AT THE END OF THE FIRST SET

AT THE END OF THE FIRST SET

Clearly, especially in best of three set matches, winning the first set is a huge advantage as the winner only needs to win one more set to win the match.

But how much of an advantage is it?

If you did a survey of gamblers and pundits, you probably would get a variety of extreme answers from ‘almost guaranteed to win’ to ‘it’s just one set out of three’. I don’t deal in speculation so I have all the statistics necessary to gauge just how much of an advantage a first set win actually is.

WTA:-

Winning the first set

Win Percentage when won the first set with SP 1.01-1.20 = 96% (1.04 implied odds) Win Percentage when won the first set with SP 1.01-1.50 = 94% (1.06 implied odds) Win Percentage when won the first set with SP 1.01-2.00 = 90% (1.11 implied odds) Win Percentage when won the first set with SP 2.00-2.99 = 77% (1.30 implied odds) Win Percentage when won the first set with SP 3.00-5.99 = 63% (1.59 implied odds) Win Percentage for top 100 players when won the first set (any SP) = 87%

Losing the first set

Win Percentage when lost the first set with SP 1.01-1.20 = 48% (2.08 implied odds) Win Percentage when lost the first set with SP 1.01-1.50 = 39% (2.56 implied odds) Win Percentage when lost the first set with SP 1.01-2.00 = 30% (3.33 implied odds) Win Percentage when lost the first set with SP 2.00-2.99 = 15% (6.67 implied odds) Win Percentage when lost the first set with SP 3.00-5.99 = 11% (9.09 implied odds) Win Percentage for top 100 players when losing the first set (any SP) = 22%

Before I analyse the statistics, the eagle eyed amongst you may have noticed that the win percentage for top 100 players when won the first set is 87% but when losing the first set is 22%, and you are probably wondering why it doesn’t add up to 100%. That’s because sometimes, top 100 players play players outside the top 100…

From these statistics there are several areas that interest me. Using the midpoint of each price range (clearly this is not exact, but it should be pretty reasonable), we can say that a player that starts at 1.11 should have a price of 1.04 if they win the first set – that’s pretty much spot on. However, if that player loses the first set, would they go to 2.08? That’s incredibly unlikely in my opinion. I feel that in the best of three WTA match, they’d still be odds-on after losing the first set. This situation is almost certainly due to the weight of money supporting the favourite when they are losing – it’s a common entry point for many traders and is clearly illustrated in that example.

This can be further considered by the success of pre-match underdogs priced 3.00-5.99 when winning the first set. These players win on average 63%, from a midpoint starting price of 4.50. Would a player starting at 4.50 be 1.59 after winning the first set? That’s fairly unlikely, in my opinion. I feel that they’d probably be a little bigger than this – again lending weight to the argument that backing heavy favourites when losing is poor value due to the market forces keeping the price of the heavy favourite lower than it should be.

Therefore we can assume that backing the favourite in the WTA when a set down, applied as a blanket strategy, is a losing one. Having said that, I’m sure that used selectively it can produce profit.

I also compiled the same statistics for the ATP.

ATP:-

Winning the first set

Win Percentage when won the first set with SP 1.01-1.20 = 99% (1.01 implied odds) Win Percentage when won the first set with SP 1.01-1.50 = 95% (1.05 implied odds) Win Percentage when won the first set with SP 1.01-2.00 = 91% (1.10 implied odds) Win Percentage when won the first set with SP 2.00-2.99 = 74% (1.35 implied odds) Win Percentage when won the first set with SP 3.00-5.99 = 61% (1.64 implied odds) Win Percentage for top 100 players when won the first set (any SP) = 84%

Losing the first set

Win Percentage when lost the first set with SP 1.01-1.20 = 67% (1.49 implied odds) Win Percentage when lost the first set with SP 1.01-1.50 = 44% (2.27 implied odds) Win Percentage when lost the first set with SP 1.01-2.00 = 37% (2.70 implied odds) Win Percentage when lost the first set with SP 2.00-2.99 = 17% (5.88 implied odds) Win Percentage when lost the first set with SP 3.00-5.99 = 9% (11.11 implied odds) Win Percentage for top 100 players when lost the first set (any SP) = 25%

As we can see from comparing the WTA stats with the ATP stats above, favourites clearly enjoy a bigger advantage in the men’s game than they do in the women’s when winning the first set. However, overall we see that the women generally win more matches when a set up, due to the increased win percentages for underdogs. The two underdog price brackets in the ATP had a lower average win percentage than for the WTA – so men’s favourites tend to be able to come back and win in three sets more often than women’s favourites. The statistics on losing the first set back that up – men priced between 1.01 and 1.20 come back on average 67% of the time which is a huge increase on the women’s 48%. Not only that, but the win percentage price 1.01-1.50 was bigger for the men (44% to 39%), and it was too for the price 1.01-2.00 (37% to 30%).

