Welcome!

With the anticipation of Monster Hunter Rise, I wanted to take a look at the previous games titles in the Monster Hunter series. Monster Hunter is a series of video games made by Capcom that you play as a hunter that hunts various different and unique monsters. Large monsters act as "bosses" so the more monsters in a title means the more potential playtime you can have. Each new installment brings new monsters, new ways to hunt the monsters and also brings older monsters from previous games. When a base version of a title is released, it is followed by a sequel (known as a ultimate expansion/version of the base game). Ultimate expansions brings more monsters and therefore more playtime!

The reason for this notebook is to practice using pandas and data visualization tools (matplotlib) to produce some neat figures for the Monster Hunter community to enjoy.

I plan on looking into:

The amount of monsters that are released in each game

New monsters are always the most exciting part of a new release of a Monster Hunter title. Seeing the amount of monsters throughout the series would also be interesting! Why not both? I plan on looking at the distribution of the total monsters, as well as new monsters in each title. Maybe we can see if we are getting more monsters as the series goes on!

Background infomation:

* Japanese Title: is the japanese release of the title. japanese title were released earlier than their international counterparts until recent.
* International Title: is the international title for a japanese release, earlier international titles in the series often had different title names to thier japanese counterpart.
* Large Monsters: Large monsters are considered boss monsters, the goal of the game is to hunt these monsters.
* Small Monsters: Small monsters are supplementary monsters to add life to the game.
* Variant Monsters: Variant monsters are altered version of large monsters. they are identified as a variant if they have a prefix on their name (ie: Azure Rathalos is a variant to the Rathalos).
* New Monsters: Monsters that are introduced in that title.

Monster types throughout the mainline series

There are many different types of monsters in the series. There are dragons and dinosaurs to mammoths and even large crabs. I want to see what monster types are favored in the series by creating a figure to see the amount of different monster types in the series.

Background infomation:

* Monster Types: every monster in Monster Hunter has a type (like how animal's are classified).

The occurrences of each monster throughout the mainline series

Which monster is the most popular? Popularity will not be judged like it was in highschool rather which monsters has appeared in the most titles (Hopefully no feelings are hurt).

Background infomation:

* Monster Types: see above.

Time between releases of base and ultimate versions and the increase of monsters in the ultimate versions

What is the worst part about a base titles release? waiting for the ultimate release of course. I will see what the average time between base and ultimate versions are and I will also look into how much more monsters are added to the ultimate version.

Background infomation:

* Base Version: is a title that is new to the series that brings a new story Monster Hunter.
* Ultimate Version: a title that adds onto a base version that adds to the story, adds new monsters and a new difficulty.

Monsters that each director introduces, do they favor their monsters when releasing a game?

We all love our children (I would if I had one I think) but do directors of Monster Hunter feel the same towards the monster they have introduced? I was curious if they did, so we will see!

Background infomation:

* Director: A Director is a person that is in charge of creating a title.

Table of Contents

  1. Methods
  2. Analysis
  3. Conclusion

Methods

Note The explaination of my process is generalized because I do not expect everyone to understand python. I wanted to inform you on the direction I am going through to process the data in order to analyze it. if you want to see the outcomes of the processing (The Figures!) skip to the Analysis.

Cleaning

There are 3 csv files:

  1. Monster Classes (Describes Monster Type and Size) Monster_Classes.csv
  2. Title Data (Describes Title information including release data, director) MonsterHunter_General_Data.csv
  3. Monster in Title (lists all monsters in a title) Monsters_in_Games.csv

I prepared the .csv files to includes data such as monsters in each games, specific monster data, and title information.

In order to join the 3 tables together, a simular key is needed to joined the tables. Monsters in Title Table is melted to make a 2 column table with columns ['Name','Title'] so it can be join with Monster Classes table on 'Name'. The combined table is joined with Title Data on 'Title' to create Monster Hunter Data table.

