DATA607 Project One
Tyler Baker
9/16/2021
Project 1
Import a text file that contains data from a chess tournament. Manipulate and clean the text file so that we can export it as .csv file.
library(tidyverse)## Warning: package 'tidyverse' was built under R version 4.1.1## -- Attaching packages --------------------------------------- tidyverse 1.3.1 --## v ggplot2 3.3.5 v purrr 0.3.4
## v tibble 3.1.4 v dplyr 1.0.7
## v tidyr 1.1.3 v stringr 1.4.0
## v readr 2.0.1 v forcats 0.5.1## Warning: package 'tibble' was built under R version 4.1.1## Warning: package 'tidyr' was built under R version 4.1.1## Warning: package 'readr' was built under R version 4.1.1## Warning: package 'purrr' was built under R version 4.1.1## Warning: package 'stringr' was built under R version 4.1.1## Warning: package 'forcats' was built under R version 4.1.1## -- Conflicts ------------------------------------------ tidyverse_conflicts() --
## x dplyr::filter() masks stats::filter()
## x dplyr::lag() masks stats::lag()url <- "https://raw.githubusercontent.com/tylerbaker01/Data-607-Project-1/main/tournamentinfo.txt"
chess_tournament_data <- read.table(url, sep= "|", header = FALSE, na.strings = c("B", "H", "U", "X"), row.names = NULL, fill = TRUE, strip.white = TRUE)Restructuring the DF
Things to do :strip bad rows, :split dataframe into two seperate dataframes.
#strip bad rows.
new_chess_tournament_data = chess_tournament_data[-c(seq(1, nrow(chess_tournament_data), 3)),]
rownames(new_chess_tournament_data)<- NULL
#split new dataframe into two
split_df<- split(new_chess_tournament_data, rep(1:2, 130))## Warning in split.default(x = seq_len(nrow(x)), f = f, drop = drop, ...): data
## length is not a multiple of split variablechess_player_outcomes<- split_df[[1]]
chess_player_info<- split_df[[2]]#clean dfs
##for chess_player_info :drop vectors [3:11], :name columns, :make rownames NULL, :take the first four digits after R. This is the opponents prematch rating.
#chess_player_info
clean_chess_player_info <- chess_player_info[1:2]
clean_chess_player_info<-clean_chess_player_info[-c(1),]
rownames(clean_chess_player_info)<- NULL
colnames(clean_chess_player_info)<- c("State", "Prematch Rating")
clean_chess_player_info$`Prematch Rating` = substr(clean_chess_player_info$`Prematch Rating`,15,18)#Combine the two dfs again.
- :add chess_player_info into chess_player_outcome
- for columns [] split the data so that one column will be the result of the match and the next column will be the opposing player. :name columns. :make rownames NULL. :create a new column for avg opponent’s prematch rating. Which is equal to sum(opponent’s prematch rating)/the number of matches played.
chess_player_outcomes<-chess_player_outcomes[1:10]
chess_player_outcomes<-chess_player_outcomes[-c(1),]
chess_player_outcomes$state = clean_chess_player_info$State
chess_player_outcomes$prematch_rating=clean_chess_player_info$`Prematch Rating`split certain columns
split columns that contained two separate bits of information.
chess_player_outcomes<- separate(chess_player_outcomes, V4, into=c("outcome 1", "round 1 opp"))
chess_player_outcomes<- separate(chess_player_outcomes, V5, into=c("outcome2", "round 2 opp"))
chess_player_outcomes<- separate(chess_player_outcomes, V6, into=c("outcome3", "round 3 opp"))
chess_player_outcomes<- separate(chess_player_outcomes, V7, into=c("outcome4", "round 4 opp"))
chess_player_outcomes<- separate(chess_player_outcomes, V8, into=c("outcome5", "round 5 opp"))
chess_player_outcomes<- separate(chess_player_outcomes, V9, into=c("outcome6", "round 6 opp"))
chess_player_outcomes<- separate(chess_player_outcomes, V10, into=c("outcome7", "round 7 opp"))#name columns
colnames(chess_player_outcomes)[1] <- "player number"
colnames(chess_player_outcomes)[2] <- "player name"
colnames(chess_player_outcomes)[3] <- "total points"#tidy convert char vectors into numeric vectors. add matches played column.
