Code walk through
Load in the usual packages, and ggthemes so I can use the fivethirtyeight ggplot2 theme.
library(tidyverse)
library(lubridate)
library(ggthemes)
board_games <- read_csv("https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2019/2019-03-12/board_games.csv")
Let’s get a look at the data.
glimpse(board_games)
## Observations: 10,532
## Variables: 22
## $ game_id <dbl> 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, ...
## $ description <chr> "Die Macher is a game about seven sequential po...
## $ image <chr> "//cf.geekdo-images.com/images/pic159509.jpg", ...
## $ max_players <dbl> 5, 4, 4, 4, 6, 6, 2, 5, 4, 6, 7, 5, 4, 4, 6, 4,...
## $ max_playtime <dbl> 240, 30, 60, 60, 90, 240, 20, 120, 90, 60, 45, ...
## $ min_age <dbl> 14, 12, 10, 12, 12, 12, 8, 12, 13, 10, 13, 12, ...
## $ min_players <dbl> 3, 3, 2, 2, 3, 2, 2, 2, 2, 2, 2, 2, 3, 3, 2, 3,...
## $ min_playtime <dbl> 240, 30, 30, 60, 90, 240, 20, 120, 90, 60, 45, ...
## $ name <chr> "Die Macher", "Dragonmaster", "Samurai", "Tal d...
## $ playing_time <dbl> 240, 30, 60, 60, 90, 240, 20, 120, 90, 60, 45, ...
## $ thumbnail <chr> "//cf.geekdo-images.com/images/pic159509_t.jpg"...
## $ year_published <dbl> 1986, 1981, 1998, 1992, 1964, 1989, 1978, 1993,...
## $ artist <chr> "Marcus Gschwendtner", "Bob Pepper", "Franz Voh...
## $ category <chr> "Economic,Negotiation,Political", "Card Game,Fa...
## $ compilation <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,...
## $ designer <chr> "Karl-Heinz Schmiel", "G. W. \"Jerry\" D'Arcey"...
## $ expansion <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA, "Elfengold,...
## $ family <chr> "Country: Germany,Valley Games Classic Line", "...
## $ mechanic <chr> "Area Control / Area Influence,Auction/Bidding,...
## $ publisher <chr> "Hans im Glück Verlags-GmbH,Moskito Spiele,Vall...
## $ average_rating <dbl> 7.66508, 6.60815, 7.44119, 6.60675, 7.35830, 6....
## $ users_rated <dbl> 4498, 478, 12019, 314, 15195, 73, 2751, 186, 12...
First off, I want to see the cumulative board game total count over the years.
cumulative <-
board_games %>%
count(year_published) %>%
mutate(cumulative = cumsum(n))
cumulative
## # A tibble: 67 x 3
## year_published n cumulative
## <dbl> <int> <int>
## 1 1950 4 4
## 2 1951 2 6
## 3 1952 3 9
## 4 1953 3 12
## 5 1954 3 15
## 6 1955 4 19
## 7 1956 6 25
## 8 1957 2 27
## 9 1958 8 35
## 10 1959 7 42
## # ... with 57 more rows
cumulative %>%
ggplot(aes(year_published, cumulative)) +
geom_point() +
geom_line() +
scale_x_continuous(breaks = seq(1950, 2020, 10)) +
scale_y_continuous(labels = scales::comma) +
labs(x = "Year",
y = "Cumulative No. of Board Games") +
theme_fivethirtyeight()
The number of board games picked up in the 70s and really got going exponential after 2000.
I want to do the same thing, but look at how the use of different game mechanics has changed over the years. A game mechanic is how a game is played like rolling dice, drawing, or storytelling. Let’s take a look at the mechanic column.