Most of the data was collected referring to the website MonsterHunter.Fandom

Amount of Monsters: Processing the Data

The analysis will be done on the following titles:

Note: Early in the series, the japanese titles would be released months before the international release, the international releases would also have the title name changed (ex: (Jpn) Monster Hunter X, (International) Monster Hunter Generations )

Note I am using the monsters from Monster Hunter: World and Monster Hunter World: Iceborne last updates and Version 3.0 of Monster Hunter Rise

I am focusing on the international named titles, if it has one, otherwise I will be looking at the japanese named title.

For each title I obtain:

  1. Amount of total monsters
  2. Amount of large monsters
  3. Amount of small monsters
  4. Amount of new monsters
  5. Amount of new large monsters
  6. Amount of new small monsters
  7. Amount of Variant/Subspecies/Deviants

For all the titles:

title_data = monster_hunter_data[monster_hunter_data['Title'] == title]

To find [ 1 ] I filtered for the title name, removed any duplicate monster names, and counted the rows.

title_monsters = title_data.drop_duplicates(subset=['Name']).count()

Similary to [ 1 ] to find [ 2 / 3 ] I filter by title name, duplicate monster names and groupby the each size and count the rows.

large_monsters = total_monsters.groupby('Size').count().Large
small_monsters = total_monsters.groupby('Size').count().Small

In order to find [ 4 ], new monsters in the title needs to be found, the title date is used to filter for previous titles. The minimum title date is taken as filter condition.

title_date = min(title_data['Date Released'])

A list of monsters from the title and a list of monsters from all previous titles are found. The new monsters are found by taking the set difference of the previous monsters and the title monsters (Removing any previous monsters from the title monster list).

previous_monsters = title_data[title_data['Date Released'] < title_date].drop_duplicates(subset=['Name'])

new_monsters = title_monsters[title_monsters['Name'].isin(previous_monsters) == False].count()

The [ 5 / 6 ] is just a groupby on 'Size' using the data from [ 4 ].

new_large_monsters = new_monsters.groupby('Size').count().Large 
new_small_monsters = new_monsters.groupby('Size').count().Small

[ 7 ] The variants (which are generalizations of subspecies/deviants/apexs(Rise)) are retrieved finding the non-variant monsters by sorting a list of all the monsters names by string length (in ascending order) and then iterating through the list from 0 to n-1 removing any names that contains the shorter names which produces a list of non-variant monsters. To get the variants we take the difference of titles monsters and the non-variant title monsters.

variant_count = monsters_in_title[monsters_in_title['Size'] == 'Large']['Name'].count() - len(
            helpers.filter_out_variants(monsters_in_title[monsters_in_title['Size'] == 'Large']))

see the filter_out_variants(df) if interested.

Monster Types: Processing the data

To get the monsters throughout the titles, duplicate monster names are dropped from monster_hunter_data and the rows counted of each type.

monster_type_data = monster_hunter_data.drop_duplicates(subset=['Name']).groupby('Type').count()

To find out when each type was introduced, I sorted the date released by ascending order and dropped duplicate release dates (keeping the first date) to make monster_type_intro_date.

monster_type_intro_date = monster_hunter_data.sort_values(by='Date Released').drop_duplicates(
                                                                               subset=['Type'],keep='first')

Occurrence of Monsters: Processing the Data

I will be using analysis_titles to make:

monster_occurrence_data = monster_hunter_data[monster_hunter_data['Title'].isin(analysis_titles) == True]

A list of occurrence of each monster and their type is produced.

monster_occurrence_data = monster_occurrence_data.groupby(['Name', 'Type'])['Title'].count()

monster_occurrence_data has 3 columns 'Name', 'Type','Occurrence', where occurrence is the amount of times it is in the analysis titles. To find the monster and types that has been in most titles, I get the max occurrences by grouping by 'Type' and using the max aggregate.

max_occ_monster_type = monster_occurance_data.groupby('Type')['Title'].max()

joining max_occ_monster_type and monster_occurrence_data on 'Type' to produce most_occurrences. The join adds the column 'Max Occurrence' that contains the highest number of occurrence of the given type of monster. To create the final table I filter the table where occurrence is equal to the max occurrence of the monster type.

most_occurrences = most_occurrences[most_occurrences['Occurrence'] == most_occurrences['Max_Occurrence']]

Base vs. Ultimate: Processing the data

For the analysis, I want to see how much time does it take between the base and ultimate versions and how much more monsters are added to the ultimate release.

