Pull xG data from StatsBomb and Opta (via FBRef) for the 2023 FIFA Women's World Cup

pull-2023-fifa-womens-world-cup-xg.md

Raw data pull

library(StatsBombR)
library(worldfootballR)
library(dplyr)
library(janitor)
library(tibble)

sb_free_comps <- FreeCompetitions()
sb_comp <- filter(
  sb_free_comps, 
  competition_name == "Women's World Cup",
  season_name == '2023'
)
sb_matches <- FreeMatches(sb_comp)
sb_shots <- free_allevents(MatchesDF = sb_matches) |> 
  as_tibble() |> 
  clean_names() |> 
  filter(!is.na(shot_statsbomb_xg))

fb_match_urls <- fb_match_urls(
  country = '', 
  gender = 'F', 
  season_end_year = 2023, 
  tier = '', 
  non_dom_league_url = 'https://fbref.com/en/comps/106/history/Womens-World-Cup-Seasons'
)

opta_shots <- fb_match_shooting(fb_match_urls) |> 
  as_tibble() |> 
  clean_names()

Extract non-penalty shots and bring the data sets together.

sb_np_shots <- sb_shots |> 
  filter(
    shot_type_name != 'Penalty'
  ) |>
  transmute(
    xg = shot_statsbomb_xg,
    g = shot_outcome_name == 'Goal'
  )

opta_np_shots <- opta_shots |> 
  mutate(across(x_g, as.numeric)) |> 
  ## heuristic for penalties
  filter(
    !(distance == '13' & x_g == 0.79)
  ) |> 
  transmute(
    xg = x_g,
    g = outcome == 'Goal'
  )
  
np_shots <- bind_rows(
  sb_np_shots |> mutate(source = 'StatsBomb'),
  opta_np_shots |> mutate(source = 'Opta')
)