Beach Volleyball
The data this week comes from Adam Vagnar who also blogged about this dataset. There’s a LOT of data here - match-level results, player details, and match-level statistics for some matches. For all this dataset all the matches are played 2 vs 2, so there are columns for 2 winners (1 team) and 2 losers (1 team). The data is relatively ready for analysis and clean, although there are some duplicated columns and the data is wide due to the 2-players per team.
Check out the data dictionary, or Wikipedia for some longer-form details around what the various match statistics mean.
Most of the data is from the international FIVB tournaments but about 1/3 is from the US-centric AVP.
The FIVB Beach Volleyball World Tour (known between 2003 and 2012 as the FIVB Beach Volleyball Swatch World Tour for sponsorship reasons) is the worldwide professional beach volleyball tour for both men and women organized by the Fédération Internationale de Volleyball (FIVB). The World Tour was introduced for men in 1989 while the women first competed in 1992.
Winning the World Tour is considered to be one of the highest honours in international beach volleyball, being surpassed only by the World Championships, and the Beach Volleyball tournament at the Summer Olympic Games.
FiveThirtyEight examined the disadvantage of serving in beach volleyball, although they used Olympic-level data. Again, Adam Vagnar also covered this data on his blog.
Get the data here
# Get the Data
vb_matches <- readr::read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/main/data/2020/2020-05-19/vb_matches.csv', guess_max = 76000)
# Or read in with tidytuesdayR package (https://github.com/dslc-io/tidytuesdayR)
# Either ISO-8601 date or year/week works!
# Install via pak::pak("dslc-io/tidytuesdayR")
tuesdata <- tidytuesdayR::tt_load('2020-05-19')
tuesdata <- tidytuesdayR::tt_load(2020, week = 21)
vb_matches <- tuesdata$vb_matchesData Dictionary
vb_matches.csv
| variable | class | description |
|---|---|---|
| circuit | character | Either AVP (USA) or FIVB (International) |
| tournament | character | Tournament City or Name |
| country | character | Country where tournament played |
| year | double | Year of tournament |
| date | double | Date of match |
| gender | character | Gender of team |
| match_num | double | Match Number |
| w_player1 | character | Winner player 1 Name |
| w_p1_birthdate | double | Winner player 1 birth date |
| w_p1_age | double | Winner player 1 age |
| w_p1_hgt | double | Winner player 1 height in inches |
| w_p1_country | character | Winner player country |
| w_player2 | character | Winner player 2 name |
| w_p2_birthdate | double | Winner player 2 birthdate |
| w_p2_age | double | Winner player 2 age |
| w_p2_hgt | double | Winner player 2 height in inches |
| w_p2_country | character | Winner player 2 country |
| w_rank | character | Winner team rank |
| l_player1 | character | Losing player 1 name |
| l_p1_birthdate | double | Losing player 1 birthdate |
| l_p1_age | double | Losing player 1 age |
| l_p1_hgt | double | Losing player 1 height in inches |
| l_p1_country | character | Losing player 1 country |
| l_player2 | character | Losing player 2 name |
| l_p2_birthdate | double | Losing player 2 birthdate |
| l_p2_age | double | Losing player 2 age |
| l_p2_hgt | double | Losing player 2 height in inches |
| l_p2_country | character | Losing player 2 country |
