Purestriker
Legend
Those are matches, so they would be more than 2.
The dreaded EOY Rating thread
- Thread starter Brilliant
- Start date
Moon Shooter
Hall of Fame
I didn't say an 11 is normal. But it is adjusting from his 3.7. Obviously something was way off to start with some players WTN rating. I think the strength of 3.5 players varies from area to area. In my area there are no 3.5c male rated players in the 30s. Most of the higher rated 3.0 rated players are about 27. One 3.0c rated player is a 25.0 but I don't think that is WTN's fault I think it is surprising that USTA gave him a 3.0 rating. I self rated as a 3.0 and the WTN self rates are assigned is 29.9.
Most 3.5 rated male players are mid to lower 20s in my area. Yes an 11 is an outlier and he will remain one for some time due to the insane starting rating. But the vast majority of players in my area have a fairly consistent WTN number that accurately shows their rating.
If they self rated as 3.0 they would have simply been assigned a 29.9 rating to start. I am not sure how sticky that initial rating is (what the k factor is for early matches) but it is what they do. I had a 6 UTR as a 3.0 player.
Just because the numbers are different that does not mean WTN is the garbage rating.
Seems like there would be some type of districts to play first?
But I guess this makes sense for an area that doesn't do districts.
Moon Shooter
Hall of Fame
Did you play USTA tennis before 2019? It seems many of the people that have a messed up rating seemed to have played in 2018 and/or earlier and the rating started off base and just remained off base. It is possible there is something wrong with the algo but these algorithms have been around so long I would be surprised. I think the problem cases - and there are plenty - stem from how they are assigning initial ratings and generally getting all these players into the WTN pool of players.
This particular issue is not a problem for UTR because it has such a short memory. But I believe after these start up issues get sorted out WTN will be recognized as an obviously superior system for adult rec tennis players.
I think the fact that players are allowed to self-rate can really play into an out of balance utr rating to ntrp comparison.
EDIT: Which also supports the idea of the USTA taking a look at players records in tournaments and league before the end of year.
Last edited:
I might have played in a handful of matches those two years as I was just getting playing again. I did play in a couple of USTA events. Really started playing league 2021 fall season through now.
CiscoPC600
Hall of Fame
Do matches for prior year that didn’t make the cutoff for 2022 EOY count towards the 2023 EOU?
Moon Shooter
Hall of Fame
Those few early matches have thrown your rating out of whack.
On the WTN site (not the USTA site) you can go to "results" and see what rating you had going into each match. So you can see how your matches effect your rating.
I would be interested in knowing if your early rating seemed even further off base (then your current rating) and therefore your rating is slowly getting more accurate/sensible. In the cases where I know someone with a rating that is way off base they had that really off base rating back in the earlier years and it has been taking a long time to adjust. But it seems people that started in in the past 2 years or so with no earlier ratings have ratings that are very sensible. But this is just my anecdotal observation I wonder if other people are seeing the same thing.
I am going to copy this over to Schmke's WTN thread.
Which Section are you in? I bet I can find 3.5s with WTNs => 30.
I just randomly selected four M3.5 players from a local team. Their WTN.d ratings are: 29.7, 30.7, 34.9, 30.0. This team finished 2nd to last in the Fall 2022 season.
Here is a random sample of four M3.5 players from a local team that finished in 2nd place in Fall 2022: 25.3, 28.7, 30.1, 30.7
Last edited:
Moon Shooter
Hall of Fame
That is interesting. If someone self rates as a 3.0 it seems they get a 29.9 rating - I did.
I am in the mid-south western district of Illinois.
Just randomly looking through those I played with and ended up 3.5c the worst wtn seems to be about 27.7 doubles but he had a 19 something in singles. Typically the M 3.5 c players are about 25 WTN at doubles lower at singles if they have a singles rating. If they are pushing 4.0 or just got bumped then I think they are mostly around 20 flat with their singles and double rating averaged.
schmke
Legend
No. You need three matches to get a new year-end rating.
