Summary

A strength of tennis as a professional sport is that it provides the most opportunities for female athletes to prosper. Yet, despite the successes of women’s professional tennis (e.g., equal prize money at the four major ‘Grand Slam’ tournaments), stubborn gender gaps remain: male players achieve higher cultural recognition and overall earnings. The visibility and earning potential of players can be influenced by the scheduling decisions at major tournaments, and the programming decisions of broadcasters. Here, I summarise gender bias in match scheduling on the top three courts at The Australian Open across 13 years (2009-2021).

Night sessions at The Australian Open attract a larger broadcast audience than day sessions, both locally and in Europe, and are therefore considered the ‘prime time’ slot. Prime time between 2009 and 2021 featured fewer women’s than men’s matches in every year. This pattern has not improved over time, and is driven by the second week of the tournament, when the majority of women’s quarterfinals and the entirety of women’s semifinals were played during day sessions. When both women’s and men’s matches were played in day sessions, women’s matches tended to be scheduled earlier in the day.

Overall, the pattern of scheduling decisions at The Australian Open since 2009 show that women’s matches have not been given the same opportunity as men’s matches to attract a large audience. This norm is set to continue, with the ticketing website for the 2021 tournament specifying “Women’s semifinals (day) and the first men’s semifinal (night) take place on Thursday 18 February, with the second men’s semifinal on Friday 19 February”. Ticket prices reinforce the impression that the women’s tournament plays second violin; ticket prices for women’s semifinals range from 94 to 495 AUD, whereas prices for just one of the men’s semifinals range between 189 and 715 AUD. To some, these patterns are unproblematic reflections of economic forces – people want to watch the stars of men’s tennis, so of course those matches are more valued by ticket buyers and the host broadcaster. But supply the public with more women’s tennis matches, and they might become equally interested in women’s tennis stars too.

The results presented here complement Yip’s 2016 research on gender bias on the official website of the 2015 Australian Open. Fans of women’s tennis might also recall frustrations at the strategic decisions of Tennis Australia to invest in a men’s only team event as a prelude to The Australian Open. Recently, Tennis Australia received $12 million funding from the Australian Federal Government to increase “opportunities for women and girls in tennis”. As an organisation Tennis Australia’s stated goals are to produce “More active players”, “More great champions”, “More devoted fans”, and “Healthier communities”. Showing equal respect to men’s and women’s by allocating equal time in the spotlight at The Australian Open should help grow the fanbase for women’s tennis, and engage more recreational players.

Methods

This Methods section is organised in three tabs. Click on each sub-heading to view the text within each tab.

Wikipedia data extraction

Wikipedia contains order-of-play tables for all “Matches on Main Courts” at The Australian Open between 2009-2021, in “Day-by-day summaries” pages: 2009, 2010, 2011, 2012, 2013, 2014, 2015, 2016, 2017, 2018, 2019, 2020, and 2021. (Note that the order of play prior to 2009 is either absent or, in the case of 2008, only partially provided).

I gathered information on the men’s and women’s singles tournaments from the 13 years from 2009-2021 using the rvest package in R, and basic text wrangling.

Metadata, data, and code

The custom function for web scraping, data, and meta-data are available from doi.org/10.17605/OSF.IO/UERPZ, and can be cited as:

  • O’Dea, R.E. 2021. “Matches on Main Courts at The Australian Open After 2008.” Open Science Framework. DOI: 10.17605/OSF.IO/UERPZ.

Please email me if you spot errors in the data or code.

Tournament Data Description

In the four grand slam tournaments, the men’s and women’s singles events begin with 128 players each. In every round, half the players are eliminated, until two players remain in the final.

Data covers 1441 matches from 2009-2021, on four courts: Rod Laver Arena (all years; 692 matches), John Cain Arena (all years; 359 matches; formerly known as Melbourne Arena or Hisense Arena), Margaret Court Arena (all years; 390 matches), and 1573 Arena (Show Court 2; 47 matches; 2019-2021 only, and not included in the below visualisations). All singles matches from the fourth round onwards are played on the largest courts, and all finals matches are played on Rod Laver Arena. Therefore, these data are complete for the second week of each tournament. Around 79%, 42%, and 22% of matches are represented from third, second, and first rounds, respectively.

Women’s singles matches are best-of-three sets, whereas men’s singles matches are best-of-five sets. The average number of sets and games for women’s matches across the tournaments was 2.3 sets and 20.8 games, while the average number for men’s singles matches was 3.6 sets and 34.7 games.

