Project1

Topic: Chess Tournament Performance Analysis

Approach:

• Created structured player data with round-by-round results (W/L/D) and opponent matchups
• Calculated total points from game outcomes and average opponent ratings
• Generated visualizations to compare player performance against pre-tournament ratings
• Analyzed discrepancies between expected and actual performance based on rating tiers

Step 1: Load Required Libraries

library(dplyr)

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

library(tidyr)
library(stringr)
library(ggplot2)
library(forcats)
library(kableExtra)

## 
## Attaching package: 'kableExtra'

## The following object is masked from 'package:dplyr':
## 
##     group_rows

Step 2: Create Sample Data with Round Results and opponents

players_data <- data.frame(
  Player = c("Gary Hua", "Alice Smith", "Bob Johnson", "Charlie Brown", "Diana Davis"),
  State = c("ON", "CA", "TX", "NY", "FL"),
  PreRating = c(1794, 1650, 1550, 1700, 1600),
  R1_Result = c("W", "L", "D", "W", "L"),
  R1_Opponent = c("Alice Smith", "Gary Hua", "Charlie Brown", "Bob Johnson", "Diana Davis"),
  R2_Result = c("W", "W", "L", "D", "W"),
  R2_Opponent = c("Bob Johnson", "Charlie Brown", "Gary Hua", "Alice Smith", "Diana Davis"),
  R3_Result = c("D", "W", "W", "L", "L"),
  R3_Opponent = c("Charlie Brown", "Diana Davis", "Alice Smith", "Gary Hua", "Bob Johnson"),
  R4_Result = c("W", "D", "L", "W", "D"),
  R4_Opponent = c("Diana Davis", "Bob Johnson", "Alice Smith", "Diana Davis", "Gary Hua"),
  R5_Result = c("W", "W", "D", "L", "W"),
  R5_Opponent = c("Charlie Brown", "Diana Davis", "Bob Johnson", "Alice Smith", "Charlie Brown"),
  OpponentRatings = c(
    "1650, 1550, 1700, 1600, 1700",
    "1794, 1700, 1600, 1550, 1600",
    "1794, 1794, 1650, 1650, 1550",
    "1550, 1650, 1794, 1600, 1650",
    "1600, 1600, 1550, 1794, 1700"
  )
)

# Display the data

players_data %>%
  kable(format = "html", escape = FALSE) %>%  # Ensure HTML format and escape special characters if needed
  kable_styling(
    bootstrap_options = c("bordered", "striped", "condensed"),  # Add borders, striped rows, and compact spacing
    position = "center", # Center the table on the page :cite[1]
    font_size = 12,      # Adjust font size for better readability
  )# %>%

Player	State	PreRating	R1_Result	R1_Opponent	R2_Result	R2_Opponent	R3_Result	R3_Opponent	R4_Result	R4_Opponent	R5_Result	R5_Opponent	OpponentRatings
Gary Hua	ON	1794	W	Alice Smith	W	Bob Johnson	D	Charlie Brown	W	Diana Davis	W	Charlie Brown	1650, 1550, 1700, 1600, 1700
Alice Smith	CA	1650	L	Gary Hua	W	Charlie Brown	W	Diana Davis	D	Bob Johnson	W	Diana Davis	1794, 1700, 1600, 1550, 1600
Bob Johnson	TX	1550	D	Charlie Brown	L	Gary Hua	W	Alice Smith	L	Alice Smith	D	Bob Johnson	1794, 1794, 1650, 1650, 1550
Charlie Brown	NY	1700	W	Bob Johnson	D	Alice Smith	L	Gary Hua	W	Diana Davis	L	Alice Smith	1550, 1650, 1794, 1600, 1650
Diana Davis	FL	1600	L	Diana Davis	W	Diana Davis	L	Bob Johnson	D	Gary Hua	W	Charlie Brown	1600, 1600, 1550, 1794, 1700

 # row_spec(0, bold = TRUE, background = "#f8f9fa")  # Style header row: bold text and light background

Step 3: Calculate Total Points from Round Results:

for each W=win pt=1, for each D=Draw pt=0.5, for each L=Lost pt=0 aded in this list

processed_data <- players_data %>%
  mutate(
    TotalPoints = (
      (R1_Result == "W") + (R2_Result == "W") + (R3_Result == "W") + 
      (R4_Result == "W") + (R5_Result == "W") +
      (R1_Result == "D")*0.5 + (R2_Result == "D")*0.5 + 
      (R3_Result == "D")*0.5 + (R4_Result == "D")*0.5 + 
      (R5_Result == "D")*0.5
    )
  )
# Display the data with calculated points
processed_data %>%
  kable(format = "html", escape = FALSE) %>%  # Ensure HTML format and escape special characters if needed
  kable_styling(
    bootstrap_options = c("bordered", "striped", "condensed"),  # Add borders, striped rows, and compact spacing
    position = "center", # Center the table on the page :cite[1]
    font_size = 12,      # Adjust font size for better readability
  )

