Yesterday I shared a wild-assed idea proposing that USTA Texas consider a pilot program that uses local progressive league play to advance teams to a Sectional Championship. Specifically, this would be for Women’s 5.0 40+ which is a division that has struggled to attract participation not only in Texas but also nationally. The first and arguably most important question is how the top players are determined at the end of the season. That is closely related to the mechanics of competition that is ultimately selected.
I briefly considered the idea that this might be a great trial use case for the World Tennis Number (WTN). However, I don’t think that the in-season rating swings would be dynamic enough to insure that every player who registered has a legitimate chance of making the team that advances to Sectionals. The concern is if players are mathematically eliminated before the season begins, engagement might be low. Additionally, other matches played outside the progressive league would also influence each player’s WTN rating.
There is no feasible way that dynamic NTRP ratings can be used either. Theoretically, NTRP dynamic ratings could be used to group the players onto courts in order of strength each week. Unfortunately, it wouldn’t take enterprising captains much time at all to figure out that progressive league court placement is a pretty good way to approximate each player’s current dynamic NTRP rating. That is way too much information leakage.
That means that this hypothetical progressive league necessarily has to use the total games won as a metric for grouping players onto courts per the standings. The courts for week 1 could be set either by using WTN or randomly.
That brings us to the format and another opportunity to embrace another tennis innovation. Specifically, I am willing to have this progressive league play out each week as three short sets, advantage scoring with a 7-point Coman standard tie-break game at 4-4. In other words, each person placed on the court would play a short set with each of the other players.
Following the first week, the players should be grouped onto courts on the basis of the total number of games won up to that point in time. At the end of the season, the players are ranked strictly on the total number of games won. I am also envisioning that the pool of players would indicate their availability each week and then be grouped accordingly. That eliminates players from being unfairly boosted or penalized by substitute players.
So who wins? The players that win the most games would be the ones ultimately named to the team. However, in the progressive format, everyone will get a lot of match play at the local level in an underserved demographic. In other words… everybody wins.
Teresa,
One of the things implemented in TMX and part of the Competition Factory is a concept called “DrawMatic” which was created by Constantine Ananiadis when he was coaching at Oberlin. Rounds of play are generated dynamically based on who is available each day and ratings, regardless of whether they start as NTRP, UTR, WTN, ELO, are calculated dynamically between each round. Several college coaches have used this feature to track the performance of their players in matches which are not submitted to UTR/WTN. I’ve long argued that in fact having many types of ratings is entirely valid. TRN is cohort based, serves a targeted purpose, and has proven more accurate for US based players (since that is the only data they process) than UTR. Anyway, you are basically suggesting an scenario specific calculation which could just as well be a rating, dynamically calculated within and for your specific context…
I like this idea.