Pokeapi Vignette Reflection
Samantha Hunter 10/7/2021
PokeAPI Vignette
My vignette allows someone to query the PokeAPI and get various information about different elements of the turn-style video games. I created six different functions to get information about the berries, Pokemon stats, Pokemon types, Pokemon abilities, and Pokemon species from each generation of the game. Game generations are usually released with a new set of Pokemon species and new game elements so we can use the Pokemon species or Pokemon generation as key variables to join the data frames that are returned from the functions.
One of the functions is to get information about the berries that were introduced in the Generation II games, but the berry information itself seems to be derived from Generation III games. In Generation III, a player was able to plant and harvest berries, so my function will return the germination period and how many berries can grow on a single plant.
The other five functions exclusively return data about each Pokemon species and sub-species of the Pokemon from the endpoints. I queried the Generation endpoint, which will tell you when the Pokemon species was first introduced. I queried Habitat, which will tell you where a Pokemon can be found, but this element of the games only makes an appearance in Generation III (see EDA). Pokemon Type is another endpoint that will return all the Pokemon of the type you specified. Ability is an endpoint that you can query the name of the desired ability and will return all the Pokemon that can have that ability. Finally, I queried the Pokemon endpoint and you will get out the base stats for that wild-caught Pokemon.
Using these endpoints, I explored the data quite a bit before deciding to investigate whether or not Pokemon became stronger with each generation. I know most of my favorite Pokemon come from the first few generations of the game and I wondered how they would fare against their newer counterparts. I decided the easiest way to do this was to sum all of the base stats and compare those. I didn’t end up finding any evidence that there was a consistent power creep along generational lines, although I could see an argument that the developers of Pokemon did make Pokemon stronger up to Generation IV before scaling power back. There also aren’t the same amount of Pokemon introduced in each Generation, which can obscure and skew the results. I did find that there were roughly the same amount of very strong Pokemon in each generation, but some of the generations introduced many Pokemon (Generation I), while others introduced less than that (Generation VI).
Because there was no evidence of a power creep. I also looked at the overall population of each generation and got a feel for the rate of occurrence of where Pokemon could be caught and what type they were. I also created scatterplots for the three stats that I care about the most when I play Pokemon - HP, Attack, and Defense. The way I play Pokemon is that I try to leverage the Pokemon type strengths and weaknesses so I don’t have to worry about Special Attack, Special Defense, and Speed. I should hold the upper hand in Special Attack and Special Defense as long as I choose my Pokemon wisely. I wanted to see if there was any relationship between HP, Attack, and Defense. I thought that these might be inversely related, but I surprisingly found that these three were generally positively correlated. It seems that Pokemon tend to be well-rounded in regards to these three stats.
Reflecting on Vignette Creation
I thought this project was pretty hard, but I am happy that I was able to accomplish what I did. I spent nearly all of my time on the getting the functions to produce a nice tibble, although I did I scale back from my original plan. For example, I originally wanted to make a table of Pokemon-types and what they were weak against and strong against, but because a Pokemon-type may be weak or strong against more than one other type and about a third of Pokemon have more than one type themselves, unlisting that into a tibble was sort of a nightmare. I definitely had trouble knowing when I needed to use ‘[’ and ‘[[’ for unlisting or listing an output as well. I know next time I will write out what I need to be a list vs an element and that will save me a ton of time trouble shooting. I didn’t know how to navigate that some lists would be empty and some would have two types so I just abandoned that idea as well, but this is something I think I can take care of at a later date with a little more time. I do want to come back to this project when I have the time to figure that out. I also had trouble with getting my graphs to print in the GitHub pages due to specifying an output, but I like to be clear.
I also want to come back and play around with the formatting and analyzations of the project. I may try running actual mathematical statistic analyses to further explore whether Pokemon from different generations have significantly different stast. For now, I am just happy to draw some tentative conclusions from graphas.