7s. This consistency suggests that Pitchfork applies a relatively stable editorial standard over time.
I also wanted to explore how scores vary by genre. Are some genres consistently scored higher or lower than others?
merged |>
inner_join(genre, by = "reviewid") |>
filter(!is.na(genre)) |>
ggplot(aes(x = fct_reorder(genre, score), y = score)) +
geom_boxplot() +
coord_flip() +
labs(title = "Pitchfork Scores by Genre", x = "Genre", y = "Score")
This boxplot reveals score distributions across genres. While most genres cluster around a similar median, some show more spread or occasional outliers. Global music had the highest average score (7.42), while Pop/R&b had the lowest (6.87).
Analyzing Pitchfork review data gave me a deeper appreciation for the consistency and subtle biases in editorial music scoring. I found that while Pitchfork rarely gives out extremely low or high scores, it maintains a clear standard over time. The trends by genre and artist also suggest ongoing editorial preferences that shape how music is evaluated and remembered.