Magnes Carlsen playing chess

Magnes Carlsen playing chess, from www.npr.org/alltechconsidered


Overview

Below is the assignment description.

In this project, you’re given a text file with chess tournament results where the information has some structure. Your job is to create an R Markdown file that generates a .CSV file (that could for example be imported into a SQL database) with the following information for all of the players:
* Player’s Name
* Player’s State
* Total Number of Points
* Player’s Pre-Rating, and
* Average Pre Chess Rating of Opponents


Data Discussion

The chess tournament data was provided as a text file which has been saved to my github account.

I can extract the player number, name, state, points won, and pre-tournament ELO score by turning the text file into a table. We can use this to calculate the average pre-tournament elo score of the opponents.


Setup

We are operating in the tidyverse.

# Load packages --------------------------------------
library(tidyverse)


Import & Tidy Data

Here we extract the relevant data from the text file into a dataframe.

Section Code Summary

  • Read data
  • Extract rows into tables
  • Create dataframe and rename columns
  • Tidy ELO ratings
  • Display dataframe
# Load data --------------------------------------
txt_data <- readLines("https://raw.githubusercontent.com/pkofy/DATA607/main/DATA607Project1.txt")
# Extract rows into the two types, ignoring headers, and put them into tables
row1 <- txt_data[seq(5, length(txt_data), 3)]
row2 <- txt_data[seq(6, length(txt_data), 3)]
t1 <- read.table(text = row1, sep = "|")
t2 <- read.table(text = row2, sep = "|")
# Create dataframe and rename columns
ctdf <- data.frame(t1$V1, t1$V2, t2$V1, t1$V3, t2$V2, t1$V4, t1$V5, t1$V6, t1$V7, t1$V8, t1$V9, t1$V10)
ctdf <- ctdf %>% rename(number = t1.V1, name = t1.V2, state = t2.V1, pointswon = t1.V3, elo = t2.V2, g1 = t1.V4, g2 = t1.V5, g3 = t1.V6, g4 = t1.V7, g5 = t1.V8, g6 = t1.V9, g7 = t1.V10)
# Extract elo begin rating
ctdf$elo_begin <- str_extract(ctdf$elo, "(R: ....)")
ctdf$elo_begin <- str_extract(ctdf$elo_begin, "....$")

# Extract elo end rating
ctdf$elo_end <- str_extract(ctdf$elo, "(->....)")
ctdf$elo_end <- str_extract(ctdf$elo_end, "....$")

# Rearrange and Remove elo initial column
ctdf <- ctdf %>% relocate(elo_end, .after = pointswon)
ctdf <- ctdf %>% relocate(elo_begin, .after = pointswon)
ctdf <- subset(ctdf, select = -elo)
# Display initial dataframe
head(ctdf, n=5)

Calculate average elo

Here we calculate the average elo score of the opponents by looking up each opponents elo score at the beginning of the tournament and then taking the average of them.

Section Code Summary

  • Create columns with the opponent numbers for each game
  • Replace opponent numbers with their beginning elo scores
  • Take the row average of these
  • Display required columns
# Create new columns with the opponent numbers
ctdf$o1 <- str_extract(ctdf$g1, "..$")
ctdf$o2 <- str_extract(ctdf$g2, "..$")
ctdf$o3 <- str_extract(ctdf$g3, "..$")
ctdf$o4 <- str_extract(ctdf$g4, "..$")
ctdf$o5 <- str_extract(ctdf$g5, "..$")
ctdf$o6 <- str_extract(ctdf$g6, "..$")
ctdf$o7 <- str_extract(ctdf$g7, "..$")

# Convert the new columns of player numbers from strings to integers
ctdf$o1 <- strtoi(ctdf$o1)
ctdf$o2 <- strtoi(ctdf$o2)
ctdf$o3 <- strtoi(ctdf$o3)
ctdf$o4 <- strtoi(ctdf$o4)
ctdf$o5 <- strtoi(ctdf$o5)
ctdf$o6 <- strtoi(ctdf$o6)
ctdf$o7 <- strtoi(ctdf$o7)
# Replace the player numbers with the player elo_begin scores
ctdf$o1 <- ctdf$elo_begin[ctdf$o1]
ctdf$o2 <- ctdf$elo_begin[ctdf$o2]
ctdf$o3 <- ctdf$elo_begin[ctdf$o3]
ctdf$o4 <- ctdf$elo_begin[ctdf$o4]
ctdf$o5 <- ctdf$elo_begin[ctdf$o5]
ctdf$o6 <- ctdf$elo_begin[ctdf$o6]
ctdf$o7 <- ctdf$elo_begin[ctdf$o7]

# Convert the player elo_begin scores from strings to integers
ctdf$o1 <- strtoi(ctdf$o1)
ctdf$o2 <- strtoi(ctdf$o2)
ctdf$o3 <- strtoi(ctdf$o3)
ctdf$o4 <- strtoi(ctdf$o4)
ctdf$o5 <- strtoi(ctdf$o5)
ctdf$o6 <- strtoi(ctdf$o6)
ctdf$o7 <- strtoi(ctdf$o7)
# Take the row average of these and assign it to a new column
ctdf$begavgofopp <- rowMeans(ctdf[ , c(14:20)], na.rm=TRUE)
# Display the required subset of requested data
ctdf_final <- ctdf[ , c(2,3,4,5,21)]
ctdf_final


Write to CSV file

Here we write the final dataframe to a .csv to be saved to the github folder.

# Writes to csv file
write.csv(ctdf_final, file = "DATA607Project1.csv")


Future Efforts

Piping

I think I can standardize my code better by doing more piping.

Full Range Coding

I probably could have been more sophisticated in my approach. I built up the approach piece by piece so I think I just need to do more projects to become fluent with these techniques.

I was happy with my vectorized approach to looking up the beginning ELO ratings however I could have done a better job of reducing white space and representing numerical values as numbers instead of characters in the initial dataframe setup.


Reference and Source Files

The referenced text and csv files and the R Markdown file for this document are saved here, github.com/pkofy/DATA607, with the name “DATA607Project1”.