DATA 607 Project 1 – Chess Tournament Approach

Author

Patricio Romero

Published

September 24, 2026

Introduction

Chess tournament cross-tables contain information about players, ratings, scores, opponents, and game results in a semi-structured text format. Although the file is readable to a person, it is not immediately suitable for analysis or database storage.

The objective of this project is to use R to read and transform the provided tournamentinfo.txt file into a structured CSV dataset. The final dataset will contain each player’s name, state, total points, pre-tournament rating, and average pre-tournament rating of the opponents faced.

The project will also validate the transformed data and perform a small exploratory data analysis before generating the final chess_tournament_results.csv file.

Data Source

The data were provided by the professor in the file tournamentinfo.txt. The file contains the results of a seven-round chess tournament with 64 players.

Each player is represented by two lines. The first line contains the player’s pairing number, name, total points, and results for each round. The second line contains the player’s state, USCF identification number, pre-tournament rating, post-tournament rating, and color played in each round.

Round results such as W 39, L 8, or D 12 identify a win, loss, or draw and the pairing number of the opponent. Codes without an opponent number, including B, H, U, and X, represent special tournament outcomes and will not be used when calculating the average opponent rating.

Data Dictionary

The final CSV file will contain one row for each player and the following variables:

Variable Description
player_name Full name of the chess player
state Player’s state or province abbreviation
total_points Total tournament points earned by the player
pre_rating Player’s rating before the tournament
average_opponent_pre_rating Average pre-tournament rating of the player’s numbered opponents

The average opponent rating will be calculated by identifying the opponent pairing numbers in the round results, matching those numbers to the corresponding players, retrieving their pre-tournament ratings, and calculating the arithmetic mean.

Planned Approach

The project will follow these steps:

  1. Read all lines from tournamentinfo.txt into R.
  2. Identify the two lines associated with each tournament player.
  3. Extract the pairing number, player name, state, total points, and pre-tournament rating.
  4. Extract the numbered opponents from the seven round-result fields.
  5. Exclude special tournament codes that do not identify an opponent.
  6. Match each opponent pairing number to the corresponding player’s pre-tournament rating.
  7. Calculate and round the average opponent pre-rating for every player.
  8. Combine the calculated values into a tidy dataset with one row per player.
  9. Validate row counts, column types, missing values, duplicate players, and rating ranges.
  10. Confirm that Gary Hua’s average opponent pre-rating is 1605, as shown in the assignment example.
  11. Perform a small exploratory data analysis of player ratings and opponent strength.
  12. Export the final dataset as chess_tournament_results.csv.

The original text file will remain unchanged. All cleaning, transformation, calculation, validation, visualization, and CSV creation will be performed reproducibly in the Quarto document.

Data Import and Initial Inspection

The required packages are loaded and the original tournament text file is read into R. Each line is initially retained as text so that the structure can be inspected before extracting individual variables.

library(readr)
library(dplyr)
library(stringr)
library(ggplot2)
library(skimr)
library(knitr)

raw_lines <- read_lines(
  "tournamentinfo.txt",
  progress = FALSE
)

length(raw_lines)
[1] 196
head(raw_lines, 8)
[1] "-----------------------------------------------------------------------------------------" 
[2] " Pair | Player Name                     |Total|Round|Round|Round|Round|Round|Round|Round| "
[3] " Num  | USCF ID / Rtg (Pre->Post)       | Pts |  1  |  2  |  3  |  4  |  5  |  6  |  7  | "
[4] "-----------------------------------------------------------------------------------------" 
[5] "    1 | GARY HUA                        |6.0  |W  39|W  21|W  18|W  14|W   7|D  12|D   4|" 
[6] "   ON | 15445895 / R: 1794   ->1817     |N:2  |W    |B    |W    |B    |W    |B    |W    |" 
[7] "-----------------------------------------------------------------------------------------" 
[8] "    2 | DAKSHESH DARURI                 |6.0  |W  63|W  58|L   4|W  17|W  16|W  20|W   7|"

Identifying Player Records

Player records are identified by lines that begin with a pairing number followed by a vertical separator. These lines contain the player’s name, total points, and round results. The following line contains the player’s state and rating information.

player_line_indices <- which(
  str_detect(raw_lines, "^\\s*[0-9]+\\s*\\|")
)

number_of_players <- length(player_line_indices)

number_of_players
[1] 64
head(player_line_indices)
[1]  5  8 11 14 17 20

Preliminary Data Inspection

The preliminary inspection confirms that the source file can be read successfully and that the expected number of player records can be identified. No tournament calculations or final transformations are performed at this stage.

inspection_summary <- data.frame(
  measure = c(
    "Raw text lines",
    "Identified player records"
  ),
  value = c(
    length(raw_lines),
    number_of_players
  )
)

knitr::kable(
  inspection_summary,
  col.names = c("Measure", "Value"),
  caption = "Preliminary inspection of the tournament file"
)
Preliminary inspection of the tournament file
Measure Value
Raw text lines 196
Identified player records 64

The source file contains 196 text lines, from which 64 player records were identified. These results are consistent with the expected structure of the tournament cross-table and support the planned transformation.

Validation Plan

The transformed dataset will be validated before it is exported. The following checks will be performed:

  1. Confirm that the final dataset contains 64 player records.
  2. Confirm that the final CSV contains the five variables required by the assignment.
  3. Verify that each pairing number identifies only one player.
  4. Check for missing player names, states, total points, and pre-tournament ratings.
  5. Confirm that total points and ratings are stored as numeric variables.
  6. Verify that numbered opponents match valid pairing numbers.
  7. Exclude tournament codes without opponent numbers from the opponent-rating calculation.
  8. Check that each calculated opponent average is based only on games with an identified opponent.
  9. Confirm that Gary Hua’s opponent rating average rounds to 1605.
  10. Verify that the generated CSV can be read back into R without changing its row count or column structure.

These checks will help detect parsing errors, invalid matches, missing values, and inconsistencies before the final dataset is published.

Expected Output

The Quarto code will generate a file named:

chess_tournament_results.csv

The CSV will contain 64 player records and the following five columns:

  • player_name
  • state
  • total_points
  • pre_rating
  • average_opponent_pre_rating

The expected first record is:

player_name state total_points pre_rating average_opponent_pre_rating
Gary Hua ON 6.0 1794 1605

The final CSV will be stored in the project folder with the original tournamentinfo.txt file and the reproducible Quarto document. This structure will allow another person to run the code and reproduce the transformed dataset.

Planned Deliverables

The completed project will include the original tournamentinfo.txt file, the reproducible Quarto source file, the generated chess_tournament_results.csv file, and the rendered HTML report. The code and data will be available through GitHub, and the rendered report will be published online for review.

AI Use

ChatGPT was used to help interpret the assignment requirements, organize the planned approach, improve the English writing, explain the tournament cross-table structure, and provide coding guidance. I will run the R code, review the transformed data, validate the results, and confirm the final conclusions myself.