I personally consider that men favourites tend to come back more than men’s because of the bigger gap in fitness, and to some extent mental strength. I truly believe that a great deal of tennis is played in the mind. Some men and women players are very, very weak in the mental strength department and also lack fitness, particularly in the third set. These deficiencies will prevent them from reaching the levels that their ability warrants. If you see a player who is strong or weak mentally or physically, make a note of that. It’s a note that could earn you a lot of money in the future…

Another factor that influences the market price after the first set is the actual first set scoreline. Logically, many people consider that the player that enjoys the bigger advantage in the first set should be so dominant that the second set is a mere formality.

However, that isn’t necessarily the case. We saw in the French Open Quarter Final on June

4^th, 2013, a prime example of that.

Serena Williams started her quarter final match with Svetlana Kuznetsova as heavy favourite at 1.07. She dominated Kuznetsova in the first set, winning it 6-1. Only the extremely optimistic Kuznetsova fan would expect Kuznetsova to be able to turn the match around but she almost managed it, taking the second set 6-3 and leading 2-0 in the third set. Kuznetsova actually traded as slight favourite at around 1.80 at that point. Clearly even with extremely heavy favourites winning the first set by a dominant margin, winning the match is far from a formality.

So as usual, several years ago I decided that having an opinion wasn’t enough and decided to investigate the statistics further…

I sampled a large number of mens and womens matches, with any starting price and player rank considered and obtained the following results:-

ATP:-

6-0 first set – win percentage = 92% 6-1 first set – win percentage = 86% 6-2 first set – win percentage = 86% 6-3 first set – win percentage = 86% 6-4 first set – win percentage = 79% 7-5 first set – win percentage = 84% 7-6 first set – win percentage = 76%

WTA:

6-0 first set – win percentage = 97% 6-1 first set – win percentage = 89% 6-2 first set – win percentage = 90% 6-3 first set – win percentage = 81% 6-4 first set – win percentage = 83% 7-5 first set – win percentage = 77% 7-6 first set – win percentage = 79%

From these statistics, we can see that generally the closer the first set score the lower the success rate is for winning the match, with there generally being a drop from the 6-0 to 6-3 scoreline in the ATP to the 6-4 to 7-6 scorelines, and the 6-0 to 6-2 scorelines in the WTA to the 6-4 to 7-6 scorelines.

If we notice an over-reaction in the market after a close set, especially in situations where a player gets broken at *5-6 or loses a tight tiebreak, we can often obtain good value because clearly in those situations the match win percentage for the first set winner will be at the lowest possible figures.

Another area we can look at when the first set is completed is a player’s record when a set down or up. Some players have a very negative mentality when they drop the first set, whereas it inspires others to come back. Some players get complacent when they drop the first set, or lack self belief that they can win. It’s important to know which players fall into the various categories.

The Australian legend Rod Laver had the highest career win percentage a set down, winning an incredible 48.4% of matches from that point. Out of the current players, Rafael Nadal’s

42.1% is the highest, and unsurprisingly he and the other current top 3 players make up the top 5 current players over the career stats, along with Lleyton Hewitt.

However, Hewitt’s inclusion here clearly shows the danger of looking at career stats. His career is clearly on the wane, and you can argue the same for Roger Federer to some extent as well. I prefer to look at the 12 month stats for match situations, personally.

Here’s some 12 month stats (correct at June 2013):-

Good players when a set up:-

Roger Federer 98.08%

Jo-Wilfried Tsonga 97.87%

Kei Nishikori 97.44%

Rafael Nadal 97.22%

Alejandro Falla 96.30%

Novak Djokovic 95.45%

Juan Martin Del Potro 93.02%

David Ferrer 92.86%

Dimitry Tursunov 92.86%

Fabio Fognini 92.59%

Tommy Robredo 91.43%

Leonardo Mayer 91.30%

Jerzy Janowicz 90.70%

Tomas Berdych 90.20%

Marcel Granollers 90.00%

Bad players when a set up:-

Carlos Berlocq 70.83%

Go Soeda 71.43%

Julien Benneteau 74.19%

Fernando Verdasco 75.00%

Martin Klizan 75.00%

Gilles Muller 76.00%

Jurgen Melzer 76.19%

Robby Ginepri 76.47%

Lukasz Kubot 76.92%

Ivan Dodig 77.50%

Edouard Roger-Vasselin 77.78%

Frederico Gil 77.78%

Steve Darcis 78.12%

Radek Stepanek 78.26%

Denis Istomin 78.57%

From these stats we can see which players thrive when a set up and which have a lot more difficulty converting a set advantage to a match win. Whilst it’s worth noting that many of those ATP players with good records a set up are top players, it’s not exclusively the case.

