Arsenal didn’t win by scoring the most goals — they won by conceding the fewest

Manchester City scored the league’s most goals. Arsenal kept 19 clean sheets, went 17–2–0 in those matches, and earned 53 of their 85 points without conceding.

Arsenal won the 2025–26 Premier League not by scoring the most goals but by conceding the fewest, keeping a league-high 19 clean sheets and earning 62% of their points without conceding. A two-panel makeover pairs a goals-for/against quadrant with a points-by-goals-conceded cliff. Built in R with ggplot2 and patchwork.

Original

The original visualization comes from Arsenal 25/26 Champions

Original visualization

Makeover

Figure 1: Two-panel data visualization titled “Arsenal didn’t win by scoring the most goals — they won by conceding the fewest.” The left panel is a scatter plot of all 20 Premier League clubs in the 2025–26 season, plotting goals scored (vertical) against goals conceded (horizontal axis reversed, so fewer conceded sit to the right). Arsenal, highlighted in red, sits alone in the top-right corner: 71 goals scored and a league-low 27 conceded. Manchester City scored the most goals (77) but conceded more (35) and finished second. The right panel is a bar chart of Arsenal’s points by goals conceded per match: 53 points came in clean-sheet games, 27 when conceding one, and just 5 when conceding two or more — 62% of their title-winning points arrived without conceding. Source: FBref, Premier League 2025–26.

Steps to Create this Graphic

1. Load Packages & Setup

```{r}
#| label: load
#| warning: false
#| message: false      
#| results: "hide"     

## 1. LOAD PACKAGES & SETUP ----
suppressPackageStartupMessages({
  if (!require("pacman")) install.packages("pacman")
  pacman::p_load(
    tidyverse, ggtext, showtext, scales, glue, janitor, patchwork
)
})

### |- figure size ----
camcorder::gg_record(
  dir    = here::here("temp_plots"),
  device = "png",
  width  = 13,
  height = 7,
  units  = "in",
  dpi    = 320
)

# Source utility functions
suppressMessages(source(here::here("R/utils/fonts.R")))
source(here::here("R/utils/social_icons.R"))
source(here::here("R/utils/image_utils.R"))
source(here::here("R/themes/base_theme.R"))
```

2. Read in the Data

```{r}
#| label: read
#| include: true
#| eval: true
#| warning: false
#| 

df_raw <- read_csv(
  here::here("data/MakeoverMonday/2026/premier_league_player_match_stats.csv")) |>
  clean_names()
```

3. Examine the Data

```{r}
#| label: examine
#| include: true
#| eval: true
#| results: 'hide'
#| warning: false

glimpse(df_raw)
skimr::skim_without_charts(df_raw)
```

4. Tidy Data

```{r}
#| label: tidy
#| warning: false

### |- one row per team per game, goals from the authoritative score ----
game_results <- df_raw |>
  distinct(game_id, match_week, team, home_team, away_team, match_score) |>
  mutate(
    home_goals = as.integer(str_extract(match_score, "^\\d+")),
    away_goals = as.integer(str_extract(match_score, "\\d+$")),
    gf = if_else(team == home_team, home_goals, away_goals),
    ga = if_else(team == home_team, away_goals, home_goals),
    points = case_when(
      gf > ga ~ 3L,
      gf == ga ~ 1L,
      TRUE ~ 0L
    )
  )

### |- Panel A: team season totals + league context ----
team_totals <- game_results |>
  summarise(
    gf = sum(gf),
    ga = sum(ga),
    .by = team
  ) |>
  mutate(
    gd        = gf - ga,
    highlight = if_else(team == "Arsenal", "Arsenal", "Other")
  )

# League average (sum of GF == sum of GA, so a single line value)
avg_line <- mean(team_totals$gf)

### |- Panel B: Arsenal points by goals-conceded bucket ----
arsenal_buckets <- game_results |>
  filter(team == "Arsenal") |>
  mutate(
    bucket = case_when(
      ga == 0 ~ "0",
      ga == 1 ~ "1",
      TRUE ~ "2+"
    )
  ) |>
  summarise(
    games = n(),
    points = sum(points),
    wins = sum(points == 3L),
    draws = sum(points == 1L),
    losses = sum(points == 0L),
    .by = bucket
  ) |>
  mutate(
    bucket    = factor(bucket, levels = c("0", "1", "2+")),
    pts_game  = points / games,
    pct_total = points / sum(points)
  ) |>
  arrange(bucket)
```

