Does greater competition load going into the second week of a Grand Slam hurt a player's win chances?
By the time of the Australian Open quarterfinals, Dominic Thiem had already played 10 hours and 23 minutes putting him in the top 20% of match time played to get to the quarters among men competing in Grand Slams in the past decade. And he only got further in the tail of total competition load as the event progressed. In this post, I look at whether the intensity of Thiem’s journey likely made his prospects for the title even more difficult.
Few players are going to go deep into a Grand Slam without some wear and tear. But for some players that journey is much more arduous than for others.
This was the situation Dominic Thiem found himself in for the last three rounds of the 2020 Australian Open. Going into his unlucky quarterfinal matchup against Rafael Nadal, Thiem had accumulated 10.4 hours of match play, this put him +1.3 hours above the average over slam quarterfinalists of the past decade (see Figure 1).
After a four-set battle over Rafael Nadal that included three tiebreaks, Thiem was as 14.6 hours played and +3.4 hours over the average. Another four-set win, this time over Alexander Zverev, brought the total to 18.2 hours and +4.5 hours above the average of slam finalists. This put Thiem in the unenviable position of being in the top 3% of competition load when he faced 7-time Australian Open winner Novak Djokovic, who, by the way, had only taken 12.5 hours on court to get to the 2020 final.
Did this make any difference for the outcome of the final?
We can get some idea about that by looking at other examples of players who had intense journeys to get to the second week of a Grand Slam. In the table at the end of this post I have put together a selection of some examples, including Thiem’s. Thiem’s 2020 Australian Open put his competition load next to ones like Kevin Anderson’s 2018 Wimbledon and Rafael Nadal’s 2017 Australian Open runs. Notably, no player in this group won a title.
But these are only a few examples. And even if many examples showed a similar trend we are ignoring a big confound. Is high competition load contributing to losses? Or is high competition load merely a byproduct of being an underdog?
What we want to know is whether high competition load has an influence on match outcomes after accounting for the difference in ability of the two players who meet in a given round of a slam. In the terminology of a regression, if $w_{ij}$
is the probability that player $i$
beats player $j$
and $load_i$ is player $i$
’s cumualtive load, we can look at the effect of load with the following,
$$ logit(w_{ij}) = \beta_0 + \beta_1 logit(p_{ij}) + \beta_2 (load_i - \bar{load}) $$
where $p_{ij}$ is a pre-match prediction that is based on the ability of the two players and any other factor other than the event competition load.
This describes a simple logistic regression model, where the main difficulty is choosing $p_{ij}$. This could be taken from pre-match bookmaker odds for example. I’m going to base it off of my own player ratings.
Using the past 10 years of rounds from the quarterfinals and later, I’ve fit the load model and looked at the expected change in Thiem’s win chances at each of the final three rounds of the 2020 Australian Open. The historical pattern suggests that Thiem’s chances might have already been hurt going into the quarterfinal (an average of 12% drop) and really took a nose dive after his long battle against Rafael Nadal. From that point on he was expected to have 30-35% less of a chance than his player rating would predict.
It might be tempting to shrug our shoulders at such findings, thinking that players have little control over the time of a match. But this would be underselling the ownership players have over how aggressively they will play and how this can influence the time on court. For slam players who are thinking more than one match at a time, turning up the aggression in matches when it reaches an average length may be an important strategy for getting a slam title.
