In this section, we present a series of visualizations created during the process of mapping Pokémon distributions to real-world biomes. These visual representations highlight key findings from our analysis and illustrate the development process of the interactive map.
Ridge Plot Analysis of Pokémon Stats and Types
This analysis presents a ridge plot to observe the distribution of Pokémon based on their stats and types.
The findings highlight interesting patterns in how Pokémon evolve and their relationship to stat totals.
Figure 1. Distribution of Pokémon stats by type. A higher peak represents more Pokémon of that type at that stat total.
By generating this ridge plot, we are able to observe the distribution of Pokémon according to their stats and types. The plot reveals two significant insights: Firstly, we notice several peaks for each Pokémon type, which reflects the evolution path of Pokémon—more evolved Pokémon tend to have higher stat totals. Secondly, the dragon type stands out as having the highest average stat, which aligns with the well-known rarity and strength of dragon-type Pokémon.
Key Findings
The ridge plot provides valuable insights into how the evolution of Pokémon correlates with their stats.
Pokémon types with higher evolutionary stages tend to have more pronounced peaks in their stat distributions,
reinforcing the link between evolution and stat increase. Additionally, the dominance of the dragon type in terms of average stat totals is evident. This type's rarity and powerful abilities make it stand out among others, furthering the understanding of its competitive edge.
Stat Distributions for Fire, Water, and Ice Pokémon
This interactive candle plot presents the maximum, minimum, and median stat values for each of the Pokémon types we focused on:
fire, water, and ice. By exploring these distributions, we gain insights into the range and central tendencies of their stats.
Figure 1. Candle plot showing the distribution of stats (max, min, and median) for the Pokémon types: Fire, Water, and Ice.
The candle plot clearly illustrates the differences in the statistical distributions between the Pokémon types.
Ice-type Pokémon, for instance, show the lowest maximum stat, indicating that their legendary Pokémon tend to have lower stat totals compared to the legendary Pokémon of Fire and Water types.
Key Findings
From this interactive candle plot, we observe significant variations in the max, min, and median stat totals for Fire, Water, and Ice types.
This reinforces the decision to separate Pokémon types before ranking them. The plot makes it clear that Ice-type Pokémon have the lowest maximum stats,
suggesting that legendary Pokémon of the Ice type are not as statistically powerful as those of the Fire and Water types.
Figure 1. Bar plot showing the distribution of primary Pokémon types.
We can see that the most common Primary Type for most Pokémon is water with 134 Pokémon. On the other hand, the least common Primary type for Pokémon is flying with 9 Pokémon. The rest of the Pokémon types are quite evenly distributed. While this may seem to imply that flying Pokémon are the rarest, this is not the case, as can be observed in the next plot below.
Figure 3. Bar plot showing the distribution of secondary Pokémon types.
For the secondary types, we can see that the flying type is the most common by far. This is not surprising, as many Pokémon are capable of flying but still retain their primary elemental type. This distribution suggests that the flying type is a common secondary feature across various Pokémon species.
Figure 4. Heat plot showing the distribution of secondary Pokémon types.
This heat plot visually emphasizes the prevalence of flying as a secondary type across the Pokémon dataset, supporting the bar plot’s findings.
What is the Most Common Colour of Pokémon?
Now that there are Pokémon in the real world, you probably want to select which one is your favourite! This diagram shows how rare certain colours of Pokémon are.
This plot was used to portray the proportion of colours of Pokémon. With each square representing 1% of the Pokémon species, we can observe that the most common colour is red, while green and purple are the least common colours, each sitting at around 3%. This is great news for people who like the colour red!
Note: The process of isolating the dominant colour of Pokémon and sorting them into colour families may not result in 100% accuracy in this chart.
Key Findings
This diagram reveals the distribution of Pokémon colours, highlighting that red is the most common colour among Pokémon, while green and purple are the least common, each accounting for about 3% of the total Pokémon species.
While the process of determining the dominant colours of Pokémon may not be 100% accurate, this plot offers a clear visual representation of colour proportions across the Pokémon species.
Effect of Latitude and Longitude on Temperature
This study examines how geographical coordinates—latitude and longitude—impact temperature variations.
We analyzed three key relationships to determine which factor plays a more dominant role in shaping temperature patterns.
Figure 1. Relationship between latitude and temperature. The trend shows that temperatures decrease as latitude increases.
The graph shows a strong negative correlation between latitude and temperature.
This supports the well-known geographical principle: temperatures are highest near the equator and decrease as we move toward the poles.
Figure 2. Relationship between longitude and temperature. Results indicate minimal influence of longitude on temperature.
Unlike latitude, longitude does not show a clear relationship with temperature. The data points are scattered, indicating
no significant correlation. Other factors such as elevation and ocean currents likely play a greater role.
Key Findings
These findings align with fundamental climatological principles, where latitude influences solar radiation
exposure and, consequently, temperature patterns. Regions closer to the equator receive more direct sunlight,
resulting in higher temperatures, while polar regions experience lower temperatures due to indirect solar rays.
Overall, this analysis reinforces that latitude is the dominant geographic factor in determining
temperature variation, while longitude plays no meaningful role in temperature prediction.
Regression Analysis: Latitude as a Primary Determinant of Temperature
The R-squared value of 0.781 indicates that 78.1% of the variation in temperature can be explained by latitude alone. This supports the well-known geographic principle that temperatures decrease as one moves away from the equator.
Coefficient:-0.7931 (As latitude increases, temperature decreases.)
Statistical Significance: p-value < 0.001 (The relationship is statistically significant.)
In contrast, longitude had an R-squared value of 0, meaning it does not explain any variation in temperature.
The regression results showed:
Coefficient:0.0113 (This suggests no meaningful relationship between longitude and temperature.)
Statistical Significance: p-value 0.874 (This indicates that longitude has no statistical significance in determining temperature.)
Temperature Variance: Hottest vs Coldest Places
This section explores the variance in temperature between the hottest and coldest places. The analysis reveals that the hottest places exhibit less variance in temperature compared to the coldest places.
Figure 1. Comparison of temperature spread in the hottest and coldest places.
The data shows that the hottest places have a more consistent temperature range, with less fluctuation, maintaining relatively stable high temperatures throughout the year.
On the other hand, the coldest places exhibit greater variability in temperature, influenced by seasonal changes and geographical factors.
Development of the Interactive Folium Map
To effectively visualize the spatial distribution of Pokémon across different biomes, we employed Folium, a Python library designed for interactive geographic mapping. The objective of this step was to create a dynamic and user-friendly tool that integrates ecological data with real-world geographical coordinates. Through an iterative design process, we tested multiple drafts before finalizing the map.
Preliminary Drafts and Iterative Refinement
During the initial development phase, three versions of the map were created to evaluate different visualization techniques and optimize the representation of Pokémon distributions. Each iteration provided insights into improving the accuracy and usability of the final model.
Figure 3: Final test version before optimization, featuring enhanced interactivity and legend customization.
Final Map
Feedback from these preliminary drafts led to significant refinements in the final version. Key improvements included better handling of overlapping Pokémon, an improved search function, and real-time weather updates.
Figure 3: Final test version before optimization, featuring enhanced interactivity and legend customization.
Feedback and Peer Review
We encourage constructive feedback to refine our visualizations and improve the interactive mapping experience. Your insights will contribute to the accuracy and usability of this research. Please share your observations below:
Public Comments