IP.R

library( package = 'dplyr' )
library( package = 'data.table' )

# Create vectors that will store column classes for the game and event files
l_e_cols <- c( 'character'                   # GAME_ID
             , rep( x = 'NULL', times = 13 )
             , 'character'                   # PIT_ID
             , rep( x = 'NULL', times = 25 )
             , 'numeric'                     # EVENT_OUTS_CT
             , rep( x = 'NULL', times = 58 )
             , 'character'                   # FLD_TEAM_ID
             , rep( x = 'NULL', times = 58 )
             )

l_g_cols <- c( rep( x = 'NULL', times = 8 )             
             , 'character'                  # HOME_TEAM_ID  
             , rep( x = 'NULL', times = 76 )             
             , 'character'                  # HOME_TEAM_LEAGUE_ID 
             , rep( x = 'NULL', times = 93 )                
             )

# Create vectors that will store column names for the game and event files
l_e_names <- c( 'GAME_ID', 'PIT_ID', 'EVENT_OUTS', 'TEAM_ID' )
l_g_names <- c( 'TEAM_ID','LEAGUE_ID' ) 

# Load the 1990 season event file into the environment
d_e_1990 <- fread( input      = 'all1990.csv'
                 , sep        = ','
                 , header     = T
                 , colClasses = l_e_cols
                 , col.names  = l_e_names
                 )

# Load the 1990 season game file into the environment
d_g_1990 <- fread( input      = 'games1990.csv'
                 , sep        = ','
                 , header     = T
                 , colClasses = l_g_cols
                 , col.names  = l_g_names
                 )

# Game dataset ( d_g_1990 ) has a lot of duplicate records, so get unique observations
d_g_1990 <- distinct( .data = d_g_1990, TEAM_ID, LEAGUE_ID )

# Associate Game dataset and Event dataset
d_1990 <- inner_join( x = d_e_1990, y = d_g_1990, by = c('TEAM_ID') )

# Get IP by every pitcher for every game played in the 1990 season.
d_g_ip <- group_by( .data = d_1990, GAME_ID, PIT_ID )       %>%
         summarise( O  = sum( x = EVENT_OUTS, na.rm = T ) ) %>%
            mutate( IP = O %/%3 + O %% 0.3 )

# Get IP by every pitcher in the 1990 season.
d_s_ip <- group_by( .data = d_1990, PIT_ID )                 %>%
         summarise( O  = sum( x = EVENT_OUTS, na.rm = T ) )  %>%
            mutate( IP = O %/%3 + O %% 0.3 )

# Get IP by every team in the 1990 season.
d_t_ip <- group_by( .data = d_1990, TEAM_ID, LEAGUE_ID )     %>%
         summarise( O  = sum( x = EVENT_OUTS, na.rm = T ) )  %>%
            mutate( IP = O %/%3 + O %% 0.3 )

# Get IP by the AL and NL leagues in the 1990 season.
d_l_ip <- group_by( .data = d_1990, LEAGUE_ID )             %>%
         summarise( O  = sum( x = EVENT_OUTS, na.rm = T ) ) %>%
            mutate( IP = O %/%3 + O %% 0.3 )

# Get the total of IP in the 1990 season
d_mlb_ip <- mutate( .data = d_1990, YEAR = 1990 )           %>%
          group_by( YEAR )                                  %>%
         summarise( O  = sum( x = EVENT_OUTS, na.rm = T ) ) %>%
            mutate( IP = O %/%3 + O %% 0.3 )