The rankings project, which is developing proposals for a modified rankings system for England, is continuing and the next phase is to develop a test platform to run modelling with ‘live’ data.
Following member consultation, Table Tennis England decided to retain a ‘relative’ system, as is currently the case, but with important modifications.
Since the results of the member consultation were published, UK Sport’s Sport Intelligence Unit has undertaken a modelling and simulation exercise, which has given the project team a robust basis on which to progress to the next stages.
As a result of this, we have established the following framework for further development:
- A relative points system, which UK Sport have tested against historical data sets to demonstrate its efficacy. This element of the rankings would be a ‘true’ (non-inflationary) ELO model, and no annual points reduction would be required. This would illustrate ability
- A model for bonus points in singles/tournament and team/league formats, with a draft framework for testing at the next stage. The bonus point element of the rankings would be time-constrained, capped (eg best x of y events), weighted by event, and illustrate form. Inactive players would slip down the rankings until their bonus points are zero and only the ELO element remains
- Time-limited protection of ELO points lost in order to avoid disincentivising play
- The ELO (ability), Bonus (form) and Protected points combined would produce the rankings
- Initial findings related to integrating age-groups and genders suggest that suitable competition patterns exist to successfully support the integration of age-groups into a single ranking list (ie combining Cadet, Junior, Senior & Veterans), but that it is unlikely to be possible to integrate genders without destabilising the integrity of the system.
This framework supports the agreed outcomes of the Rankings Consultation – to use a relative system as the basis for national rankings, to investigate whether greater integration in the ranking lists would be feasible, and to undertake simulation to consider how we could achieve a better balance between the various principles through variations in how we use bonus points; event weightings; the difference between win points and loss points; and periodic points deductions.
The next step is development of a working model (a ‘test platform’) within which we can undertake iterative testing of the variables and parameters with live data, including further definition of how to treat new players and inactive players, and potential integration or reflection of ITTF results and rankings.
This development work will be mapped, and the development time and cost prioritised against other Table Tennis England projects and commitments. This process will determine a timeframe for the remainder of the project.
At that stage, we will be in a position to explain how we intend to test the proposed framework and we will set out a further process of consultation with Table Tennis England members.