board_games %>% select(mechanic)
## # A tibble: 10,532 x 1
## mechanic
## <chr>
## 1 Area Control / Area Influence,Auction/Bidding,Dice Rolling,Hand Managem~
## 2 Trick-taking
## 3 Area Control / Area Influence,Hand Management,Set Collection,Tile Place~
## 4 Action Point Allowance System,Area Control / Area Influence,Auction/Bid~
## 5 Hand Management,Stock Holding,Tile Placement
## 6 Dice Rolling
## 7 Area Enclosure,Pattern Building,Pattern Recognition,Tile Placement
## 8 Modular Board
## 9 Area Control / Area Influence,Tile Placement
## 10 Card Drafting,Hand Management,Point to Point Movement,Route/Network Bui~
## # ... with 10,522 more rows
It looks like a board game can have multiple mechanic categories as we see here separated by commas. What I need to do to analyze this type of data is force each row to have just one category while maintaining some contextual information. In this case I just want to keep a hold of the year column. Before we start, I’m just curious, what game had the most combinations of game mechanics?
board_games %>%
select(year_published, name, mechanic) %>%
mutate(m_count = str_count(mechanic, ",")) %>%
arrange(desc(m_count))
## # A tibble: 10,532 x 4
## year_published name mechanic m_count
## <dbl> <chr> <chr> <int>
## 1 2015 504 Area Control / Area Influence,Ar~ 17
## 2 2014 Emperor's New ~ Acting,Action Point Allowance Sy~ 14
## 3 2013 Patchistory Action Point Allowance System,Ar~ 11
## 4 2013 City of Remnan~ Action Point Allowance System,Ar~ 10
## 5 2016 Pyramid Arcade Area Control / Area Influence,Be~ 10
## 6 2011 Mage Knight Bo~ Card Drafting,Co-operative Play,~ 9
## 7 2012 Exodus: Proxim~ Area Control / Area Influence,Ar~ 9
## 8 2015 Chaosmos Action Point Allowance System,Ca~ 9
## 9 2014 Sons of Anarch~ Action Point Allowance System,Ar~ 9
## 10 1985 Advanced Squad~ Auction/Bidding,Dice Rolling,Hex~ 8
## # ... with 10,522 more rows
Hmm, 504, I’m not familiar with it. Looks like settlers of catan after with a cash system after looking here. Probably very complicated.
Now on to the data manipulation. I’ll use tidyr::separate_rows() to separate the categories and puts each of them into their own row.
mechanics_count <-
board_games %>%
select(year_published, mechanic) %>%
drop_na(mechanic) %>%
separate_rows(mechanic, sep = ",")
mechanics_count
## # A tibble: 23,950 x 2
## year_published mechanic
## <dbl> <chr>
## 1 1986 Area Control / Area Influence
## 2 1986 Auction/Bidding
## 3 1986 Dice Rolling
## 4 1986 Hand Management
## 5 1986 Simultaneous Action Selection
## 6 1981 Trick-taking
## 7 1998 Area Control / Area Influence
## 8 1998 Hand Management
## 9 1998 Set Collection
## 10 1998 Tile Placement
## # ... with 23,940 more rows
Now we have all the game mechanics used with their corresponding year. Notice we went from 9.5K rows to 23.9K which is expected.
mechanics_count %>%
select(mechanic) %>%
unique() %>%
drop_na(mechanic) %>%
nrow()
## [1] 51
Before we get into plotting, I need to find the top 6 most occurring game mechanics. Having a total of 51 different game mechanics would make my plot hard to understand so let’s keep it simple.
top_mechanics <-
mechanics_count %>%
count(mechanic, sort = T) %>%
top_n(n = 6, wt = n) %>%
pull(mechanic)
top_mechanics
## [1] "Dice Rolling" "Hand Management"
## [3] "Set Collection" "Hex-and-Counter"
## [5] "Variable Player Powers" "Tile Placement"
Perfect, now lets plot them using a faceted density plot.
mechanics_count %>%
filter(mechanic %in% top_mechanics) %>%
mutate(mechanic = factor(mechanic, levels = top_mechanics)) %>%
ggplot(aes(year_published, fill = mechanic)) +
geom_density(col = "grey90", show.legend = F) +
facet_wrap(~mechanic, ncol = 3) +
scale_x_continuous(breaks = seq(1960, 2010, 10)) +
scale_fill_tableau() +
labs(title = "Most Occurring Board Game Mechanics over the years") +
theme_fivethirtyeight() +
theme(axis.text.y = element_blank(),
axis.text.x = element_text(angle = 90))
All but one game mechanic seem to have rising in popularity rather quickly over the past two decades, while the Hex-and-Counter method has maintained a good share for for almost 40 years!