I chose to use the japanese titles release dates because they have the earliest release dates.

note: Monster Hunter:Portable 3g and Monster Hunter:Rise are excluded because they do not have ultimate releases.

The game_title_data table is made by filtering for game_titles, titles are sorted by 'Release date' and any duplicate titles are dropped.

2 more tables are created by joining the respective column from game_titles to game_title_data to create base_titles and ultimate_titles

base_titles = pd.merge(game_title_data, game_titles, left_on='Title', right_on="Base", how='right').dropna()[
        ['Title', 'Date Released']]

ultimate_titles = pd.merge(game_title_data, game_titles, left_on='Title', right_on="Ultimate", how='right').dropna()[
        ['Title', 'Date Released']]

A 'Large Monster' column is added to both base_titles and ultimate_titles by applying a function to get the large monster in the titles.

lambda title: len( mh_data[mh_data['Title'] == title) & (mh_data['Size'] == 'Large')]['Name'].unique() )

The tables base_titles, ultimate_titles are joined on index of each table that relates base game to ultimate version to create bu_game_date_data. The columns 'Date Difference' and 'Large Monster Difference' are added by computing the difference of the ultimate date and large monsters by base date and large monsters.

bu_game_date_data['Date Difference'] = abs(bu_game_date_data['Date Released Ultimate'] - 
                                        bu_game_date_data['Date Released Base'])

bu_game_date_data['Monster Difference'] = abs(bu_game_date_data['Large Monsters Ultimate'] - 
                                           bu_game_date_data['Large Monsters Base'])

Directors of Monster Hunter: Processing the data

In order to find each director's monsters I sorted monster_hunter_data by 'Date Released' and dropped duplicate monster names (keeping the first occurrence) creating director_data.

director_data = monster_hunter_data.sort_values(by=['Date Released']).drop_duplicates(subset=['Name'], keep='first')

Filtering for each director, Kaname Fujioka, Yasunori Ichinose, and Yuya Tokuda.

fujioka_titles = director_data[director_data['Director'] == 'Kaname Fujioka'][
        ['Title', 'Date Released', 'Director']].drop_duplicates(subset=['Title'],keep='first')

I create a list of the directors monsters by using director_data and filtering for the director and projecting ['Name','Title']

fujioka_monsters = director_data[director_data['Director'] == 'Kaname Fujioka'][['Name', 'Title']]

I add 2 additional columns to fujioka_titles, ['Total Monsters','Director Monsters'] by apply 2 functions to the dataframe

# For Total Monsters
lambda title: monster_hunter_data[monster_hunter_data['Title'] == title]['Name'].drop_duplicates().count())

# For Director Monsters
lambda title: monster_hunter_data[(monster_hunter_data['Title'] == title) &
 (monster_hunter_data['Name'].isin(fujioka_monsters['Name']) == True)]['Name'].drop_duplicates().count())

I combined the fujioka_titles, ichinose_titles, tokuda_titles to create director_title_data and an additional dataframe is made to have some additional information that has columns ['Director','Titles','Total Director Monsters'].

Analysis

Amount of Monsters: Analysis

Using amt_monsters_df produces 'Amount of Monster in Mainline Series' figure.

Some outlier titles when it comes to new monsters are Monster Hunter, Monster Hunter Generations Ultimate, and Monster Hunter 3

Monster Hunter is the first title in the series so the 1.0 new monster ratio is not alarming.

Monster Hunter Generations Ultimate is an anniversary title and the Monster Hunter team's goal was to bring as many previous monsters in the franchise, so the low new monster ratio is valid.

Monster Hunter 3 has a high new monster ratio at 0.84 which is interesting. This title is the only base title with underwater combat. Maybe since there was such a new and unique way to play this title they wanted to add many new monsters.

Overall I think we cannot answer any questions such as Are they adding less monster over time? because the way each title gets made most likely differs in the amount of time for development to what monsters would fit the theme of the title.

