Analysis 2024

Goals of this notebook

Important

The most recent data for this analysis was released in May 2024. The MLS will update salaries again after the summer transfer window, usually in September.

We’ll explore MLS Salaries through history. We’ll start with the most recent data, then look back historically. A couple of questions that come to mind:

• Which players are getting paid the most this year?
• Which teams have the highest salary bill this year?
• How do team salary rankings compare over time?

Per the MLS Player’s Association, “compensation” is: The Annual Average Guaranteed Compensation (Guaranteed Comp) number includes a player’s base salary and all signing and guaranteed bonuses annualized over the term of the player’s contract, including option years.

Setup

library(tidyverse)
library(janitor)
library(scales)
library(ggrepel)
library(DT)

Import

Getting the cleaned data.

salaries <- read_rds("data-processed/mls-salaries.rds")

salaries |> glimpse()

Rows: 11,249
Columns: 9
$ year         <chr> "2007", "2007", "2007", "2007", "2007", "2007", "2007", "…
$ club_short   <chr> "CHI", "CHI", "CHI", "CHI", "CHI", "CHI", "CHI", "CHI", "…
$ last_name    <chr> "Armas", "Banner", "Barrett", "Blanco", "Brown", "Busch",…
$ first_name   <chr> "Chris", "Michael", "Chad", "Cuauhtemoc", "C.J.", "Jon", …
$ position     <chr> "M", "M", "F", "F", "D", "GK", "F", "D", "M", "D", "D", "…
$ base_salary  <dbl> 225000.0, 12900.0, 41212.5, 2492316.0, 106391.0, 58008.0,…
$ compensation <dbl> 225000.0, 12900.0, 48712.5, 2666778.0, 106391.0, 58008.0,…
$ club_long    <chr> "Chicago Fire FC", "Chicago Fire FC", "Chicago Fire FC", …
$ conference   <chr> "Eastern", "Eastern", "Eastern", "Eastern", "Eastern", "E…

Setting the most recent year of data

I’m creating an object called recent_year because at some point I’ll have new data and might want to just change the year.

recent_year <- "2024"

Players with highest salaries

Over all time

A searchable table of all players, all time.

sal_high <- salaries |> 
  arrange(compensation |> desc()) |> 
  select(!c(club_long, conference))

sal_high |> datatable()

Warning in instance$preRenderHook(instance): It seems your data is too big for
client-side DataTables. You may consider server-side processing:
https://rstudio.github.io/DT/server.html

Data takeaway: Messi money

Upon his signing on July 15, 2023, Lionel Messi became the highest paid player in the history of the MLS with a total compensation of $20.4 million. Lorenzo Insigne of Toronto was second at $15.5 million, the only other player earning more than $10 million within a year.

In most recent year

sal_high_recent <- salaries |> 
  filter(year == recent_year) |> 
  mutate(rank = min_rank(desc(compensation))) |> 
  relocate(rank) |> 
  arrange(compensation |> desc()) |> 
  select(!c(club_long, conference))

sal_high_recent

Teams with more than one player from top 10

sal_high_recent |> 
  filter(rank <= 10) |> 
  count(club_short, sort = T)

Data takeaway: Driussi 5th in early 2024

For the early 2024 release, Austin FC’s Sebastián Driussi is the 5th highest player in the MLS and Houston’s Hector Herrera is the 8th highest-paid player.

Both Inter Miami and Toronto FC have two top 10 earners on their rosters.

Difference with just base pay in 2024?

This looks at just the base salary as opposed to total compensation. No great changes at the top of the list.

sal_high_base <- salaries |> 
  arrange(base_salary |> desc()) |> 
  select(!c(club_long, conference, compensation))

sal_high_base |> filter(year == recent_year, base_salary >= 2000000) 

Team salaries

We’ll get per-year salaries by team, then look at just this year.

Highest team salaries over time

First get the total compensation for each team in each year.

sal_team <- salaries |> 
  group_by(year, club_long) |> 
  summarise(total_compensation = sum(compensation)) |> 
  arrange(total_compensation |> desc())

`summarise()` has grouped output by 'year'. You can override using the
`.groups` argument.

# peek at the top
sal_team |> filter(total_compensation > 20000000)

Then find the top team for each year.

top_sal_team_yr <- salaries |> 
  group_by(year, club_long) |> 
  summarise(total_compensation = sum(compensation)) |> 
  slice_max(total_compensation)

top_sal_team_yr

And note how often the teams have been the top spender.

top_sal_team_yr |> 
  ungroup() |> 
  tabyl(club_long) |> 
  adorn_totals("row") |>
  adorn_pct_formatting() |> 
  as_tibble()

Data takeaways: Toronto historically spends high

Looking at the most expensive rosters in the MLS of time, Toronto FC has six of the top 10 highest entries. Over the past 17 years, Toronto has been the top spending team seven times, or 40% of the time. The L.A. Galaxy is next with five highest-spending years.