| l_rank | character | Losing team rank |
| score | character | Match score separated by a dash and matches separated by a comma, eg 21 points to 12 points is 21-12 |
| duration | double | Duration of match in minutes |
| bracket | character | Tournament bracket |
| round | character | Tournament round |
| w_p1_tot_attacks | double | Winner player 1 number of attacks (attacking swings over the net) |
| w_p1_tot_kills | double | Winner player 1 number of kills (point ending attacks) |
| w_p1_tot_errors | double | Winner player 1 mistakes |
| w_p1_tot_hitpct | double | Winner player 1 hitting percentage - calculated as (kills-errors)/attacks - this is the player’s effectiveness at scoring |
| w_p1_tot_aces | double | Winner player 1 total aces - point ending serves |
| w_p1_tot_serve_errors | double | Winner player 1 total serving errors - mistakes made on serve |
| w_p1_tot_blocks | double | Winner player 1 total blocks - point ending blocks |
| w_p1_tot_digs | double | Winner player 1 total digs - successful defense of an attack |
| w_p2_tot_attacks | double | Winner player 2 number of attacks (attacking swings over the net) |
| w_p2_tot_kills | double | Winner player 2 number of kills (point ending attacks) |
| w_p2_tot_errors | double | Winner player 2 mistakes |
| w_p2_tot_hitpct | double | Winner player 2 hitting percentage - calculated as (kills-errors)/ |
| w_p2_tot_aces | double | Winner player 2 total aces - point ending serves |
| w_p2_tot_serve_errors | double | Winner player 2 total serving errors - mistakes made on serve |
| w_p2_tot_blocks | double | Winner player 2 total blocks - point ending blocks |
| w_p2_tot_digs | double | Winner player 2 total digs - successful defense of an attack |
| l_p1_tot_attacks | double | Losing player 1 number of attacks (attacking swings over the net) |
| l_p1_tot_kills | double | Losing player 1 number of kills (point ending attacks) |
| l_p1_tot_errors | double | Losing player 1 mistakes |
| l_p1_tot_hitpct | double | Losing player 1 hitting percentage - calculated as (kills-errors)/ |
| l_p1_tot_aces | double | Losing player 1 total aces - point ending serves |
| l_p1_tot_serve_errors | double | Losing player 1 total serving errors - mistakes made on serve |
| l_p1_tot_blocks | double | Losing player 1 total blocks - point ending blocks |
| l_p1_tot_digs | double | Losing player 1 total digs - successful defense of an attack |
| l_p2_tot_attacks | double | Losing player 2 number of attacks (attacking swings over the net) |
| l_p2_tot_kills | double | Losing player 2 number of kills (point ending attacks) |
| l_p2_tot_errors | double | Losing player 2 mistakes |
| l_p2_tot_hitpct | double | Losing player 2 hitting percentage - calculated as (kills-errors)/ |
| l_p2_tot_aces | double | Losing player 2 total aces - point ending serves |
| l_p2_tot_serve_errors | double | Losing player 2 total serving errors - mistakes made on serve |
| l_p2_tot_blocks | double | Losing player 2 total blocks - point ending blocks |
| l_p2_tot_digs | double | Losing player 2 total digs - successful defense of an attack |
skimr
── Data Summary ────────────────────────
Values
Name vb_matches
Number of rows 76756
Number of columns 65
_______________________
Column type frequency:
character 17
Date 5
difftime 1
numeric 42
________________________
Group variables None
── Variable type: character ───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