rownames(chess_player_outcomes)<- NULL
chess_player_outcomes$`player number` <- as.numeric(chess_player_outcomes$`player number`)
chess_player_outcomes$`round 1 opp` <- as.numeric(chess_player_outcomes$`round 1 opp`)
chess_player_outcomes$`round 2 opp` <- as.numeric(chess_player_outcomes$`round 2 opp`)
chess_player_outcomes$`round 3 opp` <- as.numeric(chess_player_outcomes$`round 3 opp`)
chess_player_outcomes$`round 4 opp` <- as.numeric(chess_player_outcomes$`round 4 opp`)
chess_player_outcomes$`round 5 opp` <- as.numeric(chess_player_outcomes$`round 5 opp`)
chess_player_outcomes$`round 6 opp` <- as.numeric(chess_player_outcomes$`round 6 opp`)
chess_player_outcomes$`round 7 opp` <- as.numeric(chess_player_outcomes$`round 7 opp`)
chess_player_outcomes$prematch_rating <- as.numeric(chess_player_outcomes$prematch_rating)
chess_player_outcomes$`matches played` = ((14 - rowSums(is.na(chess_player_outcomes)))/2)
#14 because the NAs will be counted twice, thus the number of possible matches should also be counted twice#handle NAs I created a dummy row that only really carried a player number and the average player rating. The average player rating will be called every time an NA appears in a match.
player_rating_avg <- mean(chess_player_outcomes$prematch_rating)
chess_player_outcomes <- rbind(chess_player_outcomes, c("player number"= 65,"player name"="NA Replacement", "total points"="0.0", "outcome 1"=NA, "round 1 opp"=65, "outcome 2"=NA, "round 2 opp"=65, "outcome 3"=NA, "round 3 opp"=65, "outcome 4"=NA, "round 4 opp"=65, "outcome 5"=NA, "round 5 opp"=65, "outcome 6"=NA, "round 6 opp"=65, "outcome 7"=NA, "round 7 opp"=65, "state"=NA, "prematch_rating"=player_rating_avg, "matches playes"=NA))
#created a new row to add the average prematch player rating. This value will replace all the NAs in a numeric column. which will only be each round's opp.#split DF again
chess_player_outcomes_numeric <- chess_player_outcomes[, c("player number", "round 1 opp", "round 2 opp", "round 3 opp", "round 4 opp", "round 5 opp", "round 6 opp", "round 7 opp", "prematch_rating")]#Tidy
chess_player_outcomes_numeric$`player number` <- as.numeric(chess_player_outcomes_numeric$`player number`)
chess_player_outcomes_numeric$`round 1 opp` <- as.numeric(chess_player_outcomes_numeric$`round 1 opp`)
chess_player_outcomes_numeric$`round 2 opp` <- as.numeric(chess_player_outcomes_numeric$`round 2 opp`)
chess_player_outcomes_numeric$`round 3 opp` <- as.numeric(chess_player_outcomes_numeric$`round 3 opp`)
chess_player_outcomes_numeric$`round 4 opp` <- as.numeric(chess_player_outcomes_numeric$`round 4 opp`)
chess_player_outcomes_numeric$`round 5 opp` <- as.numeric(chess_player_outcomes_numeric$`round 5 opp`)
chess_player_outcomes_numeric$`round 6 opp` <- as.numeric(chess_player_outcomes_numeric$`round 6 opp`)
chess_player_outcomes_numeric$`round 7 opp` <- as.numeric(chess_player_outcomes_numeric$`round 7 opp`)
chess_player_outcomes_numeric$prematch_rating <- as.numeric(chess_player_outcomes_numeric$prematch_rating)#manipulate numerics
#first, I must make NAs equal to 65. This is where the previous stated dummy row exists.
chess_player_outcomes_numeric[is.na(chess_player_outcomes_numeric)]=65
#the following gave me opposing player ratings of each match.
chess_player_outcomes_numeric$opp1rating = chess_player_outcomes_numeric$prematch_rating[order(chess_player_outcomes_numeric$`round 1 opp`)]
chess_player_outcomes_numeric$opp2rating = chess_player_outcomes_numeric$prematch_rating[order(chess_player_outcomes_numeric$`round 2 opp`)]
chess_player_outcomes_numeric$opp3rating = chess_player_outcomes_numeric$prematch_rating[order(chess_player_outcomes_numeric$`round 3 opp`)]
chess_player_outcomes_numeric$opp4rating = chess_player_outcomes_numeric$prematch_rating[order(chess_player_outcomes_numeric$`round 4 opp`)]
chess_player_outcomes_numeric$opp5rating = chess_player_outcomes_numeric$prematch_rating[order(chess_player_outcomes_numeric$`round 5 opp`)]
chess_player_outcomes_numeric$opp6rating = chess_player_outcomes_numeric$prematch_rating[order(chess_player_outcomes_numeric$`round 6 opp`)]
chess_player_outcomes_numeric$opp7rating = chess_player_outcomes_numeric$prematch_rating[order(chess_player_outcomes_numeric$`round 7 opp`)]