Results

Fewer women’s singles matches in prime time

In the below graphs, left panels (light backgrounds) show court allocations for day sessions, while right panels (dark backgrounds) show court allocations for night sessions. Horizontal columns depict either: (1) purple bar = more women’s singles matches than men’s; (2) orange bar = more men’s singles matches than women’s; (3) purple and orange bar stacked next to each other = equal numbers of women’s and men’s singles matches. The x-axis shows the gender proportion, with the vertical line at 0.5 indicating an equal gender split.

Click on the smaller sub-headings to view content in different tabs.

Centre Court

across rounds: spotlight moves in the second week

In the tournament’s first week, night sessions on Rod Laver Arena are approximately equally split between women’s and men’s matches (percentage of women’s matches in the first, second, and third round average 50%, 51%, and 47%, respectively). From the second week, the spotlight on Rod Laver Arena shines more on men. After the near-equal fourth round (49% women’s matches), only 3 women’s quarterfinals (13%) were played at night: the 2009 quarterfinal between Dinara Safina and Jelena Dokić, the 2018 quarterfinal between Caroline Wozniaki and Carla Suárez Navarro, and the 2019 quarterfinal between Petra Kvitova and Ash Barty (in two of those quarterfinals, the presence of Australian players – Dokić and Barty – likely promoted those matches to prime time). All women’s semifinals are played in the day session on Day 11 of the tournament (and are ticketed as such), followed by the women’s final in the night session on the final day. The men’s final – as is customary across all major tournaments – is played on the final day.

across years: fewer women’s matches in prime time

Since 2009, women’s singles matches on Rod Laver Arena have been allocated to more day sessions and fewer night sessions compared with men’s singles matches.

Second and Third Tier Courts

Results for the gender proportions in singles matches on John Cain Arena and Margaret Court Arena are tabbed within two sections (across rounds, and across years). Click on sub-headings to see different tabs.

across rounds

John Cain Arena

Margaret Court Arena

across years

Overall, fewer women’s matches have been played at night on John Cain Arena (26%) compared to men’s matches (74%), but this pattern has changed in the last couple of years. According to Wikipedia’s records, no singles matches were played at night on John Cain Arena in 2009, 2015, 2017, or 2018 (hence the gaps between columns in the below graph – this may be because matches played at night were not ticketed as a separate session).

Margaret Court Arena has also played fewer women’s matches at night (38%), but since 2016 the gender split has been close to equal.

John Cain Arena

Margaret Court Arena

Who plays first?

To explore how women’s and men’s singles matches are ordered within sessions, I looked at unique combinations of match orders. For example, in a session with two singles matches, did the women’s match come first, or was the men’s match scheduled before the women’s match? For each unique match combination, I counted the number of times that particular ordering occurred on the three main courts: Rod Laver Arena (RLA), John Cain Arena (JCA), and Margaret Court Arena (MCA). The below tabs (‘Night Sessions’ and ‘Day Sessions’) show which ordering were most common.

In the below table, the left column displays all combinations of singles matches within a session (night or day) on a main court. The combinations are arranged in the order of frequency on the main court (RLA = Rod Laver Arena), with the top row showing the most common night session on RLA. The right-most columns display the frequency of singles match ordering for the second and third-tier stadiums.

Night Sessions

On Rod Laver Arena, the only round when women play the lone singles match in a night session is the women’s final (on the penultimate day of the tournament). On the occasions when two matches were played, the night session matches always featured both a men’s and women’s match. Usually the women’s match is played first.

Most night sessions on John Cain Arena have featured one men’s match, while night matches on Margaret Court Arena have either featured one men’s match, or a women’s and a men’s match.

Day Sessions

The vast majority of day sessions on Rod Laver Arena begin with a women’s match.

Day sessions on John Cain and Margaret Court Arenas show a more even split in the ordering of women’s and men’s matches (and have more combinations of matches, as separate night sessions are not always sold).

Pairs of players allocated to night sessions

The below graphs show the proportion of times when, for matches allocated to a day or night session on a main court: (1) both players were seeded (top horizontal column); (2) one player in the match was seeded (middle horizontal column); or (3) neither player was seeded (bottom horizontal column). Columns are divided into purple and orange to show the relative number of women’s and men’s singles matches within each category.

Centre Court

Across the past 12 years, most matches on Rod Laver Arena involved one seeded player against one unseeded player (56% of matches). The gender ratio of matches with one seeded player was approximately equal, representing the night matches played during the first week of the tournament. In the second week, when matches between seeded players are more likely, men’s matches were more often played at night. In the very rare cases (4%) where a match between two unseeded players was scheduled on Rod Laver Arena, the match usually involved an Australian player (19 matches = 66%).