Player | Event | Round | Opponent | Won | Score | Match Hours | Cumulative Hours |
---|---|---|---|---|---|---|---|
Roberto Bautista Agut | Australian Open 2019 | R128 | Andy Murray | 1 | 6-4 6-4 6-7(5) 6-7(4) 6-2 | 4.2 | 4.2 |
Roberto Bautista Agut | Australian Open 2019 | R64 | John Millman | 1 | 6-3 6-1 3-6 6-7(6) 6-4 | 3.8 | 8.0 |
Roberto Bautista Agut | Australian Open 2019 | R32 | Karen Khachanov | 1 | 6-4 7-5 6-4 | 2.1 | 10.1 |
Roberto Bautista Agut | Australian Open 2019 | R16 | Marin Cilic | 1 | 6-7(6) 6-3 6-2 4-6 6-4 | 4.0 | 14.1 |
Roberto Bautista Agut | Australian Open 2019 | QF | Stefanos Tsitsipas | 0 | 5-7 6-4 4-6 6(2)-7 | 3.2 | 17.3 |
Kei Nishikori | French Open 2019 | R128 | Quentin Halys | 1 | 6-2 6-3 6-4 | 2.0 | 2.0 |
Kei Nishikori | French Open 2019 | R64 | Jo-Wilfried Tsonga | 1 | 4-6 6-4 6-4 6-4 | 3.0 | 5.0 |
Kei Nishikori | French Open 2019 | R32 | Laslo Djere | 1 | 6-4 6-7(6) 6-3 4-6 8-6 | 4.4 | 9.4 |
Kei Nishikori | French Open 2019 | R16 | Benoit Paire | 1 | 6-2 6-7(8) 6-2 6-7(8) 7-5 | 3.9 | 13.4 |
Kei Nishikori | French Open 2019 | QF | Rafael Nadal | 0 | 1-6 1-6 3-6 | 1.8 | 15.2 |
Marin Cilic | U.S. Open 2012 | R128 | Marinko Matosevic | 1 | 5-7 2-6 6-4 6-2 6-4 | 4.0 | 4.0 |
Marin Cilic | U.S. Open 2012 | R64 | Daniel Brands | 1 | 6-3 6-2 5-7 4-6 7-5 | 3.7 | 7.6 |
Marin Cilic | U.S. Open 2012 | R32 | Kei Nishikori | 1 | 6-3 6-4 6-7(3) 6-3 | 3.5 | 11.2 |
Marin Cilic | U.S. Open 2012 | R16 | Martin Klizan | 1 | 7-5 6-4 6-0 | 2.0 | 13.2 |
Marin Cilic | U.S. Open 2012 | QF | Andy Murray | 0 | 6-3 6(4)-7 2-6 0-6 | 3.0 | 16.2 |
Stefanos Tsitsipas | Australian Open 2019 | R128 | Matteo Berrettini | 1 | 6-7(3) 6-4 6-3 7-6(4) | 3.0 | 3.0 |
Stefanos Tsitsipas | Australian Open 2019 | R64 | Viktor Troicki | 1 | 6-3 2-6 6-2 7-5 | 2.7 | 5.6 |
Stefanos Tsitsipas | Australian Open 2019 | R32 | Nikoloz Basilashvili | 1 | 6-3 3-6 7-6(7) 6-4 | 2.9 | 8.5 |
Stefanos Tsitsipas | Australian Open 2019 | R16 | Roger Federer | 1 | 6-7(11) 7-6(3) 7-5 7-6(5) | 3.8 | 12.3 |
Stefanos Tsitsipas | Australian Open 2019 | QF | Roberto Bautista Agut | 1 | 7-5 4-6 6-4 7-6(2) | 3.2 | 15.5 |
Stefanos Tsitsipas | Australian Open 2019 | SF | Rafael Nadal | 0 | 2-6 4-6 0-6 | 1.8 | 17.3 |
Matteo Berrettini | U.S. Open 2019 | R128 | Richard Gasquet | 1 | 6-4 6-3 2-6 6-2 | 2.7 | 2.7 |
Matteo Berrettini | U.S. Open 2019 | R64 | Jordan Thompson | 1 | 7-5 7-6(5) 4-6 6-1 | 3.0 | 5.7 |
Matteo Berrettini | U.S. Open 2019 | R32 | Alexei Popyrin | 1 | 6-4 6-4 6-7(3) 7-6(2) | 3.6 | 9.2 |
Matteo Berrettini | U.S. Open 2019 | R16 | Andrey Rublev | 1 | 6-1 6-4 7-6(6) | 2.2 | 11.4 |
Matteo Berrettini | U.S. Open 2019 | QF | Gael Monfils | 1 | 3-6 6-3 6-2 3-6 7-6(5) | 4.0 | 15.4 |