I don't know if this is the case for @CiscoPC600 but if you played in an early start league and qualified for Sectionals already, you could play just Sectionals in 2023.
Oh that's right. You don't actually have to play in districts to play at sectionals.
Moon Shooter
Hall of Fame
So if someone ended their first year after 3 rated matches with a 3.49 dynamic rating ( hence a 3.5c) and then in their second year had two matches where they performed at like a 3.88 level they would stay a 3.5c for their third year? Even though their dynamic rating would be well above 3.50?
schmke
Legend
Yes. They would not get a new 3.5C for the second year, their 3.5C from the first year would simply remain the most recent published rating. A new rating is published only if you play at least three matches.
Moon Shooter
Hall of Fame
How odd. But you think their dynamic rating would go up correct? So when people play against them (in the second match of the second year and the third year) it is ratings wise like they are playing a 4.0.
The dynamic rating would likely increase unless the two "3.88 level" matches were played with or against self-rated players. But, it is possible those two matches would not increase the dynamic rating above the 4.0 threshold. It depends on the outcomes of the previous two matches. Regardless, since only those two matches were played in the new year, they would not affect the year-end rating since three qualified matches are required.
Moon Shooter
Hall of Fame
With any sensible k factor two 3.88 performance rating matches would cause a rating based on three previous matches to increase from a 3.49 to above 3.51. I don't understand why USTA does not just let the math work for them instead of making special rules to screw up the ratings. This along with throwing out so much data is why USTA ratings are not as accurate as they should be.
CiscoPC600
Hall of Fame
It’s been said before here but the fact that UTR hasn’t either makes it fairly obvious it’s USTA that is preventing data from being scraped.
Playing HS tennis doesn't mean you're an automatic 4.0. It means you're above 3.0. Competitive college play is quite a different matter. Also, the self-rate quiz takes this into account. I think anyone who played any sport at the varsity level in HS automatically starts at 3.5. I know I did.
There are plenty of high school players who are no better than 3.0. LOL. When I played in high school, there were multiple matches where our whole lineup didn't drop a single game. Those kids weren't anywhere even close to 3.5 much less 4.0. College is a little different. Almost all D3 and higher players regularly in the varsity lineup will be at least 4.5. There are exceptions for really weak teams in bad conferences, but that is why the self-rating guideline is 5.0 and you have to appeal and make your case to the section to play lower than that.
cks
Hall of Fame
I checked the USTA NTRP self rating guidelines. You could self rate as 3.0 or 3.5 for high school tennis depending on JV vs Varsity.
This so much. It really depends on the school they played at.
I played two matches recently with HS kids and they were both clearly 3.0s. They both looked like 4.0s with strokes, but just consistency and ability to attack opponent weaknesses did not exist in either player.
Kids are really difficult to level. Partly because they're often growing/developing so fast, and partly because good technique and actually knowing how to compete at the game of tennis are two very different things. Kids (and adults new to the game) often hyper focus on the former thinking that's all it takes - which means they often compete below the level you'd predict strictly from the eye test.
This is also why it's common to the point of cliche that people look at video and think "they don't look like they're x.0 players" - you're just looking at strokes (and it's especially deceiving if you're looking at highlight video and not actual match play). If you look at those same videos and instead look for what's happening in terms of court positioning and point construction, you can get a much more accurate read at where people can best compete.
I don't see any updated calculations. I also noticed that three matches are missing from December. Sad.
time_fly
Hall of Fame
They may still be trying to make adjustments based on the bump list, or maybe just waiting for the computer to chew through 3.5 months worth of calculations.
Pspielha
Rookie
I’ve noticed that even during the regular season when matches are pulled in within a week or so of being played, it still takes additional time for the calculations to be made. They sit at NC for a few days. I’m guessing it will be longer now with such a large data sweep.
cks
Hall of Fame
TR picked up my match results from last Sunday. Ratings have not been updated yet.
time_fly
Hall of Fame
I wonder what the scope of the data is? I can't imagine there's more than a few thousand matches per week, so we are talking on the order of hundreds of thousands of matches to be evaluated. Which, for a modern computer, is nothing when implemented properly. However we don't know what we're dealing with behind the scenes. Was the software created by a software engineer, or a hobbyist with MS Access and a Visual Basic macro?
silverwyvern4
Semi-Pro
We don't even know if Tennis Record is run by humans or by reptilian lizard people.