5. Visualization Parameters

```{r}
#| label: params
#| include: true
#| warning: false

### |-  plot aesthetics ----
clrs <- get_theme_colors(
  palette = list(
    arsenal = "#EF0107",
    neutral = "#8C857B",
    faint   = "#C9C3B8",
    dark    = "#403B34"
  )
)

# Unpack colors
col_arsenal <- clrs$palette$arsenal
col_neutral <- clrs$palette$neutral
col_faint <- clrs$palette$faint
col_dark <- clrs$palette$dark

### |-  titles and caption ----
title_text <- str_glue(
  "Arsenal didn't win by scoring the most goals — they won by conceding the fewest"
)

subtitle_text <- str_glue(
  "Manchester City scored the league's most goals. Arsenal kept 19 clean sheets, ",
  "went 17–2–0 in those matches,<br>and earned 53 of their 85 points without conceding."
)

caption_text <- create_mm_caption(
  mm_year     = 2026,
  mm_week     = 22,
  source_text = "FBref · Premier League 2025-26 player match stats<br>Goals for/against derived from final match scores"
)

### |-  fonts ----
setup_fonts()
fonts <- get_font_families()

### |-  plot theme ----
base_theme <- create_base_theme(clrs)

weekly_theme <- extend_weekly_theme(
  base_theme,
  theme(
    plot.title.position = "plot",
    plot.caption.position = "plot",
    axis.title = element_text(size = rel(0.85), color = col_neutral),
    axis.text = element_text(size = rel(0.8))
  )
)

theme_set(weekly_theme)
```

