Introduction

For my second dataset, I selected Pokémon Base Set card grading population data. Pokémon cards have become popular among collectors, and professionally graded cards can vary greatly in rarity. I want to explore how often cards receive a Grade 10 compared with lower grades and whether certain editions have different grading patterns.

Data Source

The dataset comes from the PriceCharting Pokémon Base Set population report.

The dataset contains card names, grading population counts for Grades 6 through 10, and total graded populations. I saved the original data in a wide-format CSV file named pokemon_grading_wide.csv.

Research Question

Which Pokémon Base Set cards have the highest and lowest Grade 10 rates, and how do grading patterns differ between 1st Edition and other card variants?

Data Tidying Approach

The original dataset is in wide format because Grade 6, Grade 7, Grade 8, Grade 9, and Grade 10 are stored as separate columns.

I plan to use pivot_longer() from the tidyr package to convert these grade columns into two variables: Grade and Population.

I will use dplyr to clean card names, convert grading counts into numeric values, identify missing values, and categorize cards by edition where the names provide that information.

I will also check for records that represent sealed products or other items that may not be appropriate for card-level comparisons. The original CSV will remain unchanged.

Planned Analysis

I plan to calculate the Grade 10 rate for each card by dividing the number of Grade 10 copies by the reported total graded population.

I will compare Grade 10 rates across cards and editions. I also plan to investigate whether cards with large grading populations necessarily have higher numbers of Grade 10 copies.

Possible visualizations include bar charts of Grade 10 rates, grade-distribution charts, and scatterplots comparing total grading population with Grade 10 rates.

Expected Challenges

Some grading counts may contain missing values or formatting that needs to be converted into numeric data. The total graded population may include grades not individually displayed in the dataset. I will also need to distinguish between cards, variants, and sealed products.

Expected Outcome

I hope to identify which cards are relatively uncommon in Grade 10 condition and demonstrate why the number of Grade 10 copies alone does not fully describe grading rarity.