Betting Masterclass Volume 13: Dan Weston on assessing player quality on the ATP Tour

Spanish Tennis Player Rafa Nadal
Rafa Nadal led the ATP Tour for these metrics in 2019...

In the latest sport covered by our Betting Masterclass, we asked our Tennis columnist, Dan Weston, to give his thoughts on some ideas to look at when assessing a tennis match-up in advance...

"We can see that Rafa Nadal, Novak Djokovic and Roger Federer - the three players remaining in the top ten from the traditional elite four - are still by some distance exhibiting better data in these areas than the remaining players on the ATP Tour."

Using metrics to assess general player quality

There are so many things to consider when looking at and pricing up a tennis match, but for this Betting Masterclass series, I wanted to discuss a few of the areas that I think are particularly important. In this first piece, I want to discuss how bettors can use either service hold/break opponent percentages or service and return points won percentages to be able to ascertain a ballpark quality of an individual player.

Of course, there are nuances to this, and caveats - such as how to translate performances and data in qualifying and Challenger events to the main tour, as well as opponent quality, which is an area we will discuss further in part two - but there is little doubt in my mind that these two metrics are very accurate at understanding the level of any player on the main tour.

Winning more points leads to holding and breaking more

The chart below shows the relationship between these two metrics (illustrating top 100 players playing a minimum of 15 ATP main tour matches across all surfaces in the 2019 season), and as you'll be able to see from it, as combined service/return points won rises, then on a general basis so does a player's combined hold/break percentages. This is very logical - it's hardly groundbreaking to suggest that if a player wins more service and return points, then their likelihood of holding and break becomes more likely as a result.

Betfair SRPW vs HB % Chart (Chart One) j.jpg

Elite three still deserve their reputation

I've highlighted a few players of some interest in this first chart.

We can see that Rafa Nadal, Novak Djokovic and Roger Federer - the three players remaining in the top ten from the traditional elite four - are still by some distance exhibiting better data in these areas than the remaining players on the ATP Tour.

We can also begin to assess some ballpark measures for working out player quality, with 110%+ combined service/return points won looking like a very good marker for a world class player, and anything above 115%+ for combined hold/break percentage being a similar marker for that particular metric. All of these three elite players fit into that bracket for both metrics.

Top 10-20 players also generally possess a combined service/return points won percentage in excess of 103% (combined hold/break above 105%) with some of the better members of the top ten, such as Daniil Medvedev currently, pushing closer to the numbers which the elite three can boast. From there onwards, we have a sliding scale going towards the 100% mark for both metrics, which is a mark of being around the world number 45.

Lower ranked players fall below this 100% mark in general - essentially on the whole, they lose more points, and therefore games, than the average opponent.

Identifying the outliers from the data

Injuries notwithstanding, it is rare for a player to possess numbers above 100% for either metric and not be ranked inside the top 50. This goes the other way as well - there aren't many examples of being below 100% and being consistently ranked inside the top 50. Quite a few top 50 players hover around these areas - Benoit Paire is a decent long-term example, as is Fabio Fognini more recently, as well as Nick Kyrgios at stages in his career.

Interestingly, all three above-named players could be described as being rather enigmatic, with a high peak level but also perhaps finding it difficult to give a consistent level of performance.

There are also several players with relatively high service points won percentages compared to their hold/break percentage - Milos Raonic, John Isner and Borna Coric are highlighted examples. The first two in particular fit firmly into the 'big server' bracket and a potential explanation for this discrepancy could be that despite some general dominance from a points won percentage, the tight nature of their matches means that a lot of their sets are won via a tiebreak (where each player has a 100% combined hold/break percentage in that set) or by a single break, such as via a 7-5 or 6-4 set.

Another fairly enigmatic player, Gael Monfils, is highlighted as an opposite example, and it would be interesting to dig a little deeper to try and understand why his combined hold/break percentage is so relatively high compared to his combined service/return points won percentage.

Service/Return points won a huge driver towards win percentage

Given that I have demonstrated that both metrics generally rise with player quality, I wanted to rubber-stamp this assertion by cross-referencing each metric with each player's win percentage.

Betfair SRPW vs Win % Chart (Chart Two)j.jpg

Here we can see that combined service/return points won percentage is a huge driver towards player win percentage with the aforementioned three elite players having the best data again for both metrics. There are a few slight outliers, such as Raonic, Dimitrov, Evans, Fucsovics, Humbert, Millman and Munar winning a lower percentage of matches than their combined service/return points won percentage would imply, while conversely, Kyrgios, Auger-Aliassime, Tsitsipas and Thiem have won slightly more matches than their combined percentage would imply.

A similar situation for combined hold/break percentages

Betfair HB% vs Match Win % (Chart Three)j.png

This final chart shows a very similar relationship to the previous chart, but this time between combined hold/break percentage and match win percentage - in general, as hold/break percentage rises, so does win percentage. Furthermore, we also see similarities with many of the same players being slight outliers. Variance, even over the course of a complete season, can still be sizeable.

Ideas for further research

With further analysis and research, it is possible to use these metrics for each player (on the relevant surface that the match is due to be played on) to derive an estimated win percentage for each player prior to each match. This win percentage can then be divided into 100 to derive a model price for each player, which can then be compared to market prices. These expected win percentages and, therefore, model prices can also then be adjusted, if required, for any other factors considered relevant to that particular match-up.

In my second piece of this Tennis Betting Masterclass, I want to move towards demonstrating how opposition quality has considerable impact on a player's general combined service points/return points won, before in the final third article, looking at some urban myths which many bettors may consider valuable metrics, but are actually quite overvalued.

***

Follow Dan on Twitter @TennisRatings

***

Be sure to check out the other volumes in our Betting Masterclass series, listed below:

Volume 1 - Ed Hawkins on Test Match Cricket
Volume 2 - Ed Hawkins on Twenty20 Cricket
Volume 3 - Ed Hawkins on how to bet on ODI Cricket
Volume 4 - Mark O'Haire on the football stats that don't matter
Volume 5 - Mark O'Haire on benefits of data and beating the closing price
Volume 6 - Mark O'Haire's perfect football punt checklist
Volume 7 - Steve Rawlings on how to make golf tournament bets
Volume 8 - Tony Calvin on how he makes racing profitable
Volume 9 - Kevin Blake on the importance of the skill of race reading
Volume 10 - Mike Carlson on betting on NFL
Volume 11 - Dan Fitch on getting the maximum from your darts betting
Volume 12 - Paul Krishnamurty on how to make your snooker bets pay

Dan Weston,

Discover the latest articles