HBK4life
Hall of Fame
It’s run by the usta to give 4.5s playing down to 4.0 with the goal of making their lives miserable by becoming obsessed with a number.
TennisHokie
New User
Question for everyone. In our local league we had a 2.5S rated 40 year old woman that hasn't picked up a racket in over 20 years. Played mixed 6.0 which is the first time playing USTA. Finished the local league 2-1 and played at state and went 1-2. Total stats are 3-3 overall record. 6-6 for Sets W/L and played in 100 games which was 49 L and 51 W. Her partner was a seasoned 3.5C player that's been to 4 sectionals and 1 Nationals. At the end of the year rating she was bumped from a 2.5S to 3.5M, her partner stayed at 3.5C. anyone see something like this before?
schmke
Legend
Sure. Sounds like the partner played men's league and so their rating was based on their men's matches (3.5C) and the Mixed matches didn't factor into his rating at all. If the woman only played Mixed, her rating would be based solely on her Mixed play and she would get an 'M' rating, in this case a 3.5M. It very well could be that the partner's M rating, had it been published, may have been higher, but that doesn't matter.
Last edited:
silverwyvern4
Semi-Pro
If you have a mixed rating and never played men's league before, do you have to self rate again to join a men's league?
Moon Shooter
Hall of Fame
Thanks schmke so when usta determined the female’s mixed rating would the male’s “mixed rating” have mattered or just his dynamic rating from adult league?
So for example let’s a guy plays 3 adult matches in 2021, ends up with 3.32 dynamic rating, and so gets a 3.5c rating at the end of the year. Then in 2022 he plays 33 7.0 and 8.0 mixed doubles matches without any adult matches mixed in. Would his rating for purposes of deciding his partners and his opponents mixed performance rating always be 3.32 for every single one of thes 33 mixed doubles matches? Of course if they had enough adult rating to get a c rating none of these games would matter anyway. But if they only played mixed and so their mixed rating was all that mattered then it would always be the same number (3.32) used for his rating right?
schmke
Legend
I believe that the USTA is calculating mixed ratings for all players, but only publishes the levels for those that only played Mixed and have their Adult/C rating expire.
Moon Shooter
Hall of Fame
ok so do you know if that mixed dynamic rating would include both the adult and mixed matches or just the mixed matches?
schmke
Legend
Just the mixed.
TennisHokie
New User
I understand why one is "Mixed Rating" and why the other is "C" that really wasn't my question - What I was questioning is to why a new self rated 2.5S player that went basically .500 in 6 matches got bumped a FULL point. Bumping up that player from a 2.5 to a 3.5 is an injustice when you see her play and it would essentially drive that person out of wanting to play again because she could barely play with 3.0 women players let alone 3.5. I'm just curious if anyone has seen a full point bump like that with those type of results.
silverwyvern4
Semi-Pro
If someone had a 3.5M from only playing mixed and then had to self rate for men's league and he self rated as a 4.0, would that replace his mixed rating? He could no longer play as a 3.5 in mixed league?
TennisHokie
New User
That is correct - But the question would be why did he re-self rate as a 4.0?
silverwyvern4
Semi-Pro
Prestige
TennisHokie
New User
Ahhh the the nice Badge of 4.0 --- i'd rather have state / sectional / national medals but to each their own...lol
This is unusual, but in a limited number of matches for an S-rate, it happens sometimes, especially since half her matches were playoffs. It seems unfair in this case, though, so I would appeal to the sectional head to manually adjust it. We had a guy bumped from 4.0 to 5.0 many years ago who got manually adjusted back to 4.5 (he was borderline 4.0/4.5, but would have been a waste of time to even attempt to play 5.0, so the adjustment was more than justified). Then it would just depend on how reasonable your section is.
