It is often said by the media that if a tennis player is
going to succeed in a Grand Slam event, they need to avoid playing 5 set
matches. The obvious implications of
playing a 5 set match is that the player faces more mental and especially
physical strain in these gruelling encounters.
The most notable of these has to be the match between
John Isner and Nicolas Mahut at Wimbledon in 2010 which ended 70-68 to Isner in
the final set, in a match spanning 3 days if my memory serves me
correctly. Unsurprisingly, Isner lost
his subsequent second round match to Thiemo de Bakker (a player he’d have
probably been considered a favourite against) only winning 5 games in 3 sets.
Several players seem to have a propensity to play 5 set
matches more than others. Notable
mentions here should go to Almagro, Federer, Isner, Kohlschreiber, Petzschner
and Wawrinka, who regularly appeared in my research. Indeed, it has been discussed in the media
that Isner’s poor record in Grand Slam events is due to the fact that he
regularly plays 5 set matches, due to the often tight nature of his matches.
So what I thought I’d do is to assess the impact of
playing a 5 set match on the winning player’s subsequent match, to see whether
either backing or laying these players would work well. Would the market take into account the fact
that the player had previously played an arduous match, or not? My sample was
all 4 Grand Slams, from 2009-2012.
In the 4 years, there were 365 outcomes where the player
playing had won a 5 set match in the previous round. Out of those 365 matches, only 143 players
won after winning a 5 set match in the previous round, a win percentage of
39.2%. Clearly this is an interesting
stat in itself, although further research was required to analyse an edge.
If I was to stake £100 on every player who had won a 5
set match in the previous round, I would have been £6,661 down overall (based
on Pinnacle closing prices). This
obviously is a horrendous result, and the return on investment of -18.2% is
terrible. At this point we can draw a
clear conclusion that blanket laying of every player that has won a 5 set match
in the previous round is a winning strategy.
This is fantastic in itself, but what I then did was
split the results into 2 categories: top 50 players and out of top 50
players. The results were even more
conclusive.
There were 218 outcomes where a top 50 player played a 5
set match the previous round, and they won 110 times (50.5%). Outside the top 50, there were 147 outcomes
but only 33 victories (22.4%). Here we
can see a clear distinction between how players of different ranks deal with
fatigue. The higher ranked players can
seemingly deal with this much better and there was less of an edge laying these
players in these situations.
In the 218 outcomes for top 50 players, I would have lost
£1848 (ROI -8.5%), but with the players outside the top 50 I would have been
£4813 down with an absolutely horrific ROI of -32.7%.
So the conclusion we can draw from my research is that it
is highly profitable to lay all players outside the top 50 in a Grand Slam
match after they won a 5 set match the round previously. It is profitable to lay top 50 players too,
but the edge is much smaller.
Update: Australian
Open 2013
The Australian Open proceeded in the same vein as
previous Grand Slams that were included in the sample. In fact, this trend was even more prevalent
with there not being a single player ranked over 50 winning their subsequent
match after winning a 5 set match, throughout the whole tournament!
The results for the matches where a player outside the
top 50 won a 5 set match in the previous round are as follows:-
Round of 64:-
Nieminen (SP 1.49) lost in 5 sets to Dodig.
Duckworth (SP 3.26) lost in 5 sets to Kavcic.
Roger-Vasselin (SP 5.06) lost in 4 sets to Benneteau.
Gimeno-Traver (SP 9.58) lost in 3 sets to Almagro.
Ito (SP 10.55) lost in 4 sets to Baghdatis.
Bautista-Agut (SP 2.28) lost in 5 sets to Fognini.
Baker (SP 5.84) was forced to retire during his loss to
Querrey.
Round of 32:-
Dodig (SP 9.00) lost in 4 sets to Gasquet.
Kavcic (SP 25.87) lost in 3 sets to Tsonga.
Monfils (SP 2.62) lost in 5 sets to Simon.
Donskoy (SP 6.74) lost in 3 sets to Nishikori.
Whilst it’s fair to say that the majority of these
players were expected to lose based on their starting prices, with only one
player as favourite, it’s undeniable that players ranked outside the top 50 are
at a huge disadvantage in the subsequent round following a 5 set victory.
Regarding players in the top 50, whom you may recall have
a lower return on investment (but still heavily negative) when laying them
after a 5 set match, the results were as follows:-
Round of 64:-
F Mayer (SP 2.13) lost in 3 sets to Berankis.
Almagro (SP 1.10) defeated Gimeno-Traver in 3 sets.
Youzhny (SP 1.34) lost in 5 sets to Donskoy.
Baghdatis (SP 1.08) defeated Ito in 3 sets.
Verdasco (SP 1.61) defeated Malisse in 3 sets.
Stepanek (SP 2.09) defeated Lopez in 3 sets.
Round of 32:-
Seppi (SP 3.11) defeated Cilic in 5 sets.
Tipsarevic (SP 1.38) defeated Benneteau in 5 sets.
Janowicz (SP 3.31) lost in 3 sets to Almagro.
Melzer (SP 12.17) lost in 3 sets to Berdych.
Round of 16:-
Simon (SP 24.94) lost in 3 sets to Murray.
Seppi (SP 1.59) lost in 3 sets to Chardy.
Chardy (SP 2.59) defeated Seppi in 3 sets.
Tipsarevic (SP 2.23) retired during his loss to Almagro.
Anderson (SP 4.84) lost in 3 sets to Berdych.
Quarter Finals:-
Djokovic (SP 1.23) defeated Berdych in 4 sets.
Semi-Finals:-
Federer (SP 2.30) lost in 5 sets to Murray.
Ferrer (SP 8.04) lost in 3 sets to Djokovic.
Final:-
Murray (SP 2.73) lost in 4 sets to Djokovic.
Assessing these matches, I applied stakes of £100 to back
each player, and I wanted to see if this would produce a similar negative
Return on Investment as witnessed previously.
Out of these matches, 8 players won and 11 were eliminated (a win
percentage of 42.11%), which is below the previous 50.05% success rate for
players inside the top 50.
If I had backed each player for £100 based on closing
prices, I would have made a loss of £481 from an investment of £1900 (-25.32%
ROI). Clearly again, this would be a
terrible result and from this we can also assume that laying these top 50
players after winning 5 set match in the previous rounds would prove lucrative.