6. Plot

```{r}
#| label: plot
#| warning: false

### |-  annotation coordinates ----
# Panel A — point coordinates
ars_ga <- 27
ars_gf <- 71
city_ga <- 35
city_gf <- 77

# Panel A — label anchors
ars_lab_x <- 40
ars_lab_y <- 66
city_lab_x <- 40
city_lab_y <- 82
avg_lab_x <- max(team_totals$ga)

# Panel B — label anchors
cliff_note_x <- 2.5
cliff_note_y <- 50

### |-  Panel A — Championship Quadrant ----
p_quadrant <- ggplot(team_totals, aes(x = ga, y = gf)) +
  # Geoms
  geom_hline(yintercept = avg_line, color = col_faint, linewidth = 0.4) +
  geom_vline(xintercept = avg_line, color = col_faint, linewidth = 0.4) +
  geom_point(
    data = filter(team_totals, highlight == "Other"),
    color = col_neutral, size = 2.8, alpha = 0.9
  ) +
  geom_point(
    data = filter(team_totals, highlight == "Arsenal"),
    color = col_arsenal, size = 4.5
  ) +
  # Annotate
  annotate(
    "text",
    x = avg_lab_x, y = avg_line + 1.8,
    label = "League average", color = col_neutral,
    family = fonts$text, size = 3, hjust = 0, vjust = 0
  ) +
  annotate(
    "richtext",
    x = ars_lab_x, y = ars_lab_y,
    label = "**Arsenal** — Champions<br>71 scored · 27 conceded",
    color = col_arsenal, family = fonts$text, size = 3.6,
    hjust = 0, vjust = 1, fill = NA, label.color = NA
  ) +
  annotate(
    "richtext",
    x = city_lab_x, y = city_lab_y,
    label = "**Manchester City**<br>77 scored · 35 conceded<br>Finished 2nd",
    color = col_neutral, family = fonts$text, size = 3.4,
    hjust = 0, vjust = 0, fill = NA, label.color = NA
  ) +
  # Scales
  scale_x_reverse() +
  coord_cartesian(clip = "off") +
  # Labs
  labs(
    x = "Fewer goals conceded →",
    y = "More goals scored →",
    subtitle = "Premier League clubs, 2025–26 season"
  ) +
  # Theme
  theme(
    plot.subtitle = element_text(
      family = fonts$text, size = rel(0.95), color = col_neutral,
      margin = margin(b = 10)
    )
  )

### |-  Panel B — Clean-Sheet Points Cliff ----
p_cliff <- ggplot(arsenal_buckets, aes(x = bucket, y = points, fill = bucket)) +
  geom_col(width = 0.66) +
  # Geoms
  geom_text(
    aes(label = points),
    vjust = -0.4, family = fonts$text, fontface = "bold",
    size = 4.4, color = col_neutral
  ) +
  # Annotate
  annotate(
  "richtext",
  x = cliff_note_x - 1.1,
  y = cliff_note_y,
  label = str_glue(
  "**53 of 85 points**<br>",
  "62% of Arsenal's<br>",
  "title-winning points came<br>",
  "in clean-sheet matches<br>",
  "<span style='font-size:9pt;color:{col_neutral}'>17–2–0 when conceding zero</span>"
),
  family = 'sans',
  color = col_dark,
  size = 3.2,
  hjust = 0,
  vjust = 1,
  lineheight = 1.2,
  fill = NA,
  label.color = NA
) +
  # Scales
  scale_fill_manual(
    values = c("0" = col_arsenal, "1" = col_neutral, "2+" = col_neutral)
  ) +
  scale_y_continuous(expand = expansion(mult = c(0, 0.15))) +
  # Labs
  labs(
    x = "Goals conceded in a match",
    y = NULL,
    subtitle = "The title was built on clean sheets"
  ) +
  guides(fill = "none") +
  # Theme
  theme(
    panel.grid = element_blank(),
    axis.text.y = element_blank(),
    axis.ticks = element_blank(),
    plot.subtitle = element_text(
      family = fonts$text, size = rel(0.95), color = col_dark,
      face = "bold", margin = margin(b = 10)
    )
  )

### |-  Combine plots ----
p_combined <- p_quadrant + p_cliff +
  plot_layout(widths = c(1.15, 0.85)) +
  plot_annotation(
    title = title_text,
    subtitle = subtitle_text,
    caption = caption_text,
    theme = theme(
      plot.title = element_text(
        family = fonts$title_1, face = "bold", size = rel(1.8),
        margin = margin(b = 6)
      ),
      plot.subtitle = element_textbox_simple(
        family = fonts$subtitle, size = rel(0.95), color = col_neutral,
        lineheight = 1.1, margin = margin(b = 14)
      ),
      plot.caption = element_markdown(
      size = 6.5, family = 'sans', color = col_neutral,
      hjust = 0, margin = margin(t = 10), lineheight = 1.3
      )
    )
  )
```

7. Save

```{r}
#| label: save
#| warning: false

### |-  plot image ----  
save_plot_patchwork(
  plot = p_combined , 
  type = "makeovermonday", 
  year = current_year,
  week = current_week,
  width = 13, 
  height = 7
  )
```

8. Session Info

TipExpand for Session Info

R version 4.5.3 (2026-03-11 ucrt)
Platform: x86_64-w64-mingw32/x64
Running under: Windows 11 x64 (build 26100)

Matrix products: default
  LAPACK version 3.12.1

locale:
[1] LC_COLLATE=English_United States.utf8 
[2] LC_CTYPE=English_United States.utf8   
[3] LC_MONETARY=English_United States.utf8
[4] LC_NUMERIC=C                          
[5] LC_TIME=English_United States.utf8    

time zone: America/New_York
tzcode source: internal

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
 [1] here_1.0.2      patchwork_1.3.2 janitor_2.2.1   glue_1.8.0     
 [5] scales_1.4.0    showtext_0.9-8  showtextdb_3.0  sysfonts_0.8.9 
 [9] ggtext_0.1.2    lubridate_1.9.5 forcats_1.0.1   stringr_1.6.0  
[13] dplyr_1.2.1     purrr_1.2.2     readr_2.2.0     tidyr_1.3.2    
[17] tibble_3.3.1    ggplot2_4.0.3   tidyverse_2.0.0 pacman_0.5.1   