Clearly the list of players above who have bad records a set up will give you an idea of who it could be viable to lay when a set up.

What is also useful is to see which players do well when a set down, and which players tend to wave the white flag in those situations. Again, these are 12 month stats correct at June 2013.

Good players a set down:-

Rafael Nadal 70.00%

Novak Djokovic 61.90%

Andy Murray 55.56%

David Ferrer 50.00%

Ernests Gulbis 42.31%

Stanislas Wawrinka 40.91%

Richard Gasquet 39.13%

Jerzy Janowicz 38.46%

Juan Martin Del Potro 38.10%

Tomas Berdych 36.36%

Sam Querrey 35.71%

Daniel Brands 35.71%

Thomaz Bellucci 34.62%

Note: Roger Federer won only 31.58% of matches where he dropped the first set in the last 12 months. You may well wonder why I say ‘only’ when this is still above ATP average. However, it’s important to realise that his starting price will almost always be extremely low

– which cannot be said for the average player. Furthermore, in his career he has won 41.35% of matches where he lost the first set, so the 12 month figures signal a significant decline in his fortunes when in this situation.

Bad players a set down (12 month stats at June 2013):-

Alejandro Falla 3.57%

Fernando Verdasco 5.26%

Ivo Karlovic 5.56%

Bjorn Phau 6.67%

Michael Llodra 6.67%

Santiago Giraldo 8.70%

Jesse Levine 9.52%

Tobias Kamke 9.68%

Juan Monaco 10.00%

Albert Ramos 11.54%

Grigor Dimitrov 12.00%

Ivan Dodig 12.50%

Ryan Harrison 12.50%

Flavio Cipolla 12.90%

Radek Stepanek 13.33%

Robin Haase 13.33%

Jan Hajek 13.64%

Go Soeda 13.79%

Lukas Lacko 13.79%

Philipp Petzschner 14.29%

Roberto Bautista-Agut 15.00%

Adrian Ungur 15.38%

Xavier Malisse 16.00%

Nikolay Davydenko 16.00%

Pablo Andujar 16.13%

Whilst it’s definitely fair to say that most of the above players that perform badly when a set down are at the lower end of the ATP tour, there are some notable higher level players in this sample – the likes of Fernando Verdasco, Juan Monaco, Grigor Dimitrov, Radek Stepanek and Nikolay Davydenko are all much more illustrious names than most on the list. It would be wise to be cautious about backing these players when losing.

Naturally, I also performed the same research for the WTA, and here’s the 12 month stats

(also correct at June 2013):-

Good players when a set up:-

Greta Arn 100.00%

Sofia Arvidsson 100.00%

Serena Williams 98.53%

Victoria Azarenka 98.15%

Urzsula Radwanska 97.22%

Maria Sharapova 96.43%

Agniezska Radwanska 95.56%

Venus Williams 95.45%

Maria Kirilenko 95.00%

Ayumi Morita 93.10%

Elena Vesnina 92.59%

Ana Ivanovic 92.31%

Anastasia Rodionova 92.31%

Eugenie Bouchard 92.31%

Melanie Oudin 92.31%

Polona Hercog 92.31%

Jamie Hampton 92.00%

Bad players when a set up:-

Jarmila Gajdosova 50.00% (small sample of 10 matches)

Maria Jose Martinez Sanchez 58.33%

Anabel Medina Garrigues 65.00%

Olga Govortsova 65.22%

Yaroslava Shvedova 65.38%

Mathilde Johansson 66.67%

Ksenia Pervak 68.75%

Tsvetana Pironkova 68.75%

Francesca Schiavone 70.83%

Arantxa Rus 71.43%

Shahar Peer 73.68%

Misaki Doi 74.19%

Sabine Lisicki 75.00%

Vania King 76.19%

Andrea Petkovic 76.47%

Vera Dushevina 77.42%

Julia Goerges 78.12%

Heather Watson 78.26%

Lourdes Dominguez Lino 78.57%

Mandy Minella 79.17%

Dominika Cibulkova 79.31%

Jelena Jankovic 79.49%

As with the ATP, many of the players with a good record a set up are top 20 players.

However, it’s worth noting that Arn, Arvidsson, Bouchard, Morita, Oudin, Rodionova and Hercog are much lower ranked and I’d be dubious about currently opposing them when they take the first set.

