speedrunr

The goal of speedrunr is to easily access data from speedrun.com.

Installation

You can install the released version of speedrunr from GitHub with:

remotes::install_github("jemus42/speedrunr")

Example

Let’s say you want to plot the times of all Ocarina of TIme 100% runs.
Let’s get started:

library(speedrunr)
library(dplyr) # Data manip
library(knitr) # Tables

Identifiyng the game you’re looking for

You can either search for “Ocarina of Time”, or supply 'oot', the game’s abbreviation on speedrun.com.

games <- get_games(name = "Ocarina of Time")

games %>% 
  select(id, name_international, name_abbr) %>%
  head() %>%
  kable()

id	name_international	name_abbr
j1l9qz1g	The Legend of Zelda: Ocarina of Time	oot
kdkjex1m	The Legend of Zelda: Ocarina of Time Master Quest	ootmq
268vqkdp	The Legend of Zelda: Ocarina of Time 3D	oot3d
76rkv4d8	Ocarina of Time Category Extensions	ootextras
m1zromd0	Ocarina of Time Beta Quest	ootbq
v1pol9m6	SM64: Ocarina of Time	sm64oot

Turns out j1l9qz1g is the id we’re looking for.

Get the game’s categories

categories <- get_categories(id = "j1l9qz1g")

categories %>%
  select(id, name, type) %>%
  head() %>%
  kable()

id	name	type
q255jw2o	100%	per-game
824qn3k5	100%	per-level
zdnoz72q	All Dungeons	per-game
q25g198d	Any%	per-game
02qe4z2y	Any%	per-level
zd35jnkn	Glitchless	per-game

So apparently we’re looking for q255jw2o, the full-game 100% category.

Get the runs in that category

Now we can fetch the runs. By default, 100 runs are returned, ordered by submit date in descending order, so newest runs first. This also means you will only be able to fully assess the WR progression if you make sure to get all the runs.

runs <- get_runs(game = "j1l9qz1g", category = "q255jw2o")

glimpse(runs)
#> Rows: 100
#> Columns: 22
#> $ id              �[3m�[90m<chr>�[39m�[23m "y9015e2y", "yd55novm", "z5j7l4nm", "zpr4kk8m", "mrke…
#> $ weblink         �[3m�[90m<chr>�[39m�[23m "https://www.speedrun.com/oot/run/y9015e2y", "https:/…
#> $ game            �[3m�[90m<chr>�[39m�[23m "j1l9qz1g", "j1l9qz1g", "j1l9qz1g", "j1l9qz1g", "j1l9…
#> $ level           �[3m�[90m<lgl>�[39m�[23m NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, N…
#> $ category        �[3m�[90m<chr>�[39m�[23m "q255jw2o", "q255jw2o", "q255jw2o", "q255jw2o", "q255…
#> $ videos          �[3m�[90m<chr>�[39m�[23m "https://www.twitch.tv/videos/608790240", "https://ww…
#> $ status          �[3m�[90m<chr>�[39m�[23m "verified", "verified", "verified", "verified", "veri…
#> $ comment         �[3m�[90m<chr>�[39m�[23m NA, "3rd try Dampe", "This run sucks.", NA, "Finally …
#> $ player_id       �[3m�[90m<chr>�[39m�[23m "zx73zzvj", "v813d3xp", "pj00zwjw", "v8l62g78", "v8lr…
#> $ player_url      �[3m�[90m<chr>�[39m�[23m "https://www.speedrun.com/user/JoeCool", "https://www…
#> $ player_name     �[3m�[90m<chr>�[39m�[23m "JoeCool", "glitchymon", "EnNopp112", "EricDaCleric",…
#> $ player_role     �[3m�[90m<chr>�[39m�[23m "user", "user", "user", "user", "user", "user", "user…
#> $ player_signup   �[3m�[90m<dttm>�[39m�[23m 2019-09-10 02:55:31, 2015-03-05 22:12:19, 2015-06-25…
#> $ date            �[3m�[90m<date>�[39m�[23m 2020-05-01, 2020-04-06, 2020-04-12, 2020-04-21, 2020…
#> $ submitted       �[3m�[90m<dttm>�[39m�[23m 2020-05-02 04:21:10, 2020-04-28 11:01:30, 2020-04-21…
#> $ time_primary    �[3m�[90m<int>�[39m�[23m 18240, 13465, 13805, 13847, 35313, 14660, 20783, 1362…
#> $ time_realtime   �[3m�[90m<int>�[39m�[23m 18240, 13465, 13805, 13847, 35313, 14660, 20783, 1362…
#> $ time_ingame     �[3m�[90m<int>�[39m�[23m 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,…
#> $ time_hms        �[3m�[90m<time>�[39m�[23m 05:04:00, 03:44:25, 03:50:05, 03:50:47, 09:48:33, 04…
#> $ system_platform �[3m�[90m<chr>�[39m�[23m "nzelreqp", "nzelreqp", "nzelreqp", "nzelreqp", "w89r…
#> $ system_emulated �[3m�[90m<lgl>�[39m�[23m FALSE, FALSE, FALSE, FALSE, FALSE, FALSE, FALSE, FALS…
#> $ system_region   �[3m�[90m<chr>�[39m�[23m "o316x197", "o316x197", "o316x197", "o316x197", "o316…