loaded via a namespace (and not attached):
 [1] gtable_0.3.6       xfun_0.57          htmlwidgets_1.6.4  tzdb_0.5.0        
 [5] yulab.utils_0.2.4  vctrs_0.7.3        tools_4.5.3        generics_0.1.4    
 [9] curl_7.0.0         parallel_4.5.3     gifski_1.32.0-2    pkgconfig_2.0.3   
[13] ggplotify_0.1.3    skimr_2.2.2        RColorBrewer_1.1-3 S7_0.2.1          
[17] lifecycle_1.0.5    compiler_4.5.3     farver_2.1.2       textshaping_1.0.5 
[21] repr_1.1.7         codetools_0.2-20   snakecase_0.11.1   litedown_0.9      
[25] htmltools_0.5.9    yaml_2.3.12        crayon_1.5.3       pillar_1.11.1     
[29] camcorder_0.1.0    magick_2.9.1       commonmark_2.0.0   tidyselect_1.2.1  
[33] digest_0.6.39      stringi_1.8.7      labeling_0.4.3     rsvg_2.7.0        
[37] rprojroot_2.1.1    fastmap_1.2.0      grid_4.5.3         cli_3.6.6         
[41] magrittr_2.0.5     base64enc_0.1-6    withr_3.0.2        rappdirs_0.3.4    
[45] bit64_4.6.0-1      timechange_0.4.0   rmarkdown_2.31     bit_4.6.0         
[49] otel_0.2.0         hms_1.1.4          evaluate_1.0.5     haven_2.5.5       
[53] knitr_1.51         markdown_2.0       gridGraphics_0.5-1 rlang_1.2.0       
[57] gridtext_0.1.6     Rcpp_1.1.1         xml2_1.5.2         svglite_2.2.2     
[61] rstudioapi_0.18.0  vroom_1.7.1        jsonlite_2.0.0     R6_2.6.1          
[65] fs_2.0.1           systemfonts_1.3.2 

9. GitHub Repository

TipExpand for GitHub Repo

The complete code for this analysis is available in mm_2026_22.qmd.

For the full repository, click here.

10. References

TipExpand for References

Primary Data (Makeover Monday): 1. Makeover Monday 2026 Week 22: Arsenal 25/26 Champions 2. Original Chart: FBref — Arsenal Standard Stats (2025-26 Premier League) - Article context: Arsenal Analysed: How We Won the Premier League - Coverage: all 20 Premier League clubs, 380 matches, matchweeks 1–38, 2025-26 season

Source Data: 3. FBref — Football Statistics and History - Coverage: player-level statistics for every Premier League fixture, 2025-26 season - Unit: per-player, per-match counts (minutes, goals, assists, shots, cards, tackles, etc.) 4. MakeoverMonday 2026 W22 — data.world - CSV file: premier_league_player_match_stats.csv; 11,492 rows × 39 columns (league, season, game, team, player, minutes, goals, assists, match_score, match_week, …)

Note: Team goals for and against are derived from each match’s final score (match_score) rather than summed player goals — summing player goals omits own goals scored for a team and understates goals for (Arsenal: 71 from the final scores vs. 67 player-scored). Match outcomes (win/draw/loss and points) are reconstructed from the same final scores. The right panel groups Arsenal’s 38 matches by goals conceded (0, 1, 2+); “53 of 85 points” and “62%” refer to championship points earned in matches where Arsenal conceded zero goals (a 17–2–0 record). No causal relationship between clean sheets and the title is asserted — the panel describes the association between defensive outcomes and points accumulation.

11. Custom Functions Documentation

Note📦 Custom Helper Functions

This analysis uses custom functions from my personal module library for efficiency and consistency across projects.

Functions Used:

• fonts.R: setup_fonts(), get_font_families() - Font management with showtext
• social_icons.R: create_social_caption() - Generates formatted social media captions
• image_utils.R: save_plot() - Consistent plot saving with naming conventions
• base_theme.R: create_base_theme(), extend_weekly_theme(), get_theme_colors() - Custom ggplot2 themes

Why custom functions?
These utilities standardize theming, fonts, and output across all my data visualizations. The core analysis (data tidying and visualization logic) uses only standard tidyverse packages.

Source Code:
View all custom functions → GitHub: R/utils