Vital to understand opposition quality
Understanding opposition quality is a critical area to consider when comparing two tennis players. While we looked at combined service/return points won percentage and combined hold/break percentage in part one of this series, which in my view are fantastic metrics to compare player ability levels, it's also worth factoring in opposition quality.
At this point, some readers might be thinking that this isn't really necessary, taking the view that throughout the course of seasons, or multiple seasons, everyone kind of plays each other and it all balances out. Such a point of view might be tempting, but it's actually quite a flawed mindset, as I will demonstrate below in several areas.
Challenger Tour players can struggle at ATP level
On a general basis, it is quite obvious that the easier the average level of opposition that a player faces then the higher chance they have to win a service or return point, and therefore win that match. As men's tennis players develop their career - often as young players - they must use the Challenger Tour as a stepping stone towards the main ATP Tour, and if they perform well on the Challenger Tour, they will start getting their ranking high enough to get themselves into a position where they can get direct entry for some smaller ATP 250 tournaments, and also the qualifying rounds for Grand Slam and Masters 1000 level events.
At this stage and beyond, these players are going to find that their opponents markedly improve, making life tougher for them in general on court. It's not unusual to see players who thrive on the Challenger Tour initially struggle to translate that level to the main tour, against better opponents, in higher profile tournaments and in front of bigger crowds.
Quantifying the gap between Challenger and ATP
This can be illustrated by the graph below, which shows the combined serve/return points won percentages for top 100 ATP players on both the main ATP Tour and also the Challenger Tour, across all surfaces in 2019 (minimum eight matches on the Challenger Tour):-

Adopting the same scale for both the x-axis and y-axis of this graph is vital, because it allows us to easily show the general grouping of players in the top-left hand corner, which is captioned as 'weak on ATP Tour, strong on Challenger Tour'. Note that there are no players in the bottom-right corner (strong on ATP Tour, weak on Challenger Tour), which is completely logical (why would a player perform better as standard of opposition increases?!) and also none really covering the top-right corner (strong on both tours), which again is logical because any player good enough to be strong on the ATP Tour wouldn't need to play on the Challenger Tour. The odd exceptional young player who is breaking through to the main ATP Tour (for example, by playing the first six months of the year in Challengers, and the second half of the year on the main tour) might fit into this category if we looked back throughout many years, though.
Of the players in this graph, only 10 had a combined serve/return points won percentage of greater than 100% on the main ATP Tour, but all recorded over this mark on the Challenger Tour against generally lower ranked opposition. So - to put it very basically - the majority of these players were below average ATP players, but every single one of them were above average Challenger Tour players.
The average player here increased their ATP Tour combined percentage by 7.4% on the Challenger Tour, and while this isn't necessarily the most accurate way of measuring the difference in standard - I use historical data covering all players playing both tours over a similar time period, for example - it should give readers a useful ballpark figure.
Higher-ranked players tend to face better opposition
Another way we can measure opposition quality is to look at how various types of players perform against different ranking brackets, and to make an assessment as to how often they have to play each bracket as opponents. It's certainly not a perfect way to measure opposition quality, given players' ability levels sometimes differ wildly from one surface to another, but it's a very useful general guide as well as being a very time-efficient one for bettors to use.
On a basic level, there's a trade-off for top-level players. While their ranking often assures them of seeded player status and the benefits that come with this, such as playing an unseeded player in the first round of a tournament, and/or first-round byes (which as a side note, I find an extremely unfair concept - how can it be right that a player needs to win fewer matches to win a tournament than another?), top players also have the downside that they have to play higher ranked players more often.
This phenomenon is created because, as is logical, other top players also get to the business end of tournaments more often, and therefore play each other on a fairly regular basis, at least from a ranking bracket perspective - for example top ten players playing rival top ten players more often than players outside the top 50 face top ten opposition. Top players also participate in ATP 250 level tournaments less frequently than lower ranked players, and therefore get less exposure to lower-ranked opposition on average (compared to, say a Masters 1000 tournament, or a high quality ATP 500 with numerous top 20 players competing in it).
This discussion is illustrated by the numbers in the chart below:-

From this, we can see that if we look at matches across the last 12 months (to April 24, 2020) for two ranking brackets - top ten players and those ranked 41-50 (around the 'average' player at ATP level) we can see that top-10 ranked players play around three times as many matches against top-10 opposition on average, compared to those ranked 41-50. They also play more matches against 11-20 ranked players, in addition to facing players ranked 21-50 more often as well. The tide turns when looking at opposition outside the top-50, which rank 41-50 play considerably more than top-10 opposition.
Opposition quality has huge impact on underlying data
Finally, it's worth discussing the effect of opposition quality on a player's hold/break data. As an example, here is some analysis of Novak Djokovic over the last 12 months, with the graph below illustrating his hold/break percentages against the four ranking brackets detailed in the chart above:-

Here we can see a marked difference between Djokovic's hold/break percentages against top-10 players (bottom-left corner) compared to against other rankings (top-right corner). His break opponent percentages rose as the ranking brackets increased, but what is also interesting to see is that he also held well in excess of 90% against all of the three non-top-10 ranking brackets. Essentially, when not facing top-10 opposition, Djokovic produces service numbers that elite servers such as John Isner would be proud of. In conjunction with breaking opponents in these ranking brackets in the mid-30s% region, it isn't difficult to see why non-elite players find it incredibly difficult to beat the world number one.
Naturally, it is possible for bettors to assess this information for any player on tour, and this would be a very useful exercise for any bettor, particularly in conjunction with assessing which players have played a higher or lower percentage of matches against high or low ranked opposition compared to rival players of a similar rank to themselves. Understanding this type of data would allow bettors to understand current ability ceilings for players and of course, to understand how given player's hold/break numbers vary based on their quality of opposition - it would also be valuable to profile players who struggle to have a high peak level but consistently produce high levels against players worse than them.
***
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
Volume 13 - Dan Weston on assessing player quality on the ATP Tour