Merge pull request #2 from jruizcabrejos/understat_match_

Understat match review

R/understat_match_players.R

@@ -1,51 +1,68 @@
 
 #' Get Understat match player data
 #'
-#' Returns player values for a selected match from Understat.com
+#' Returns player values for a selected match from Understat.com.
 #'
-#' @param match_url the URL of the match played
+#' @param match_url A `character` string with the URL of the match played.
 #'
-#' @return returns a dataframe with data for all players for the match
+#' @return returns a `data.frame` with data for all players for the match.
 #'
 #' @importFrom magrittr %>%
 #'
 #' @export
 
 understat_match_players <- function(match_url) {
-  # .pkg_message("Scraping all shots for match {match_url}. Please acknowledge understat.com as the data source")
 
-match_id <- gsub("[^0-9]", "", match_url)
+  match_id <- gsub("[^0-9]", "", match_url)
 
-match_player_data <- .get_understat_json(page_url = match_url)  %>%
+  match_player_data <- .get_understat_json(page_url = match_url)  %>%
     rvest::html_nodes("script") %>%
     as.character()
 
-match_player_data <- match_player_data[grep("rostersData\t=", match_player_data)] %>%
-  stringi::stri_unescape_unicode() %>%
+  match_player_data <- match_player_data[grep("rostersData\t=", match_player_data)] %>%
+    stringi::stri_unescape_unicode() %>%
 
-  substr(41,nchar(.)) %>%
-  substr(0,nchar(.)-13) %>%
-  paste0('[', . , ']') %>%
+    substr(41,nchar(.)) %>%
+    substr(0,nchar(.)-13) %>%
+    paste0('[', . , ']') %>%
 
-  unlist() %>%
-  stringr::str_subset("\\[\\]", negate = TRUE)
+    unlist() %>%
+    stringr::str_subset("\\[\\]", negate = TRUE)
 
-match_player_data <- lapply(match_player_data, jsonlite::fromJSON) %>%
-  do.call("rbind", .)
+  match_player_data <- lapply(match_player_data, jsonlite::fromJSON) %>%
+    do.call("rbind", .)
 
-match_player_data_home <- do.call(rbind.data.frame, match_player_data$h)
-match_player_data_away <- do.call(rbind.data.frame, match_player_data$a)
+  match_player_data_home <- do.call(rbind.data.frame, match_player_data$h)
+  match_player_data_away <- do.call(rbind.data.frame, match_player_data$a)
 
-match_player_data <- bind_rows(match_player_data_home,match_player_data_away) %>%
-  mutate(match_id = match_id)  %>%
+  match_player_data <- dplyr::bind_rows(match_player_data_home,match_player_data_away) %>%
 
-  select(match_id,  team_id,
-         team_status = h_a,
-         player_id, swap_id = id,
-         player, position, positionOrder,
-         time_played = time,
-         everything()) %>%
-  mutate(team_status = ifelse(team_status=="h","home","away"))
+    dplyr::mutate(match_id = as.integer(match_id),
+                  team_id = as.integer(team_id),
+                  team_status = as.character(team_status),
+                  player_id = as.integer(player_id),
+                  swap_id = as.integer(id),
+                  player = as.character(player),
+                  position = as.character(position),
+                  positionOrder = as.integer(positionOrder),
+                  time_played =  as.integer(time),
+                  goals = as.integer(goals),
+                  own_goals = as.integer(own_goals),
+                  shots = as.integer(shots),
+                  xG = as.numeric(xG),
+                  yellow_card = as.integer(yellow_card),
+                  red_card = as.integer(red_card),
+                  roster_in = as.integer(roster_in),
+                  roster_out = as.integer(roster_out),
+                  key_passes = as.integer(key_passes),
+                  assists = as.integer(assists),
+                  xA = as.numeric(xA),
+                  xGChain = as.numeric(xGChain),
+                  XGBuildup = as.numeric(XGBuildup))  %>%
+
+    dplyr::select(-c(time,id)) %>%
+
+    dplyr::mutate(team_status = ifelse(team_status=="h","home","away"))
 