And now we can basically re-create the leaderboard, but including obsoleted runs:

library(hms)

runs %>%
  arrange(time_primary) %>%
  head(20) %>%
  select(submitted, time_primary, player_name) %>%
  mutate(time_primary = hms(seconds = time_primary)) %>%
  kable()

submitted	time_primary	player_name
2020-04-28 11:01:30	03:44:25	glitchymon
2020-04-03 08:26:30	03:45:59	glitchymon
2020-04-12 11:36:02	03:47:00	AxelLarsen
2020-04-11 02:34:15	03:48:01	AxelLarsen
2020-03-19 00:14:45	03:49:23	glitchymon
2020-02-18 18:45:03	03:49:58	glitchymon
2020-04-21 12:30:19	03:50:05	EnNopp112
2020-03-28 11:59:22	03:50:31	AxelLarsen
2020-04-21 10:02:01	03:50:47	EricDaCleric
2020-02-29 09:55:40	03:50:47	Marco
2019-12-20 00:23:32	03:51:34	glitchymon
2020-03-22 10:06:41	03:52:14	AxelLarsen
2019-11-26 11:43:07	03:52:21	glitchymon
2019-12-19 17:59:29	03:53:27	Marco
2019-10-29 02:06:15	03:54:15	glitchymon
2020-03-20 12:07:33	03:54:49	AxelLarsen
2020-03-06 08:34:33	03:55:24	AxelLarsen
2019-11-16 07:33:14	03:55:32	Marco
2019-11-01 18:23:51	03:55:52	Marco
2020-03-30 17:31:17	03:57:18	EricDaCleric

More data

Wanna resolve those platforms? Just join with this table:

get_platforms() %>%
  head() %>%
  kable()

id	name	released
mr6km09z	MS-DOS	1970
8gej2n93	PC	1970
3167od9q	Plug & Play	1970
vm9vkn63	Tabletop	1970
w89ryw6l	Apple II	1977
o0644863	Atari 2600	1977

Same can be done with regions:

get_regions() %>%
  kable()

id	name
ypl25l47	BRA / PAL
mol4z19n	CHN / PAL
e6lxy1dz	EUR / PAL
o316x197	JPN / NTSC
p2g50lnk	KOR / NTSC
pr184lqn	USA / NTSC

There are also convenience functions to pipe your runs object into:

• add_platforms()
• add_regions()
• add_players(), which only makes on API call per unique player.

All of them work in the following way:

runs %>% 
  add_regions() %>%
  add_platforms() %>%
  select(time_primary, system_region, system_platform) %>%
  sample_n(5) %>%
  knitr::kable()

time_primary	system_region	system_platform
16178	JPN / NTSC	Wii Virtual Console
27921	JPN / NTSC	Wii Virtual Console
13894	JPN / NTSC	Wii Virtual Console
17804	JPN / NTSC	Wii Virtual Console
13831	JPN / NTSC	Wii Virtual Console

Code of Conduct

Please note that this project is released with a Contributor Code of Conduct. By participating in this project you agree to abide by its terms.