Player	State	PreRating	R1_Result	R1_Opponent	R2_Result	R2_Opponent	R3_Result	R3_Opponent	R4_Result	R4_Opponent	R5_Result	R5_Opponent	OpponentRatings	TotalPoints
Gary Hua	ON	1794	W	Alice Smith	W	Bob Johnson	D	Charlie Brown	W	Diana Davis	W	Charlie Brown	1650, 1550, 1700, 1600, 1700	4.5
Alice Smith	CA	1650	L	Gary Hua	W	Charlie Brown	W	Diana Davis	D	Bob Johnson	W	Diana Davis	1794, 1700, 1600, 1550, 1600	3.5
Bob Johnson	TX	1550	D	Charlie Brown	L	Gary Hua	W	Alice Smith	L	Alice Smith	D	Bob Johnson	1794, 1794, 1650, 1650, 1550	2.0
Charlie Brown	NY	1700	W	Bob Johnson	D	Alice Smith	L	Gary Hua	W	Diana Davis	L	Alice Smith	1550, 1650, 1794, 1600, 1650	2.5
Diana Davis	FL	1600	L	Diana Davis	W	Diana Davis	L	Bob Johnson	D	Gary Hua	W	Charlie Brown	1600, 1600, 1550, 1794, 1700	2.5

Step 4: Calculate Average Opponent Rating:

AvgOpponentRating=sum(OpponentRatings)/total_number_of_rounds

processed_data <- processed_data %>%
  rowwise() %>%
  mutate(
    OpponentList = strsplit(OpponentRatings, ","),
    AvgOpponentRating = mean(as.numeric(trimws(OpponentList)), na.rm = TRUE)
  )

# Display the data with average opponent rating
processed_data_display <- processed_data  %>%
select(Player, State, PreRating, TotalPoints, OpponentList, AvgOpponentRating)

processed_data_display %>%
  kable(format = "html", escape = FALSE) %>%  # Ensure HTML format and escape special characters if needed
  kable_styling(
    bootstrap_options = c("bordered", "striped", "condensed"),  # Add borders, striped rows, and compact spacing
    position = "center", # Center the table on the page :cite[1]
    font_size = 12,      # Adjust font size for better readability
  )

Player	State	PreRating	TotalPoints	OpponentList	AvgOpponentRating
Gary Hua	ON	1794	4.5	1650 , 1550, 1700, 1600, 1700	1640.0
Alice Smith	CA	1650	3.5	1794 , 1700, 1600, 1550, 1600	1648.8
Bob Johnson	TX	1550	2.0	1794 , 1794, 1650, 1650, 1550	1687.6
Charlie Brown	NY	1700	2.5	1550 , 1650, 1794, 1600, 1650	1648.8
Diana Davis	FL	1600	2.5	1600 , 1600, 1550, 1794, 1700	1648.8

Step 5: Select Final Columns and Save to CSV

final_data <- processed_data %>%
  select(Player, State, TotalPoints, PreRating, AvgOpponentRating, 
         R1_Result, R1_Opponent, R2_Result, R2_Opponent, 
         R3_Result, R3_Opponent, R4_Result, R4_Opponent, 
         R5_Result, R5_Opponent)

# Save as CSV
write.csv(final_data, "player_stats_with_rounds_and_opponents.csv", row.names = FALSE)

Step 6: Read and Verify the CSV File

# Read the CSV file to verify it was saved correctly
saved_data <- read.csv("player_stats_with_rounds_and_opponents.csv")

Step 7: Create a Summary of Results by Player

player_summary <- saved_data %>%
  select(Player, State, PreRating, TotalPoints, AvgOpponentRating)

# Display player summary
player_summary %>%
  kable(format = "html", escape = FALSE) %>%  # Ensure HTML format and escape special characters if needed
  kable_styling(
    bootstrap_options = c("bordered", "striped", "condensed"),  # Add borders, striped rows, and compact spacing
    position = "center", # Center the table on the page :cite[1]
    font_size = 12,      # Adjust font size for better readability
  )

Player	State	PreRating	TotalPoints	AvgOpponentRating
Gary Hua	ON	1794	4.5	1640.0
Alice Smith	CA	1650	3.5	1648.8
Bob Johnson	TX	1550	2.0	1687.6
Charlie Brown	NY	1700	2.5	1648.8
Diana Davis	FL	1600	2.5	1648.8

Step 8: Create a Detailed Round-by-Round View

# Create a detailed round-by-round view
round_details <- saved_data %>%
  select(Player, starts_with("R")) %>%
  pivot_longer(
    cols = -Player,
    names_to = c("Round", ".value"),
    names_sep = "_"
  ) %>%
  arrange(Player, Round)

# Display round details
round_details %>%
  kable(format = "html", escape = FALSE) %>%  # Ensure HTML format and escape special characters if needed
  kable_styling(
    bootstrap_options = c("bordered", "striped", "condensed"),  # Add borders, striped rows, and compact spacing
    position = "center", # Center the table on the page :cite[1]
    font_size = 12,      # Adjust font size for better readability
  )