Highest salaries this year

And let’s look at this year.

sal_team_recent <- sal_team |> filter(year == recent_year)

# peek
sal_team_recent |> head(10)

Let’s round the numbers for our chart.

sal_team_recent_mil <- sal_team_recent |> 
  mutate(total_millions = (total_compensation / 1000000) |> round(1))

sal_team_recent_mil

Let’s chart this

sal_team_recent_mil_plot <- sal_team_recent_mil |> 
  filter(club_long != "Major League Soccer") |> 
  ggplot(mapping = aes(
    x = total_millions,
    y = club_long |> reorder(total_compensation)
  )) +
  geom_bar(stat='identity') +
  geom_text(aes(label = paste("$", as.character(total_millions), sep = "")), color = "white", hjust = 1.25) +
  scale_x_continuous(labels = label_dollar()) +
  labs(
    x = "Total team spending in $ millions",
    y = "",
    title = "Messi makes Miami top MLS spender in 2023",
    subtitle = str_wrap("Salaries includes each player's base salary plus all signing and guaranteed bonuses annualized over the term of the player's contract, including option years."),
    caption = "By: Christian McDonald. Source: Major League Soccer Players Association"
  ) +
    theme_minimal()


ggsave("figures/team-salary-recent.png")

Saving 7 x 5 in image

One more look to see how many high-paid players on each team.

sal_high_recent |> 
  filter(compensation >= 5000000) |> 
  count(club_short, sort = T)

Data Takeaway: Miami, Toronto tops

Given the historic signing of Lionel Messi in 2023, it is no surprise that Inter Miami have the highest team salary for the 2024 season. Toronto ranks second on the power of having two players making over $5 million, Lorenzo Insigne and Federico Bernardeschi.

Team spending over time

Let’s look at team spending over the past five years. To do this, we have to create a ranking for the spending.

• I’m removing players not affiliated with teams
• When I added a third column to the group because I wanted to use long names for something, the ranking broke. I had to break the group then use the .by argument for rank().

sal_team_rank <- salaries |> 
  filter(club_short != "MLS" | club_short |> is.na()) |> 
  group_by(year, club_short, club_long) |> 
  summarise(
    total_comp = sum(compensation, na.rm = TRUE)
  ) |> 
  arrange(year, total_comp |> desc()) |> 
1  ungroup() |>
2  mutate(rank = rank(-total_comp), .by = year)

1

I break the group_by here.

2

Then I set the ranking to work by year.

`summarise()` has grouped output by 'year', 'club_short'. You can override
using the `.groups` argument.

# peek
sal_team_rank |> head(20)

Visualizing all of them would be tricky. Let’s do the top five over last five years.

sal_team_rank_top <- sal_team_rank |> 
  filter(rank <= 5,
         year >= (as.numeric(recent_year) - 4))

sal_team_rank_top

Peek at this a different way

sal_team_rank_top |> 
  select(-total_comp) |> 
  pivot_wider(names_from = year, values_from = rank)

Let’s visualze spending rank

This has to be adjusted when new data is added.

I want to use a list of team-specific colors. I’m trying this manually first, based on the results of the chart. The colors were pulled from here, though I wish I could instead join with a data package and pull from the team colors, but I haven’t found one that is up-to-date.

sal_team_rank_top |> count(club_short)

club_color_list <- c(
  "#80000A", # ATL
  "#7CCDEF", # CHI
  "#FE5000", # CIN or 003087
  "#FEDD00", # CLB
  "#00245D", # LA
  "#C39E6D", # LAFC
  "#F7B5CD", # MIA
  "#CE0E2D", # NE
  "#ECE83A", #NSH
  # "#5D9741", # SEA
  "#B81137" # TOR
  # "#EF3E42" # DC
  # "#00B140", # AUS
)

And now the chart

sal_team_rank_top_plot <- sal_team_rank_top |> 
  # filter()
  ggplot(aes(x=year, y=rank, group = club_short)) +
  geom_line(aes(color = club_short)) +
  geom_point(aes(color = club_short, size = 3)) +
  geom_label_repel(aes(label = club_short), size = 3) +
  scale_y_reverse() +
  scale_colour_manual(values = club_color_list) +
  scale_size_continuous(guide = "none") +
  labs(
    title = "Miami's spending was increasing before Messi",
subtitle = str_wrap("The L.A. Galaxy are the only MLS team to rank as a top five spender in each of the past five years."),
    color = "Club",
    x = NULL,
    y = "Spending Rank",
    caption = "By: Christian McDonald. Source: Major League Soccer Players Association"
  ) +
  theme_minimal() +
  guides(color = FALSE)

Warning: The `<scale>` argument of `guides()` cannot be `FALSE`. Use "none" instead as
of ggplot2 3.3.4.

# sal_team_rank_top_plot

ggsave("figures/sal_team_rank.png")

Saving 7 x 5 in image

Let’s count how many times each team is in this list.

sal_team_rank_top |> 
  count(club_long, sort = T)

Data Takeaway: Miami and LA

Among the teams spending the most on their rosters over the past five years, Miami has ranked in the top five each year. The LA Galaxy and Toronto FC have been top spenders in four of the last five years.