chess_player_outcomes_numeric$`Opp Avg` = rowMeans(chess_player_outcomes_numeric[10:16])#merge DFs
chess_tournament_dirty<- cbind(chess_player_outcomes, chess_player_outcomes_numeric)
chess_tournament_dirty<- select(chess_tournament_dirty, "player number", "player name", "state", "total points", "prematch_rating", "matches played", "Opp Avg", "round 1 opp", "outcome 1", "opp1rating", "round 2 opp", "outcome2", "opp2rating", "round 3 opp", "outcome3", "opp3rating", "round 4 opp", "outcome4", "opp4rating", "round 5 opp", "outcome5", "opp5rating", "round 6 opp", "outcome6", "opp6rating", "round 7 opp", "outcome7", "opp7rating")#Clean up DF :I want to remove my dummy row. :remove the columns of the form “oppXrating”. :round “Opp Avg” to the tenth place. :make “total points” into numeric
chess_tournament <- chess_tournament_dirty[-c(65),-c(10,13,16,19,22,25,28)]
chess_tournament$`total points`<- as.numeric(chess_tournament$`total points`)
chess_tournament$`Opp Avg` = round(chess_tournament$`Opp Avg`, 1)#renaming vectors
#renaming vectors
names(chess_tournament)[names(chess_tournament) == "player number"] <- "Player_Number"
names(chess_tournament)[names(chess_tournament) == "player name"] <- "Player_Name"
names(chess_tournament)[names(chess_tournament) == "state"] <- "State"
names(chess_tournament)[names(chess_tournament) == "total points"] <- "Total_Points"
names(chess_tournament)[names(chess_tournament) == "prematch_rating"] <- "Prematch_Rating"
names(chess_tournament)[names(chess_tournament) == "matches played"] <- "Matches_Played"
names(chess_tournament)[names(chess_tournament) == "Opp Avg"] <- "Opp_Avg"
names(chess_tournament)[names(chess_tournament) == "round 1 opp"] <- "Round_1_Opp"
names(chess_tournament)[names(chess_tournament) == "outcome 1"] <- "Outcome_1"
names(chess_tournament)[names(chess_tournament) == "round 2 opp"] <- "Round_2_Opp"
names(chess_tournament)[names(chess_tournament) == "outcome2"] <- "Outcome_2"
names(chess_tournament)[names(chess_tournament) == "round 3 opp"] <- "Round_3_Opp"
names(chess_tournament)[names(chess_tournament) == "outcome3"] <- "Outcome_3"
names(chess_tournament)[names(chess_tournament) == "round 4 opp"] <- "Round_4_Opp"
names(chess_tournament)[names(chess_tournament) == "outcome4"] <- "Outcome_4"
names(chess_tournament)[names(chess_tournament) == "round 5 opp"] <- "Round_5_Opp"
names(chess_tournament)[names(chess_tournament) == "outcome5"] <- "Outcome_5"
names(chess_tournament)[names(chess_tournament) == "round 6 opp"] <- "Round_6_Opp"
names(chess_tournament)[names(chess_tournament) == "outcome6"] <- "Outcome_6"
names(chess_tournament)[names(chess_tournament) == "round 7 opp"] <- "Round_7_Opp"
names(chess_tournament)[names(chess_tournament) == "outcome7"] <- "Outcome_7"#Change back some vectors into numeric
chess_tournament$Player_Number <- as.numeric(chess_tournament$Player_Number)
chess_tournament$Total_Points <- as.numeric(chess_tournament$Total_Points)
chess_tournament$Prematch_Rating <- as.numeric(chess_tournament$Prematch_Rating)
chess_tournament$Matches_Played <- as.numeric(chess_tournament$Matches_Played)
chess_tournament$Opp_Avg <- as.numeric(chess_tournament$Opp_Avg)
chess_tournament$Round_1_Opp <- as.numeric(chess_tournament$Round_1_Opp)
chess_tournament$Round_2_Opp <- as.numeric(chess_tournament$Round_2_Opp)
chess_tournament$Round_3_Opp <- as.numeric(chess_tournament$Round_3_Opp)
chess_tournament$Round_4_Opp <- as.numeric(chess_tournament$Round_4_Opp)
chess_tournament$Round_5_Opp <- as.numeric(chess_tournament$Round_5_Opp)
chess_tournament$Round_6_Opp <- as.numeric(chess_tournament$Round_6_Opp)
chess_tournament$Round_7_Opp <- as.numeric(chess_tournament$Round_7_Opp)#graph I want to see to how the average opponent’s outcome effected the player’s performance. Are there some player’s that still won every match, even though their average opponent’s rating was high?
ggplot(chess_tournament, mapping = aes(x= Total_Points, y= Opp_Avg)) +
geom_point()
So in fact, the data shows that the player’s who had harder competition actually performed better. This surprised me at first, but now my guess is that the tournament is a round robin, thus by the end of it, most of the good players will be at the end to boost up the average opponent player rating.
#write as a csv
write.csv(chess_tournament, file = "chess_tournament.csv")