   return(match_player_data)
 }

R/understat_match_stats.R

@@ -1,18 +1,19 @@
 
 #' Get Understat match stats table data
 #'
-#' Returns the Stats values for a selected match from Understat.com
+#' Returns the Stats values for a selected match from Understat.com.
 #'
-#' @param match_url the URL of the match played
+#' @param match_url A `character` string with the URL of the match played.
 #'
-#' @return returns a dataframe with data from the stats table for the match
+#' @return returns a `data.frame` with data from the stats table for the match.
+#'
+#' @details For definitions on each variable (e.g. PPDA), you can hover with your mouse over each of them in the Stats section of the match at understat.com. For `draw_chances`, `home_chances` and `away_chances`, values below 10% will be retrieved as NA, however they sometimes have a value inside the match website (e.g. "5%").
 #'
 #' @importFrom magrittr %>%
 #'
 #' @export
 
 understat_match_stats <- function(match_url) {
-  # .pkg_message("Scraping all shots for match {match_url}. Please acknowledge understat.com as the data source")
 
   match_stats <- .get_understat_json(page_url = match_url) %>%
     rvest::html_nodes("div.scheme-block.is-hide[data-scheme='stats']") %>%
@@ -24,29 +25,29 @@ understat_match_stats <- function(match_url) {
 
   match_stats <- data.frame(
 
-    match_id = gsub("[^0-9]", "", match_url),
+    match_id = as.integer(gsub("[^0-9]", "", match_url)),
 
-    home_team = away[1],
-    home_chances = away[2],
-    home_goals = home[3],
-    home_xG = home[4],
-    home_shots = home[5],
-    home_shot_on_target = home[6],
-    home_deep = home[7],
-    home_PPDA = home[8],
-    home_xPTS = home[9],
+    home_team = as.character(away[1]),
+    home_chances = as.integer(gsub("[^0-9]", "", away[2]))/100,
+    home_goals = as.integer(home[3]),
+    home_xG = as.numeric(home[4]),
+    home_shots = as.integer(home[5]),
+    home_shot_on_target = as.integer(home[6]),
+    home_deep = as.integer(home[7]),
+    home_PPDA = as.numeric(home[8]),
+    home_xPTS = as.numeric(home[9]),
 
-    draw_chances = home[2],
+    draw_chances = as.integer(gsub("[^0-9]", "", home[2]))/100,
 
     away_team = home[1],
-    away_chances = away[3],
-    away_goals = away[4],
-    away_xG = away[5],
-    away_shots = away[6],
-    away_shot_on_target = away[7],
-    away_deep = away[8],
-    away_PPDA = away[9],
-    away_xPTS = away[10]
+    away_chances = as.integer(gsub("[^0-9]", "", away[3]))/100 ,
+    away_goals = as.integer(away[4]),
+    away_xG = as.numeric(away[5]),
+    away_shots = as.integer(away[6]),
+    away_shot_on_target = as.integer(away[7]),
+    away_deep = as.integer(away[8]),
+    away_PPDA = as.numeric(away[9]),
+    away_xPTS = as.numeric(away[10])
 
   )
 

tests/testthat/test-understat.R

@@ -47,3 +47,19 @@ test_that("understat_team_stats_breakdown() works", {
   expect_equal(ncol(team_stats), 11)
   expect_gt(nrow(team_stats), 0)
 })
+
+test_that("understat_match_players() works", {
+  testthat::skip_on_cran()
+  match_players <- understat_match_players(match_url = "https://understat.com/match/14789")
+  expect_true(any("data.frame" == class(match_players)))
+  expect_equal(ncol(match_players), 22)
+  expect_gt(nrow(match_players), 0)
+})
+
+test_that("understat_match_stats() works", {
+  testthat::skip_on_cran()
+  match_stats <- understat_match_stats(match_url = "https://understat.com/match/14789")
+  expect_true(any("data.frame" == class(match_stats)))
+  expect_equal(ncol(match_stats), 20)
+  expect_equal(nrow(match_stats), 1)
+})