Matteo Berrettini | U.S. Open 2019 | SF | Rafael Nadal | 0 | 6(6)-7 4-6 1-6 | 2.6 | 18.0 |
Andy Murray | French Open 2016 | R128 | Radek Stepanek | 1 | 3-6 3-6 6-0 6-3 7-5 | 3.7 | 3.7 |
Andy Murray | French Open 2016 | R64 | Mathias Bourgue | 1 | 6-2 2-6 4-6 6-2 6-3 | 3.6 | 7.2 |
Andy Murray | French Open 2016 | R32 | Ivo Karlovic | 1 | 6-1 6-4 7-6(3) | 1.9 | 9.2 |
Andy Murray | French Open 2016 | R16 | John Isner | 1 | 7-6(9) 6-4 6-3 | 2.7 | 11.8 |
Andy Murray | French Open 2016 | QF | Richard Gasquet | 1 | 5-7 7-6(3) 6-0 6-2 | 3.4 | 15.2 |
Andy Murray | French Open 2016 | SF | Stan Wawrinka | 1 | 6-4 6-2 4-6 6-2 | 2.6 | 17.8 |
Andy Murray | French Open 2016 | Final | Novak Djokovic | 0 | 6-3 1-6 206 4-6 | 3.0 | 20.9 |
Kevin Anderson | Wimbledon 2018 | R128 | Norbert Gombos | 1 | 6-3 6-4 6-4 | 1.9 | 1.9 |
Kevin Anderson | Wimbledon 2018 | R64 | Andreas Seppi | 1 | 6-3 6-7(5) 6-3 6-4 | 2.6 | 4.6 |
Kevin Anderson | Wimbledon 2018 | R32 | Philipp Kohlschreiber | 1 | 6-3 7-5 7-5 | 2.1 | 6.7 |
Kevin Anderson | Wimbledon 2018 | R16 | Gael Monfils | 1 | 7-6(4) 7-6(2) 5-7 7-6(4) | 3.5 | 10.2 |
Kevin Anderson | Wimbledon 2018 | QF | Roger Federer | 1 | 2-6 6-7(5) 7-5 6-4 13-11 | 4.2 | 14.4 |
Kevin Anderson | Wimbledon 2018 | SF | John Isner | 1 | 7-6(6) 6-7(5) 6-7(9) 6-4 26-24 | 6.6 | 21.0 |
Kevin Anderson | Wimbledon 2018 | Final | Novak Djokovic | 0 | 2-6 2-6 6(3)-7 | 2.3 | 23.3 |
Rafael Nadal | Australian Open 2017 | R128 | Florian Mayer | 1 | 6-3 6-4 6-4 | 2.1 | 2.1 |
Rafael Nadal | Australian Open 2017 | R64 | Marcos Baghdatis | 1 | 6-3 6-1 6-3 | 2.2 | 4.3 |
Rafael Nadal | Australian Open 2017 | R32 | Alexander Zverev | 1 | 4-6 6-3 6-7(5) 6-3 6-2 | 4.1 | 8.3 |
Rafael Nadal | Australian Open 2017 | R16 | Gael Monfils | 1 | 6-3 6-3 4-6 6-4 | 2.9 | 11.3 |
Rafael Nadal | Australian Open 2017 | QF | Milos Raonic | 1 | 6-4 7-6(7) 6-4 | 2.7 | 14.0 |
Rafael Nadal | Australian Open 2017 | SF | Grigor Dimitrov | 1 | 6-3 5-7 7-6(5) 6-7(4) 6-4 | 4.9 | 18.9 |
Rafael Nadal | Australian Open 2017 | Final | Roger Federer | 0 | 4-6 6-3 1-6 6-3 3-6 | 3.6 | 22.6 |
Dominic Thiem | Australian Open 2020 | R128 | Adrian Mannarino | 1 | 6-3 7-5 6-2 | 2.4 | 2.4 |
Dominic Thiem | Australian Open 2020 | R64 | Alex Bolt | 1 | 6-2 5-7 6-7(5) 6-1 6-2 | 3.4 | 5.7 |
Dominic Thiem | Australian Open 2020 | R32 | Taylor Harry Fritz | 1 | 6-2 6-4 6-7(5) 6-4 | 2.8 | 8.6 |
Dominic Thiem | Australian Open 2020 | R16 | Gael Monfils | 1 | 6-2 6-4 6-4 | 1.8 | 10.4 |
Dominic Thiem | Australian Open 2020 | QF | Rafael Nadal | 1 | 7-6(3) 7-6(4) 4-6 7-6(6) | 4.2 | 14.6 |
Dominic Thiem | Australian Open 2020 | SF | Alexander Zverev | 1 | 3-6 6-4 7-6(3) 7-6(4) | 3.7 | 18.2 |
Dominic Thiem | Australian Open 2020 | Final | Novak Djokovic | 0 | 4-6 6-4 6-2 3-6 4-6 | 4.0 | 22.2 |