The Second Serve Podcast ran an episode that shared “solutions” to NTRP Ratings sandbagging last August. From comments that the hosts made subsequently, I know some pushback was received on those ideas. That isn’t a surprise. Sandbagging is one of those hot topics of USTA tennis. It generates passionate opinions across the spectrum. Today I start breaking down the ideas shared on that podcast. It is a great illustration of how we need to shift our perspectives on NTRP leveled competition.
Before I get too deep into that topic today, I want to make it clear that I am a fan of the Second Serve Podcast. The Fiend at Court blog fundamentally speaks to the same target audience, recreational tennis players who want to maximize their competitive level. Additionally, the Second Serve Podcast has released episodes where I found myself engaged in topics that I am not ostensibly interested in. That is a credit to their interview and conversational skills. They consistently generate entertaining and informative content for us tennis nerds.
In the “Our Solution to Sandbagging” episode, hosts Carolyn Roach and Erin Conigliaro are joined by Carolyn’s husband Mike to discuss sandbagging. Their opening thesis is that the vast majority of sandbagging would be eliminated by implementing a rule that players who are “self-rated” or with an “appealed” NTRP rating should be ineligible for post-season play. That includes local playoffs and whatever progression method their USTA Section has implemented for advancement to Nationals.
It would have been helpful had that episode first articulated the problem that they were actually trying to solve. Additionally, a definition of sandbagging would have also been helpful before launching into solutions. Those are the fundamental reasons that their proposal did not resonate with me. I consider the ongoing issues with self-rated players to be one of initialization rather than sandbagging. Additionally, players with appealed ratings is a completely separate topic to be tabled for later this weekend.
My working definition of sandbagging is “intentionally concealing one’s true skill to create an advantage in competition.” When tennis players first enter (or re-enter) the NTRP ecosystem, the computer algorithm requires declaration of a starting initial value. The USTA’s current method for determining a player’s first NTRP rating is the self-rating process. It is far from perfect, but quite frankly there probably isn’t a better solution that scales as effectively.
When a player goes through the self-rating process, there are three fundamental outcomes and two massive caveats. The outcomes of the self-rating can be too high, too low, or the Goldilocks “just right.” The fundamental thing we need to remember about “just right” is that it is wholly dependent on our perspectives.
A USTA League captain who has a new self-rated player on their team that loses all matches played, will likely believe that the self-rating was too high. In my experience, people who lose to a self-rated player almost always immediately jump to the conclusion that the self-rating was too low. “Just right” is seldom held by all stakeholders simultaneously about the same self-rated player.
Additionally, the USTA League regulations caution that “athletic” people might improve rapidly resulting in disqualification. That is one situation where the self-rate could be “just right” initially, after which the player makes significant progress throughout the year. Tennis is arguably the only sport that punishes people for quick mastery. That is one of the two big caveats, an initially valid self-rating can dynamically skew.
The second caveat is that tennis is a game with high variance. A player can play significantly above level one day only to play dramatically lower the next. A single match doesn’t mean that the self-rate was egregious one way or another. In recognition of that fact, the USTA has implemented a three strike threshold for disqualification. For the sake of transparency, I personally believe that dynamic disqualification should be eliminated. The previous post “It’s Time to DQ Disqualification” outlines my reasons.
In addition to the outcome of the self-rating process, there is a critical second dimension of how we intellectually process self-rated players in general. As an established playing community, we can choose the assumptions we make about the motivations and intent of self-rated players. Those choices drive player experience for everybody.
We can choose to believe that any “errors” in the self-rating process were either driven by an honest mistake or malfeasance. I choose to believe that the USTA self-rating process is extraordinarily challenging and that any mistakes will work themselves out over time as ratings are updated.
This is where the “Yeah, but…” response typically comes in about the “ridiculous” self-rated player that someone lost to in post-season play. Those complaints miss the point. That is actually how the progression system is supposed to work. The top teams advance through successive tiers of competition until one is crowned as National Champion. At some point along the way, every team except the champion plays someone who is just a little better.
As a participant in the tennis ecosystem, players are entitled to compete against ever increasing levels of competition as long as they keep winning. Winning in and of itself isn’t an entitlement. Unfortunately, USTA League play has built a culture where the majority of players seem to regard it that way. That is a significant root cause issue for the USTA to solve.
I didn’t even get past the two minute introduction of the “Our Solution to Sandbagging” podcast in this post. It is apparent this entire “Unplugged” weekend will continue to riff on that theme. To close out today, I will observe that I do think that dishonest self-rating is an issue for the USTA. However, a much higher priority for the organization is to be as welcoming and inclusive to new players as possible.
Excluding self-rated players from the full competitive process is the exact opposite of welcoming new people into the competitive ecosystem. It also makes it a challenge to field teams at all in underserved tennis communities. We should be focused on building participation to the point where self-rated players are an insignificant minority.
The full scope of sandbagging is much more insidious than an error here and there in initialization. My fundamental reservation about this podcast episode is that by narrowly focusing on the initialization aspects, the most critical issues with sandbagging were overlooked entirely.
Buckle up. We’re taking another lap around that track tomorrow.
- Our Solution to Sandbagging, Second Serve Tennis Podcast, Episode 125, August 17, 2022.
I have read many of your posts and think they are great. I don’t always agree but I love the perspective you have as a high-level player. After reading many comments from listeners of Second Serve and from you, we still think our “sit out one year from post-season play” makes sense. We are even willing to adjust that idea… maybe it’s X amount of matches before they get a computer rating.
If a player that is just getting back into tennis self-rates or appeals, or if a player is just starting to play and has to self-rate, we don’t think sitting out of post-season play for 1 year (or some other timeframe) is a big deal in their “tennis career”. If the intention is to play tennis long-term, a small window “penalty” is a just a small blip in time. It’s not a perfect solution by any means but at least on-level players won’t pay money to travel to state, district, sectional, and national tournaments and not have a fair advantage by, very likely, facing multiple sandbaggers.
One of the arguments for the World Tennis Number (WTN), with the ability to eneter results of any matches, is that more data will bring about better and more reliable ratings.
In our seniors leagueteh match format is three courts of doubles where the courts are ranked in order of strength. However, the league struggles, despite having movement rules in consecutive matches, to control ‘sandbagging’ where a court 1 pair manoeuvres to play on court 3 in a crucial match, effectively throwing court 1 in an attempt to win courts 2 and 3.
One of the issues is a team can be better than two good individual players and so ranking by player ofetn do not correctly define teh strength of a team.
I undertand the USTA did have a rule about ranking of players in strength order – but this was abandoned after proving it was not possible to police.
Another factor in matches between players rated differently, is often players match up affects the outcome of a match. Some players do better against certain opponents. Most notably this can happen when a ‘lefty’ plays another ‘lefty’. When a ‘lefty’ plays a ‘righty’… which happens most of the time, they can develop quite a high rating – but when they face a nother ‘lefty’, the rating often doesn’t count for much!
Your comment touches on a couple of post ideas that I have been carrying on my prospective topics list for a while.
The first topic is straight up how the requirement to order doubles teams on order of strength was impossible to police and only benefitted the cheaters.
The second topic is that there is no transitive property of tennis. Due to matchups, player A beats player B who beats player C, who beats player A on a consistent basis. Or mathematically A > B > C > A, which simplifies to A > A, which is obviously invalid and proves that there is no transitive property of tennis.
The “Appeals” post from a couple of days later generated some interesting comments on the FiendAtCourt Facebook page that merit follow up. Maybe I will hit those ideas as a trio November 12-14 which is the first place on my publishing calendar to squeeze them in.