R/fetch_fixtures.R

In Chrisjb/fploptimiser: Fantasy Premier League Team Optimisation

#' Fetch Fixtures for next n gameweeks
#'
#' gets fixtures for the next n gameweeks and optionally plots the difficulty.
#'
#' @param n number of fixtures to fetch
#'
#' @import jsonlite
#' @import tidyr
#' @import glue
#' @importFrom magrittr %>%
#' @import ggplot2
#' @export
#'
#' @return a data.frame of fixtures.
#'
#' @examples
#' fixtures <- fetch_fixtures(n = 5)
#'
#' # visualise fixtures
#' plot(fixtures)
#'
#' # find teams with low correlation fixures (good for rotation)
#' fixture_rotation(fixtures)
#'
#'

fetch_fixtures <- function(n) {
  root <- jsonlite::fromJSON("https://fantasy.premierleague.com/api/bootstrap-static/")

  fix <- jsonlite::fromJSON('https://fantasy.premierleague.com/api/fixtures')

  fixtures_df <- data.frame(
    time = as.Date(fix$kickoff_time),
    gw = fix$event,
    home = fix$team_h,
    away = fix$team_a,
    difficulty_home = fix$team_h_difficulty,
    difficulty_away = fix$team_a_difficulty,
    home_opponent = fix$team_a,
    away_opponent = fix$team_h,
    played = fix$finished,
    stringsAsFactors = FALSE
  )

  team_names <- root$teams[c('id','name')]


  fixture_difficulty <- fixtures_df %>%
    tidyr::pivot_longer(cols=c('home','away'), names_to = 'ha',values_to = 'id') %>%
    dplyr::left_join(team_names, by =c('id'='id')) %>%
    mutate(difficulty = if_else(ha == 'home', difficulty_home, difficulty_away),
           difficulty = as.factor(difficulty),
           opponent_id = if_else(ha == 'home', home_opponent, away_opponent)) %>%
    dplyr::left_join(team_names, by = c('opponent_id'='id'), suffix = c('','_opponent')) %>%
    filter(played != T) %>%
    select(time, gw, team_id= id, team_name = name, ha, played, opponent = name_opponent, difficulty)

  # deal with double fixtures
  fd_summary <- fixture_difficulty %>%
    mutate(ha = if_else(ha =='home','h','a')) %>%
    group_by(gw, team_name) %>%
    arrange(difficulty) %>%
    filter(!duplicated(team_name)) %>%
    ungroup()



  next_n<- fd_summary %>%
    filter(gw %in% seq(min(gw,na.rm=T), min(gw, na.rm=T)+n-1))

  avg_diff <- next_n %>%
    group_by(team_name) %>%
    summarise(mean_difficulty = mean(as.numeric(as.character(difficulty))),
              median_difficulty = median(as.numeric(as.character(difficulty)))) %>%
    arrange(mean_difficulty)

  next_n  <- next_n %>%
    left_join(avg_diff, by = c('team_name' = 'team_name'))

  class(next_n) <- c('fpl_fixtures',class(next_n))

  next_n %>%
    arrange(time, team_name)


}