Urszula Radwanska is an interesting case. She has a superb record when a set up, but is a highly inconsistent player generally, with the ability to test higher ranked players, but equally to lose in two comfortable sets to much lower ranked opposition. It would appear from stats, and also when I’ve seen her play, that she is a very confidence orientated player.

Some of the women’s records when a set up are horrific. Gamblers in particular speculate about some women’s players mental strength – the phrase ‘mental midget’ is often used, and clearly these statistics provide some evidence of that.

It’s very interesting to see that Carlos Berlocq was the worst ATP performer when a set up with a 70.83% winrate – however, that would have put him just equal eighth on the WTA list. Considering that the market usually prices women’s favourites as a shorter price (correctly) than the equivalent priced men’s favourites when a set up, opposing the names above with a bad record when a favourite should provide a positive expectation over the long term. Interestingly there were many high profile players in the list of poorly performing players when a set up that often start as a strong favourite – Schiavone (particularly on clay), Lisicki, Petkovic, Goerges, Cibulkova and Jankovic all regularly start odds on, and often short odds on. Opposing these players when a set up should be extremely lucrative.

Good players when a set down:-

Serena Williams 71.43%

Kirsten Flipkens 50.00%

Victoria Azarenka 45.45%

Lauren Davis 40.00%

Maria Sharapova 38.46%

Petra Kvitova 37.04%

Monica Puig 35.71%

Stefanie Voegele 35.29%

Vania King 34.78%

Svetlana Kuznetsova 33.33%

Venus Williams 33.33%

Lourdes Dominguez Lino 32.14%

Nadia Petrova 31.58%

Polona Hercog 31.25%

Bad players when a set down:-

Arantxa Rus 0.00%

Pauline Parmentier 4.00%

Coco Vandeweghe 5.26%

Greta Arn 5.56%

Sofia Arvidsson 7.69%

Jarmila Gajdosova 8.33%

Varvara Lepchenko 8.70%

Anastasia Rodionova 9.09%

Kristina Mladenovic 9.09%

Yaroslava Shvedova 9.09%

Alberta Brianti 9.52%

Shahar Peer 9.52%

Andrea Petkovic 10.00%

Maria Jose Martinez Sanchez 10.00%

Lucie Hradecka 10.53%

Marina Erakovic 11.11%

Tsvetana Pironkova 12.50%

Christina McHale 13.04%

Julia Goerges 13.04%

Kimiko Date Krumm 13.04%

Su-Wei Hsieh 13.64%

Laura Pous-Tio 14.29%

Mona Barthel 14.29%

Sara Errani 14.29%

Lucie Safarova 14.81%

Misaki Doi 14.81%

Yanina Wickmayer 14.81%

Marion Bartoli 15.00%

Daniela Hantuchova 15.15%

Madison Keys 15.79%

Bethanie Mattek-Sands 15.79%

Sam Stosur 16.00%

Considering the average for winning the match when a set down in the WTA is 22%, some of these figures are as horrific as the figures regarding win percentage when a set up. There would have to be huge value offered to want to back a player on this list when a set down. Again, there are several top players on the list – Goerges, Barthel, Errani, Safarova, Wickmayer, Hantuchova, Stosur all have a decent reputation and in some cases their name will guarantee a short price in the market, even when losing. If they start as heavy favourite, which they often do, and their price is around evens or a little bigger when a set down, it’s hard to justify backing them on the basis of these statistics.

The final thing we need to consider at the end of the first set is our staking. If, for some reason, we decide to back the winner of the first set, this is slightly less of a concern because there will be a point during the second set or third set where we can at least get some of our stake back. However, if we lay the winner of the first set, and then they take the lead in the second set, or get close to winning it, we have a problem because their price will be very short – hence leading to a large percentage loss of our stake. This is less of a problem if we have laid the winner of the first set at a very low price already, as our risk is much smaller compared to our reward. However, if a heavy underdog wins the first set, and we decide to lay them, we have a bigger problem.

For example – a heavy favourite starting between 1.20 and 1.25 loses the first set. Their price will, at that point, generally be around the 2.00 mark, give or take a few ticks either way. If you back them at this price, and the second set goes on serve, with no breaks, to a scoreline like 4-4, and the favourite gets broken, you are in severe danger of losing almost all your stake if you trade out when the underdog is serving for the match at *5-4. Whilst having a high level of risk isn’t necessarily a problem to experienced traders or those with a high risk tolerance, it’s vital that all traders, especially novices, are aware of the potential implications. As I mentioned previously – favourites that lose the first set do not always come back! This is something that in-play gamblers definitely need to bear in mind if they decide to take a longer term position at the end of the first set.