Monster Types: Analysis

Monster Types that are the most popular are Flying Wyvern (47), Elder Dragon (43) and Bird Wyvern (37).

The monster types in the top half are monster types that were released in the first title Monster Hunter with the exception of Neopteron (11), Lynian (7) and Herbivore (13). Lynian and Herbivore are small monsters that are included in each title to add life and are usually reintroduced therefore the amount is fair. Neopteron are also small monsters but they do include some large monsters (Seltas, Seltas Queen, etc.).

The monster type Snake Wyvern (3) has a low amount of monsters that belong to its type.

This means that the last time a new Snake Wyvern was released was at the end of 2014! This got me curious, so I looked for what are the monster type(s) that have the longest time since a new monster of that type has been released.

Snake Wyvern and the Unknown types both have not had new monsters since the end of 2014!

Occurrence of Monsters: Analysis

The monsters with the most occurrences throughout the series is Rathalos, Rathian and Kelbi which have shown up in all analysis titles.

Rathalos Rathian Kelbi
Rathalo Rathian Kelbi

Images from MonsterHunter.fandom

The Kelbi being in every game is necessary because there is a helpful item that you use in the game that uses material that come it.

It is interesting that both the Rathian and Rathalos are the most occurring large monsters. They are the male and female counterpart to each other. If one is featured in a title, the other one will surely be appearing in that title too! I suppose you can say that both the Rathian and Rathalos are Monster Hunter's iconic monsters. They are dragons, what's not to like about them!

Base vs. Ultimate: Analysis

Let us use the mean time between games because with Covid, the time of game devolpment has been affected for many game studios. Monster Hunter Rise was released on March 26/2021 so maybe Monster Hunter Rise Ultimate will be released in mid July 2022 (sooner I hope!)

I also would like to see how much more monsters and variant monsters are added to the ultimate releases throughout the series.

Note I am using the japanese titles because they are released before their international counterparts.

The amount of large monsters added has changed from after Monster Hunter Portable 2nd. I think this is hard to answer the question "are there more monsters being added to ultimate releases" because as time goes on, the budget and popularity of the series increases and makes adding more monsters possible. Starting with Monster Hunter: World they have been slowly adding more monsters through constant updates made possible with access to the internet. Monster Hunter games in the past have always been released and never had new monsters being added.

Directors of Monster Hunter: Analysis

Let us see if the directors of Monster Hunter, Kaname Fujioka, Yasunori Ichinose, and the newest director Yuya Tokuda favor monsters they have introduced.

It is not possible to see if Yuya Tokuda favors his own monsters because he has only directed 2 titles, a base and ultimate release. Both Kaname Fujioka and Yasunori Ichinose have 7 title each to compare.

To see the differnce between Kaname Fujioka and Yasunori Ichinose we can take the mean of each 'Director Monster Ratio'.

Kaname Fujioka seems to favor his own monsters compared to Yasunori Ichinose. But Kaname Fujioka did work on 2 titles, Monster Hunter and Monster Hunter G before Yasunori Ichinose. As we found out in the Occurrence of Monsters: Analysis the most occuring monsters came from the first title Monster Hunter that was directed by Kaname Fujioka. By comparing the most_occurrences list to the monsters introduced by Kaname Fujioka.

In the most_occurrences list, monsters that Kaname Fujioka have introduced populated about ~81% of that list!

Conclusion

I do not think I have discovered any meaningful information about Monster Hunter but I do believe I have found some nice little details from Director Monsters and Occuring Types of Monsters sections. I made some nice figures that I hope the Monster Hunter community will enjoy. I feel much more comforatable with using pandas and I do feel there is much more I can do with them.

A big thanks to my SO for helping me proofread this notebook. And also the community that runs MonsterHunter.Fandom, it made this project much easier to collect the data.

Thank you so much for taking the time to read this, I am still trying to find a writing style that suits me. I plan on keeping on practicing skills that will help me in the future and sharing interesting and fun topics with many communities!

Completed: August 28th 2020