PREMATCH TRADING
I thought that I
would start of the technical part of the handbook with a chapter on pre-match
trading. After all – what can be better than locking in a profit before the
match starts?
For a
player’s price to move pre-match, the market needs to disagree with the opening
lines that the bookmakers introduce. The opening lines at a company like
Pinnacle Sports can be moved fairly easily with a few medium-large bets and the
limits tend to reflect that – they’re much smaller than nearer the match start
time. On Betfair it can be difficult to get a big trade matched very early but
if you are persistent and are a little flexible with your prices you should be
able to get a fair sum matched eventually, especially in a high profile or live
streamed match.
I feel that a lot
of tennis traders, punters and pundits are often perplexed about a player’s
propensity to drift and consider that it’s guesswork whether to take an early
pre-match price or wait until closer to the start of the match. However I’ve
done a great deal of work regarding price movements and I think that actually
the movement is highly predictable.
The market tends to disagree with a price
for several reasons.
The first reason is simply the names of
the two players.
For example – a big
name will always attract market interest. Caroline Wozniacki is a prime example
of that. At the recent 2013 WTA Brussels tournament, she started her first
match against Jie Zheng at 1.25. This price was insane considering her run of
horrendous form, when recently she had lost to Qiang Wang, Garbine Muguruza and
Stefanie Voegele, all of whom were ranked outside the top 50. Before the Zheng
match she had lost her last four outings, yet you could lay her at 1.25
pre-match! That was all because of her reputation.
In the men, Marcos
Baghdatis is a notable player. Not close to his previous ability, he is still
very often priced very low by the bookmakers due to his reputation. His record
in the last 12 months at the time of writing is 28 wins and 25 losses – if you
had blind backed him in all of those matches you would have achieved a terrible
Return on Investment of -18.6%.
Andy Murray on clay
is another example. His clay stats are very poor for a player of his rank and
as I’ve mentioned on the website several times he is a great value lay on the
surface.
The bookmakers know
they will attract attention just because of his name and therefore price him
very short. At the time of writing, based on Pinnacle Sports closing prices,
Murray has a career Return of Investment on clay of -16.1%, and in the last
year that figure is - 15.5%. When you consider that if you look at his record
across all surfaces instead of clay Murray has a positive Return on Investment,
it shows how short he is priced on clay.
The second reason
that a price changes is that modellers like myself with mathematical projected
hold models (based on service points/return points won or service hold/break
percentages)
realise that a certain price is value according to their models. In my opinion,
and I’m not alone in thinking it, this is the only way of effectively modelling
a starting price based on stats. Of course, a modeller then must adjust for
other matchup factors, but it is the only effective way of setting a base price
for a player to start with. These modellers tend to be either individual big
hitters or work with betting syndicates and they can move the market very
easily. Since I started using projected holds to predict starting prices, I’ve
noticed that I can accurately predict which players price will lengthen or
shorten in the market on the majority of occasions.
The final reason is
information. Sometimes a player leaks information on social media regarding an
injury or illness that they have picked up, or have not recovered from.
Sometimes they might just leak it to associates – who then can profit from that
information.
And of course, some
matches might have strong suspicions that they are fixed. I wouldn’t say that
there are many matches that are suspicious, but I would say that there are
some. It definitely does go on.
What always worries
me pre-match is when a player is clearly no value based on my projected hold
model, there is nothing leaked on social media yet their price keeps falling.
That’s when I get very suspicious.
I performed drift analysis of the 2012
season, and found some very interesting information.
In the WTA, if you
had opposed every player of the 659 that had drifted over 5% in implied odds
for level stakes, you would have lost 1.52% (Based on Pinnacle Sports closing
prices) of your stake on average each bet. So we can clearly see that the
market, on the whole, corrects itself well by the start of the match.
There were certain players that drifted
often…
Alberta Brianti
drifted over 5% in 8 out of her 25 matches and when drifting did so from an
opening price range of 1.66-3.21.
Julia Cohen drifted
over 5% in 7 out of her 11 matches and when drifting did so from an opening
price range of 1.99-3.66.
Eleni Daniilidou
drifted over 5% in 8 out of her 30 matches and when drifting did so from an
opening price range of 2.01-4.00 (and won in none of these 8 matches)
Marina Erakovic
drifted over 5% in 10 out of her 40 matches and when drifting did so from an
opening price range of 1.5-3.08.
Stephanie
Foretz-Gacon drifted over 5% in 7 out of her 20 matches and when drifting did
so from an opening price range of 1.55-4.64.
Jarmila Gajdosova
drifted over 5% in 8 out of her 26 matches and when drifting did so from an
opening price range of 1.53-3.85.
Edina
Gallovits-Hall drifted over 5% in 5 out of her 14 matches and when drifting did
so from an opening price range of 1.75-3.68.
Daniela Hantuchova
drifted over 5% in 11 out of her 47 matches and when drifting did so from an
opening price range of 1.26-2.60.
Jelena Jankovic
drifted over 5% in 13 out of her 59 matches and when drifting did so from an
opening price range of 1.16-2.42.
