Vince Ketchum is a famous Pokémon trainer. Recently he has become very curious about the nature of Pokémon and has hired you, a young, budding statistician, to conduct a research project.

Vince has collected data on 802 species of Pokémon and has entered his data into the challenge.csv file (right-click to download to your home desktop). Vince has also noted that there is a more detailed dataset that can be found here.

The dataset contains 802 observations and 12 variables:

• Name: Name of Pokémon
• Type: Type of Pokémon
• Height: Height of Pokémon in metres
• Weight: Weight of Pokémon in kilograms
• Capture Rate
• Happiness: Baseline happiness of Pokémon
• HP: Hit Points of a given Pokémon
• Attack: Base Attack of Pokémon
• Defense: Base Defense of Pokémon
• Speed: Base Speed of Pokémon
• Generation: Generation in which the Pokémon was first introduced
• Is_legendary: Whether the Pokémon is legendary or non-legendary

Vince has formulated some important questions that will help him defeat the notorious Team Rocket:

• What is the strongest Pokémon? Why?
  What makes a Pokémon strong? Do some Pokémon have higher attributes than others?

• Do a Pokémon’s attributes (Attack, Defense, weight, height, Speed, etc.) vary based on its type?
  Perhaps “Dragon” Pokémon are significantly heavier than “Bug” Pokémon. Perhaps “Fire” Pokémon have higher attack than “Water” Pokémon.

• Are legendary Pokémon significantly stronger than non-legendary Pokémon?
  Vince believes “legendary” Pokémon are stronger; does the data support this?

• Emotional Pikachu:
  Vince’s Pikachu has a base happiness of 120. Is this significantly “high” or “low”?

• Is there an association between a Pokémon’s weight and height?
  Do heavier Pokémon tend to be taller?

• Have Pokémon from newer generations become stronger or weaker than those in older generations?
  Is there a generational difference in attributes?

For your report, Vince asks that you explore 2 (or more) of these questions. You are also encouraged to raise questions of your own. Your report may be delivered in any format (PDF, Word document, video, website, etc.). You will be judged on a 20-point scale:

• Mathematical and statistical argumentation (10 points)
• Aesthetic (10 points)

Good luck, and have fun!

Challenge submission

• Submissions are now closed. Check out the winning report here.