Second and Third Courts

Matches on the second and third show courts, as with the centre court, are most likely to contain one seeded player, and least likely to contain two unseeded players. Seeded women players have been less likely than men to be scheduled on John Cain Arena (formerly Hisense and Melbourne Arena), which has often been the largest court accessible by people with the cheapest tickets (groundspass holders). On Margaret Court Arena, women’s matches with one seeded player are more likely to be scheduled during the day session than at night.

John Cain Arena

Margaret Court Arena

Star Power: Who has played on centre court the most?

Since 2009, 22 people have played at least 14 matches on Rod Laver Arena:
player_name singles_event n_matches n_years
Roger Federer Men’s Singles 69 12
Novak Djokovic Men’s Singles 68 13
Rafael Nadal Men’s Singles 64 12
Serena Williams Women’s Singles 57 11
Maria Sharapova Women’s Singles 35 10
Andy Murray Men’s Singles 34 9
Victoria Azarenka Women’s Singles 28 8
Angelique Kerber Women’s Singles 22 9
Caroline Wozniacki Women’s Singles 22 9
Simona Halep Women’s Singles 22 7
Samantha Stosur Women’s Singles 19 8
Ashleigh Barty Women’s Singles 18 6
Garbiñe Muguruza Women’s Singles 17 8
Petra Kvitová Women’s Singles 17 7
Venus Williams Women’s Singles 17 8
Li Na Women’s Singles 16 5
Lleyton Hewitt Men’s Singles 16 8
Agnieszka Radwańska Women’s Singles 15 7
Karolína Plíšková Women’s Singles 15 6
Naomi Osaka Women’s Singles 15 5
Kim Clijsters Women’s Singles 14 3
Tomáš Berdych Men’s Singles 14 10

The distribution of the number of matches played by individuals on Rod Laver Arena has an extremely long right-tail. Just over half of all tennis players who played an Australian Open match on Rod Laver Arena since 2009 have not returned (50% of players).

The below graph is interactive. Hover your cursor over the bars to read players names.

Acknowledgements

This project was inspired by a tweet from Svenja Mastroberardino during the 2020 Australian Open.

Data collection relied on Wikipedia, which relies on donations.

For tennis fans and podcast listeners, my favourites are No Challenges Remaining, The Tennis Podcast, The Body Serve, and Tennish.

To explore the data yourself, go to doi.org/10.17605/OSF.IO/UERPZ.

R Session Information

## R version 4.0.2 (2020-06-22)
## Platform: x86_64-apple-darwin17.0 (64-bit)
## Running under: macOS Catalina 10.15.4
## 
## Matrix products: default
## BLAS:   /Library/Frameworks/R.framework/Versions/4.0/Resources/lib/libRblas.dylib
## LAPACK: /Library/Frameworks/R.framework/Versions/4.0/Resources/lib/libRlapack.dylib
## 
## attached base packages:
## [1] stats     graphics  grDevices utils     datasets  methods   base     
## 
## other attached packages:
## [1] kableExtra_1.1.0 knitr_1.28       plotly_4.9.2.1   cowplot_1.0.0   
## [5] extrafont_0.17   ggplot2_3.3.2    stringr_1.4.0    dplyr_1.0.0     
## 
## loaded via a namespace (and not attached):
##  [1] Rcpp_1.0.5        highr_0.8         pillar_1.4.6      compiler_4.0.2   
##  [5] tools_4.0.2       digest_0.6.25     viridisLite_0.3.0 jsonlite_1.7.2   
##  [9] evaluate_0.14     lifecycle_0.2.0   tibble_3.0.3      gtable_0.3.0     
## [13] pkgconfig_2.0.3   rlang_0.4.10      rstudioapi_0.11   crosstalk_1.1.0.1
## [17] yaml_2.2.1        xfun_0.20         Rttf2pt1_1.3.8    xml2_1.3.2       
## [21] withr_2.2.0       httr_1.4.2        hms_0.5.3         generics_0.0.2   
## [25] vctrs_0.3.2       htmlwidgets_1.5.1 webshot_0.5.2     grid_4.0.2       
## [29] tidyselect_1.1.0  glue_1.4.1        data.table_1.13.0 R6_2.4.1         
## [33] rmarkdown_2.6     pacman_0.5.1      farver_2.0.3      readr_1.3.1      
## [37] tidyr_1.1.0       purrr_0.3.4       extrafontdb_1.0   magrittr_1.5     
## [41] scales_1.1.1      ellipsis_0.3.1    htmltools_0.5.1.1 rvest_0.3.5      
## [45] colorspace_1.4-1  labeling_0.3      stringi_1.4.6     lazyeval_0.2.2   
## [49] munsell_0.5.0     crayon_1.3.4      Cairo_1.5-12.2