Mathilde Johansson
drifted over 5% in 7 out of her 27 matches and when drifting did so from an
opening price range of 1.43-3.97.
Anne Keothavong
drifted over 5% in 7 out of her 24 matches and when drifting did so from an
opening price range of 1.07-3.36.
Anabel Medina
Garrigues drifted over 5% in 12 out of her 49 matches and when drifting did so
from an opening price range of 1.43-3.37.
Arantxa Rus drifted
over 5% in 8 out of her 25 matches and when drifting did so from an opening
price range of 1.74-5.24.
Francesca Schiavone
drifted over 5% in 12 out of her 43 matches and when drifting did so from an
opening price of 1.27-3.68.
Anna Tatishvili
drifted over 5% in 9 out of 34 matches and when drifting did so from an opening
price of 1.25-3.78.
Elena Vesnina
drifted over 5% in 9 out of 31 matches and when drifting did so from an opening
price of 1.32-3.31.
Venus Williams
drifted over 5% in 8 out of 33 matches and when drifting did so from an opening
price of 1.2-2.55.
Jie Zheng drifted
over 5% in 12 out of 49 matches and when drifting did so from an opening price
of 1.45-3.05.
I’d split the majority of these players
into two categories. They are either occasional WTA
Tour
players with fairly low rankings or very low projected hold percentages (e.g.
Cohen, Rus, Tatishvili), or veterans (e.g. Hantuchova, Jankovic, Schiavone,
Williams) that are possibly not in favour with the betting market due to their
ability not being nearly as high as when they were at the top of their game.
In the ATP,
the drifting figures appear to show that the market still isn’t fully corrected
by the start of the match. If you had backed every ATP player that had drifted
over 5% in implied odds in 2012 (674 outcomes) for level stakes you’d have a
-6.48% Return on Investment (based on Pinnacle Sports closing prices). So even
taking into account any profit margin from the bookmakers, if you took the best
closing price on the players with odds shortening by over 5% pre-match
you’d be looking at reasonable profit.
As with the WTA, there were players that
drifted with regularity.
Igor Andreev
drifted over 5% in 8 out of 27 matches and when drifting did so from an opening
price of 1.3-3.40.
Pablo Andujar
drifted over 5% in 17 out of 48 matches and when drifting did so from an
opening price of 1.33-6.24.
Alex Bogomolov Jr
drifted over 5% in 11 out of 39 matches and when drifting did so from an
opening price of 1.41-7.98.
Tatsuma Ito drifted
over 5% in 7 out of 22 matches and when drifting did so from an opening price
of 1.55-7.48.
Igor Kunitsyn
drifted over 5% in 7 out of 23 matches and when drifting did so from an opening
price of 1.58-4.22.
Paolo Lorenzi
drifted over 5% in 7 out of 25 matches and when drifting did so from an opening
price of 1.83-6.58.
Nicolas Mahut
drifted over 5% in 9 out of 34 matches and when drifting did so from an opening
price of 1.32.6.60.
Eric Prodon drifted
over 5% in 5 out of 7 matches and when drifting did so from an opening price of
3.01-4.91.
Pere Riba drifted
over 5% in 5 out of 8 matches and when drifting did so from an opening price of
2.13-6.10.
Gilles Simon
drifted over 5% in 14 out of 66 matches and when drifting did so from an
opening price of 1.24-5.44
Joao Souza
drifted over 5% in 7 out of 14 matches and when drifting did so from an opening
price of 1.84-4.60.
Bernard Tomic drifted
over 5% in 11 out of 48 matches and when drifting did so from an opening price
of 1.20-3.61.
Viktor Troicki
drifted over 5% in 13 out of 48 matches and when drifting did so from an
opening price of 1.25-3.04.
Filippo Volandri
drifted over 5% in 12 out of 31 matches and when drifting did so from an
opening price of 1.62-5.04.
Looking at these
players it’s clear to see that the vast majority are journeyman Challenger Tour
players that sometimes play on the ATP Tour, and when they do they generally have
a very mediocre record. It’s worth noting that none of the young Challenger
Tour decent prospects made this list when stepping up to the ATP main tour.
Furthermore, as with the WTA, there are a lot of weak servers on this list
(hence the player’s projected hold will be low, leading to interest from
modellers).
Certain players
also shortened a great deal in the pre-match betting. The following players
shortened over 5% on a regular basis in 2012.
Brian Baker (29% shortened over 5% of
implied odds)
Thomaz Bellucci
(22%)
Julien Benneteau (22%)
Ricardas Berankis
(47% - small sample)
James Blake (30% - small sample)
Alexandr Dolgopolov
(22%)
Grigor Dimitrov (30%)
Ryan Harrison (21%)
Gael Monfils (21%)
Kei Nishikori (21%)
Sam Querrey (29%)
Andreas Seppi (21%)
I feel that with a
good projected hold model and the knowledge about which players are often
favoured by the market, you can accurately predict which players’ price will
change in the run-up to a match. Armed with this information will give traders
a fantastic chance to lock in profits before the match starts, and pre-match
punters the opportunity to get maximum value for their bets.
No comments