Player	Round	Result	Opponent
Alice Smith	R1	L	Gary Hua
Alice Smith	R2	W	Charlie Brown
Alice Smith	R3	W	Diana Davis
Alice Smith	R4	D	Bob Johnson
Alice Smith	R5	W	Diana Davis
Bob Johnson	R1	D	Charlie Brown
Bob Johnson	R2	L	Gary Hua
Bob Johnson	R3	W	Alice Smith
Bob Johnson	R4	L	Alice Smith
Bob Johnson	R5	D	Bob Johnson
Charlie Brown	R1	W	Bob Johnson
Charlie Brown	R2	D	Alice Smith
Charlie Brown	R3	L	Gary Hua
Charlie Brown	R4	W	Diana Davis
Charlie Brown	R5	L	Alice Smith
Diana Davis	R1	L	Diana Davis
Diana Davis	R2	W	Diana Davis
Diana Davis	R3	L	Bob Johnson
Diana Davis	R4	D	Gary Hua
Diana Davis	R5	W	Charlie Brown
Gary Hua	R1	W	Alice Smith
Gary Hua	R2	W	Bob Johnson
Gary Hua	R3	D	Charlie Brown
Gary Hua	R4	W	Diana Davis
Gary Hua	R5	W	Charlie Brown

Plot Players by Total Points (Highest to Lowest)

# Read the data
player_data <- read.csv("player_stats_with_rounds_and_opponents.csv")

# Order players by total points (highest to lowest)
player_data_ordered <- player_data %>%
  arrange(desc(TotalPoints)) %>%
  mutate(Player = fct_reorder(Player, TotalPoints))

# Create the plot
ggplot(player_data_ordered, aes(x = Player, y = TotalPoints, fill = TotalPoints)) +
  geom_bar(stat = "identity") +
  scale_fill_gradient(low = "lightblue", high = "darkblue") +
  labs(title = "Player Performance by Total Points",
       subtitle = "Ordered from highest to lowest points",
       x = "Player",
       y = "Total Points",
       fill = "Points") +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1)) +
  geom_text(aes(label = TotalPoints), vjust = -0.3, size = 3.5)

Plot Comparing Player Pre-Rating vs Average Opponent Rating

# Read the data
player_data <- read.csv("player_stats_with_rounds_and_opponents.csv")

# Prepare data for plotting
rating_comparison <- player_data %>%
  select(Player, PreRating, AvgOpponentRating) %>%
  pivot_longer(cols = c(PreRating, AvgOpponentRating), 
               names_to = "RatingType", 
               values_to = "Rating") %>%
  mutate(RatingType = factor(RatingType, 
                             levels = c("PreRating", "AvgOpponentRating"),
                             labels = c("Player's Rating", "Avg Opponent Rating")))

# Order players by their pre-rating (highest to lowest)
player_order <- player_data %>%
  arrange(desc(PreRating)) %>%
  pull(Player)

rating_comparison <- rating_comparison %>%
  mutate(Player = factor(Player, levels = player_order))

# Create the comparison plot
ggplot(rating_comparison, aes(x = Player, y = Rating, fill = RatingType)) +
  geom_bar(stat = "identity", position = "dodge", alpha = 0.8) +
  scale_fill_manual(values = c("Player's Rating" = "steelblue", 
                               "Avg Opponent Rating" = "darkorange")) +
  labs(title = "Comparison of Player Ratings vs Average Opponent Ratings",
       subtitle = "Ordered by player rating (highest to lowest)",
       x = "Player",
       y = "Rating",
       fill = "Rating Type") +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1),
        legend.position = "top") +
  geom_text(aes(label = Rating, group = RatingType), 
            position = position_dodge(width = 0.9), 
            vjust = -0.5, size = 3)

Rating vs Performance Insights

• Gary Hua: Highest pre-rating (1794) and highest tournament points - demonstrating expected dominance from the top-rated player
• Charlie Brown: Second-highest pre-rating (1700) but finished third in points - underperformed relative to rating expectations
• Alice Smith: Third-highest pre-rating (1650) but achieved second-highest points - exceeded expectations by outperforming higher-rated competitors

Italicized item This analysis reveals that while rating generally predicts performance, notable exceptions occur where players either exceed or fall short of their expected results based on pre-tournament ratings.

Summary of Achievements in the CSV File

We have successfully created a comprehensive chess tournament data analysis CSV file that includes:

Player Performance Metrics

• Player Identification: Name and state for each participant
• Skill Assessment: Pre-tournament rating for each player
• Performance Measurement: Total points calculated from round results

Competition Quality:

• Average rating of opponents faced

Detailed Round-by-Round Analysis

• Results Tracking: Win (W), Loss (L), or Draw (D) for each of the 5 rounds
• Opponent Identification: Specific opponent faced in each round
• Complete Match History: Full record of all games played in the tournament