skim_variable n_missing complete_rate min max empty n_unique whitespace
1 circuit 0 1 3 4 0 2 0
2 tournament 0 1 3 22 0 177 0
3 country 0 1 4 22 0 51 0
4 gender 0 1 1 1 0 2 0
5 w_player1 0 1 6 29 0 3388 0
6 w_p1_country 12 1.00 4 20 0 85 0
7 w_player2 0 1 5 30 0 3431 0
8 w_p2_country 5 1.00 4 20 0 87 0
9 w_rank 148 0.998 1 7 0 812 0
10 l_player1 0 1 5 29 0 5713 0
11 l_p1_country 18 1.00 4 20 0 109 0
12 l_player2 0 1 5 30 0 5689 0
13 l_p2_country 10 1.00 4 20 0 111 0
14 l_rank 1240 0.984 1 7 0 837 0
15 score 22 1.00 4 25 0 6624 0
16 bracket 0 1 6 21 0 36 0
17 round 4939 0.936 7 8 0 10 0
── Variable type: Date ────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
skim_variable n_missing complete_rate min max median n_unique
1 date 0 1 2000-09-16 2019-08-29 2009-08-25 658
2 w_p1_birthdate 383 0.995 1953-06-13 2004-07-15 1981-10-30 2805
3 w_p2_birthdate 408 0.995 1952-10-11 2004-06-08 1981-10-15 2847
4 l_p1_birthdate 1059 0.986 1953-06-13 2004-12-01 1982-03-28 4236
5 l_p2_birthdate 959 0.988 1949-12-04 2004-08-12 1982-03-20 4282
── Variable type: difftime ────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
skim_variable n_missing complete_rate min max median n_unique
1 duration 2249 0.971 120 secs 8040 secs 42'00" 108
── Variable type: numeric ─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
skim_variable n_missing complete_rate mean sd hist
1 year 0 1 2010. 5.48 ▃▇▆▅▇
2 match_num 0 1 31.8 23.5 ▇▅▂▁▁
3 w_p1_age 383 0.995 28.7 5.05 ▂▇▃▁▁
4 w_p1_hgt 3966 0.948 73.7 3.64 ▁▅▇▃▁
5 w_p2_age 408 0.995 28.8 4.85 ▁▇▆▁▁
6 w_p2_hgt 4016 0.948 73.7 3.69 ▁▃▇▆▁
7 l_p1_age 1059 0.986 28.3 5.26 ▂▇▂▁▁
8 l_p1_hgt 6988 0.909 73.4 3.62 ▁▃▇▅▁
9 l_p2_age 959 0.988 28.4 5.12 ▂▇▁▁▁
10 l_p2_hgt 6983 0.909 73.5 3.65 ▁▃▇▅▁
11 w_p1_tot_attacks 62178 0.190 25.9 10.0 ▇▅▁▁▁
12 w_p1_tot_kills 62178 0.190 14.7 5.34 ▂▇▅▁▁
13 w_p1_tot_errors 62413 0.187 2.90 2.27 ▇▁▁▁▁
14 w_p1_tot_hitpct 62185 0.190 0.480 0.230 ▇▁▁▁▁
15 w_p1_tot_aces 60560 0.211 1.32 1.45 ▇▂▁▁▁
16 w_p1_tot_serve_errors 62417 0.187 2.03 1.65 ▇▃▁▁▁
17 w_p1_tot_blocks 60560 0.211 1.70 2.15 ▇▂▁▁▁
18 w_p1_tot_digs 62178 0.190 8.35 5.48 ▇▅▁▁▁
19 w_p2_tot_attacks 62174 0.190 26.1 10.1 ▇▇▁▁▁
20 w_p2_tot_kills 62174 0.190 14.8 5.33 ▂▇▅▁▁
21 w_p2_tot_errors 62413 0.187 2.92 2.29 ▇▁▁▁▁
22 w_p2_tot_hitpct 62181 0.190 0.477 0.164 ▁▇▁▁▁
23 w_p2_tot_aces 60556 0.211 1.19 1.36 ▇▁▁▁▁
24 w_p2_tot_serve_errors 62413 0.187 1.93 1.62 ▇▃▁▁▁
25 w_p2_tot_blocks 60556 0.211 1.69 2.19 ▇▂▁▁▁
26 w_p2_tot_digs 62174 0.190 8.54 5.56 ▇▃▁▁▁
27 l_p1_tot_attacks 62179 0.190 27.1 11.1 ▇▁▁▁▁
28 l_p1_tot_kills 62179 0.190 12.8 5.76 ▃▇▃▁▁
29 l_p1_tot_errors 62413 0.187 4.38 2.76 ▇▂▁▁▁
30 l_p1_tot_hitpct 62189 0.190 0.313 0.176 ▃▇▁▁▁
31 l_p1_tot_aces 60561 0.211 0.776 1.04 ▇▂▁▁▁
32 l_p1_tot_serve_errors 62418 0.187 2.10 1.66 ▇▃▁▁▁
33 l_p1_tot_blocks 60561 0.211 0.997 1.53 ▇▁▁▁▁
34 l_p1_tot_digs 62179 0.190 7.19 5.17 ▇▂▁▁▁
35 l_p2_tot_attacks 62178 0.190 26.7 10.8 ▃▇▁▁▁
36 l_p2_tot_kills 62178 0.190 12.6 5.66 ▃▇▃▁▁
37 l_p2_tot_errors 62413 0.187 4.32 2.71 ▇▃▁▁▁
38 l_p2_tot_hitpct 62189 0.190 0.313 0.176 ▂▇▁▁▁
39 l_p2_tot_aces 60560 0.211 0.775 1.06 ▇▁▁▁▁
40 l_p2_tot_serve_errors 62417 0.187 2.05 1.66 ▇▂▁▁▁
41 l_p2_tot_blocks 60560 0.211 1.06 1.56 ▇▁▁▁▁
42 l_p2_tot_digs 62178 0.190 7.14 5.18 ▇▃▁▁▁Cleaning Script
Data is already pretty clean! You may want to pivot the data by team or optionally separate out the winning/losing scores by match.
