In July 2019 the LTA and the ITF announced the World Tennis Number project. Local Tennis Leagues has been connected to the project since its earliest days. We believe WTN offers a huge opportunity to grow the sport and that it could have a profoundly impact on the tennis you play.
WTN is expected to go live in 2021
The basic scale has now been revealed. It will be from 40 to 1, with 1 being the highest standard. This scale will be displayed with 1 decimal place as a player progresses, for example, 32.2, 20.4, 15.7, 6.9.
Here we answer some of your questions and provide links to where you can find out more about the project as it takes shape. This page will be updated frequently as the official launch date approaches. (Please note all answers are our interpretation and subject to clarification as more information about the project is released.)
Introducing WTN, the ratings game changer - how LTL revealed the WTN project in July 2019
LTA/WTN FAQs - Frequently asked questions answered by the LTA
ITF/WTN FAQs - Frequently asked questions answered by the ITF
Your questions answered
Please use the contact form to submit a new question. We are particulary concerned to address questions about the opportunities the WTN brings to Local Tennis Leagues and its players.
What's the difference between traditional ratings and this new rating?
Traditional LTA ratings measure a standard of play attained and below a certain level, your "standard" as measured by the rating doesn't fall. The WTN is intrinsically dynamic. Theoretically, it goes up (and down) with every set of tennis you play and it is relative to everyone else playing. We have learned to think of it as not a measure of finite achievement but a measure of your real standard (relative to others) that gets more accurate the more you play.
What is the algorithm that is used to create the number?
Local Tennis Leagues has been trialing an algorithm behind the scenes since the summer of 2018. The algorithm used by the WTN has not yet been published, but it will be similar to the one we have used. Our algorithm is a variant of that used by the Glicko rating system and went through many iterations before we settled on the most accurate. It is our understanding the principals behind the WTN calculations will be published in full nearer to launch, but it too will have been extensively tested against an enormous amount of tennis data.
Can you give an example of what the WTN will look like for players receiving one? It sounds like a 'ranking number' but is referred to as a rating
The new rating will be from 40 to 1, with 1 being the highest. This scale will be displayed with 1 decimal place as a player progresses, for example, 32.2, 20.4, 15.7, 6.9. So rather than a 9.1, it will be three digits long (and behind the scenes possibly even longer). It will change frequently. A ranking is very much for players competing in tournaments whereas a rating is for everyone across the sport and can cover all types of competition (box leagues, team events etc). The WTN is a dynamic "snapshot" representing at any given moment a player's standard relative to other players, not a measure of where you are in a specific points race.
I understand that the WTN will work at set level rather than game level. Doesn't this make it less accurate as it won't distinguish between a set lost 6-0 or 7-6?
In our own study undertaken by Imperial College London, we found a set-level algorithm as accurate as game level approach. In addition, there are practical difficulties about working at game level. We hope to be able to expand on this answer in due course.
Will the WTN be relevant to the average, forty-something recreational player?
Yes... because of the huge amount of data from all types of tennis players, the aim is that the WTN will have relevance to all players whatever their age or gender.
At our level, people's performance levels can change dramatically over the months (either improving as we play more, or taking a big dip when life gets in the way...). How will the algorithm take account of that. A glorious win two years ago should presumably not count for as much as a scratchy loss last week?
Definitely the more you play, the more accurate your WTN will be. But because it incorporates an "uncertainty factor", one off results where there is sparce data, won't disproportionally skew the result. Similarly, the algorithm adjusts for likely and unlikely results. In our test algorithm an unexpectedly good result is likely to have a greater effect on your number than an unexpected loss. Good wins from several years ago slowly become history and numbers are updated to reflect matches played more recently by you and the people you have played.
How will the WTN be used in the leagues and will it be compulsory?
We have been using our own trial version as an additional "advisory" tool for creating groups for over a year. Our intention is to use the WTN as one of the principal factors in creating groups in the future, while preserving the principal that group winners earn promotion to a group above them. We will be consulting widely with players about this in the months to come and as a part of this making a decision about whether to make WTN a necessary part of league membership.