En esta edición de Tidy Tuesday del 04 de abril de 2023 trabajamos con los datos de la Premier League temporada 2021-2022, donde nos entrega la data de cada partido y sus resultados. Con esto, quisimos consultar la cantidad de goles de local y de visita de cada equipo, para saber si la capacidad goleadora se ve afectada por el estadio.
Cargamos los paquetes y los datos
pacman::p_load(tidyverse,
tidytuesdayR,
lubridate,
showtext)
data_14 <- tt_load("2023-04-04")---- Compiling #TidyTuesday Information for 2023-04-04 ----
--- There is 1 file available ---
── Downloading files ───────────────────────────────────────────────────────────
1 of 1: "soccer21-22.csv"epl_21_22 <- data_14$`soccer21-22`Luego, cargamos las fuentes elegidas para este gráfico
font_add_google("Fira Sans", "fira")
font_add_google("Quicksand", "quick")
showtext_auto()Calculamos la cantidad de goles, tiros y tiros apuerta de los equipos locales y los equipos visitantes.
home_goals <- epl_21_22 %>%
group_by(HomeTeam) %>%
summarise(home_goals = sum(FTHG),
home_shots = sum(HS),
home_shot_target = sum(HST))
away_goals <- epl_21_22 %>%
group_by(AwayTeam) %>%
summarise(away_goals = sum(FTAG),
away_shots = sum(AS),
away_shot_target = sum(AST))
head(home_goals)# A tibble: 6 × 4
HomeTeam home_goals home_shots home_shot_target
<chr> <dbl> <dbl> <dbl>
1 Arsenal 35 347 116
2 Aston Villa 29 235 86
3 Brentford 22 233 77
4 Brighton 19 280 67
5 Burnley 18 229 73
6 Chelsea 37 326 115Unimos los resultados de ambas tablas en una sola consolidada
tot_goal_shot <- home_goals %>%
left_join(away_goals, by = c("HomeTeam" = "AwayTeam")) %>%
rename("Team" = HomeTeam)
head(tot_goal_shot)# A tibble: 6 × 7
Team home_goals home_shots home_shot_target away_goals away_shots
<chr> <dbl> <dbl> <dbl> <dbl> <dbl>
1 Arsenal 35 347 116 26 241
2 Aston Villa 29 235 86 23 219
3 Brentford 22 233 77 26 209
4 Brighton 19 280 67 23 210
5 Burnley 18 229 73 16 178
6 Chelsea 37 326 115 39 266
# ℹ 1 more variable: away_shot_target <dbl>Creamos el dataset segmentos que permite dibujar las líneas entre puntos.
segmentos <- tot_goal_shot %>%
select(c(Team, home_goals, away_goals)) %>%
mutate(goal_dif = home_goals - away_goals,
Team = fct_reorder(Team, desc(Team)))
# levels(segmentos$Team)
head(segmentos)# A tibble: 6 × 4
Team home_goals away_goals goal_dif
<fct> <dbl> <dbl> <dbl>
1 Arsenal 35 26 9
2 Aston Villa 29 23 6
3 Brentford 22 26 -4
4 Brighton 19 23 -4
5 Burnley 18 16 2
6 Chelsea 37 39 -2Luego selecciinamos solo los goles, reordenamos los equipos para que aparezcan alfabéticamente, y modificamos la forma de la tabla, para que quede en un formato “long”.
tot_long <- tot_goal_shot %>%
select(c(Team, home_goals, away_goals)) %>%
mutate(Team = fct_reorder(Team, desc(Team))) %>%
pivot_longer(cols = 2:3, names_to = "H_A", values_to = "goals")
head(tot_long)# A tibble: 6 × 3
Team H_A goals
<fct> <chr> <dbl>
1 Arsenal home_goals 35
2 Arsenal away_goals 26
3 Aston Villa home_goals 29
4 Aston Villa away_goals 23
5 Brentford home_goals 22
6 Brentford away_goals 26Pasamos el dataset en formato “long” a ggplot2 y configuramos todos los detalles.
tot_long %>%
ggplot(aes(y = Team, group = Team)) +
geom_segment(data = segmentos,
aes(x = home_goals, xend = away_goals, yend = Team),
color = "gray40",
linewidth = 0.9) +
geom_point(aes(x = goals, color = H_A), size = 2.5) +
geom_text(data = segmentos,
aes(y = Team, x = (away_goals + goal_dif*0.5), label = abs(goal_dif)),
position = position_nudge(y = 0.4),
size = 6) +
scale_color_manual(values = c("#e90052", "#38003c"), # Colores oficiales PL
labels = c("Goles de visita", "Goles en casa"),
name = NULL) +
labs(x = "Goles totales", y = NULL, caption = "@AdasmeBenja - Datos de TidyTuesday 04-04-2023",
title = "De local se golea más, (casi) siempre",
subtitle = "Goles de local y de visita para cada club de Premier League 21-22") +
coord_cartesian(expand = F,
clip = "off") +
theme_minimal(base_family = "quick", base_size = 16) +
theme(
#plot.background = element_rect(fill = "#07F2F2"),
title = element_text(family = "fira"),
axis.text.y = element_text(face = "bold"),
legend.position = "bottom",
legend.direction = "horizontal",
legend.text = element_text(size = 14)#,
# legend.key.size = unit(0.5, "cm")
)
Guardamos el gráfico como imágen para exportar y compartir.
ggsave("2023-w14-premierleague2.jpg", dpi = 300, width = 6, height = 4)