library(tidyverse)
col_types_vb <- cols(
circuit = col_character(),
tournament = col_character(),
country = col_character(),
year = col_double(),
date = col_date(format = ""),
gender = col_character(),
match_num = col_double(),
w_player1 = col_character(),
w_p1_birthdate = col_date(format = ""),
w_p1_age = col_double(),
w_p1_hgt = col_double(),
w_p1_country = col_character(),
w_player2 = col_character(),
w_p2_birthdate = col_date(format = ""),
w_p2_age = col_double(),
w_p2_hgt = col_double(),
w_p2_country = col_character(),
w_rank = col_character(),
l_player1 = col_character(),
l_p1_birthdate = col_date(format = ""),
l_p1_age = col_double(),
l_p1_hgt = col_double(),
l_p1_country = col_character(),
l_player2 = col_character(),
l_p2_birthdate = col_date(format = ""),
l_p2_age = col_double(),
l_p2_hgt = col_double(),
l_p2_country = col_character(),
l_rank = col_character(),
score = col_character(),
duration = col_time(format = ""),
bracket = col_character(),
round = col_character(),
w_p1_tot_attacks = col_double(),
w_p1_tot_kills = col_double(),
w_p1_tot_errors = col_double(),
w_p1_tot_hitpct = col_double(),
w_p1_tot_aces = col_double(),
w_p1_tot_serve_errors = col_double(),
w_p1_tot_blocks = col_double(),
w_p1_tot_digs = col_double(),
w_p2_tot_attacks = col_double(),
w_p2_tot_kills = col_double(),
w_p2_tot_errors = col_double(),
w_p2_tot_hitpct = col_double(),
w_p2_tot_aces = col_double(),
w_p2_tot_serve_errors = col_double(),
w_p2_tot_blocks = col_double(),
w_p2_tot_digs = col_double(),
l_p1_tot_attacks = col_double(),
l_p1_tot_kills = col_double(),
l_p1_tot_errors = col_double(),
l_p1_tot_hitpct = col_double(),
l_p1_tot_aces = col_double(),
l_p1_tot_serve_errors = col_double(),
l_p1_tot_blocks = col_double(),
l_p1_tot_digs = col_double(),
l_p2_tot_attacks = col_double(),
l_p2_tot_kills = col_double(),
l_p2_tot_errors = col_double(),
l_p2_tot_hitpct = col_double(),
l_p2_tot_aces = col_double(),
l_p2_tot_serve_errors = col_double(),
l_p2_tot_blocks = col_double(),
l_p2_tot_digs = col_double()
)
raw_df <- c("https://raw.githubusercontent.com/BigTimeStats/beach-volleyball/master/data/match_archive_2000_to_2017_v2.csv",
"https://raw.githubusercontent.com/BigTimeStats/beach-volleyball/master/data/match_update_20170729_to_20170912.csv",
"https://raw.githubusercontent.com/BigTimeStats/beach-volleyball/master/data/match_update_20170913_to_20180314.csv",
"https://raw.githubusercontent.com/BigTimeStats/beach-volleyball/master/data/match_update_20180315_to_20180821.csv",
"https://raw.githubusercontent.com/BigTimeStats/beach-volleyball/master/data/match_update_20180822_to_20190409.csv",
"https://raw.githubusercontent.com/BigTimeStats/beach-volleyball/master/data/match_update_20190410_to_20190818.csv",
"https://raw.githubusercontent.com/BigTimeStats/beach-volleyball/master/data/match_update_20190818_to_20190902.csv") %>%
map_dfr(read_csv, col_types = col_types_vb)
raw_df %>%
skimr::skim()
# Georgios Karamanis noticed that the birthdates are
# incorrect for anyone born before 1970 (off by 100 years)
clean_df <- mutate_at(
raw_df,
vars(contains("birthdate")),
list(~ if_else(. >= as.Date("2020-01-01"),
. - lubridate::years(100),
.
))
)
write_csv(clean_df, "2020/2020-05-19/vb_matches.csv")