The dreaded EOY Rating thread

Not to drop names but we’ve had a private discussion on this and his numbers weren’t far off from TR.
Oh, that is who it is.

To be fair, if TR had him at 4.38, I had him no where close to that high. Solid 4.0 yes so 3.5 seems egregiously low, but if we are going to consider the USTA's year-end level to be accurate/correct, I was off less than half the amount TR was off. I'll still take my rating over TR's any day.
 
Mixed doubles is the Achilles heel of UTR. Unless both genders are playing against one another in singles and unbalanced co-ed competition over multiple years, UTR will always have a skewed data set. 90% of each person’s rating is against same gender competition and therefore the rating is not validated in mixed competition.

I am not sure it would take that many matches to sort it out. I think those matches would be very helpful. Remember UTR completely ignores results from over 12 months.

In my area it seems the women have slightly inflated ratings compared to men. (or men are slightly deflated). I think the same is true of WTN - the women seem to have a bit better WTN rating then I think compared to the men.

You would think some High school guys looking for a scholarship would play some official tournament matches against some female players to increase their UTR.
 
All I am saying is that UTR takes a rating based on a single gender data set and then applies it in a different environment against other players who are also only validated against themselves. I do not see how this can be accurate, but maybe one if you with a phd in statistics can explain it.
 
All I am saying is that UTR takes a rating based on a single gender data set and then applies it in a different environment against other players who are also only validated against themselves. I do not see how this can be accurate, but maybe one if you with a phd in statistics can explain it.

The same can happen geographically. But they have a few sectional and national matches as well as people that move to try to sort it all out. I am not so convinced that those few matches really sort it out either. But the lady that is in charge of USTA ratings assures us that it does. She probably has enough data where individuals move and play in other areas and the dynamic ratings hold up. So even though I have my doubts I still think she is honest and would have good ways to check. But I just wonder if no ratings in 2020 really set us back. Also when many people on these national teams are self rated or potentially sand bagging it can really be hard to tell. So much seems to ride on so few games.
 
For USTA, I bet the most challenging matches are ones for players with very few matches. Like, a guy with 4 doubles matches every year year plays a guy in singles with 2 singles matches last year and two more four years ago and no more, good luck with THAT prediction lol
 
Decided to look back at my own matches this year to see how well TR did at predicting the W/L result for matches where they had pre-match ratings for everyone involved.

As a low-end 4.0 player it should in theory be pretty easy to get a good W/L prediction rate for me, since I'm the underdog in most matches, especially because I almost never have a high-end partner in doubles. If you just predicted a loss for all my matches, you'd get much better than 50% correct without even looking at my opponents.

So how did TR do? In my men's rating-eligible matches, they predicted 11 out of 19 (58%) correct. That's 3/6 correct for singles and 8/13 for doubles. I had 4 TR-upset wins and 4 TR-upset losses.

Shockingly, for 11 mixed doubles matches, TR predicted every single one of them correctly via their mixed-specific (red) pre-match rating numbers. 5 wins and 6 losses, all predicted correctly. Just a lucky fluke, maybe??

It'd be interesting to test how well UTR did for the same matches, though I don't think it's possible to tell by looking back, even with a paid UTR membership? They update the ratings across the board every night and you can't see what everyone's rating was before a past match was played, I don't think.
 
Decided to look back at my own matches this year to see how well TR did at predicting the W/L result for matches where they had pre-match ratings for everyone involved.

As a low-end 4.0 player it should in theory be pretty easy to get a good W/L prediction rate for me, since I'm the underdog in most matches, especially because I almost never have a high-end partner in doubles. If you just predicted a loss for all my matches, you'd get much better than 50% correct without even looking at my opponents.

So how did TR do? In my men's rating-eligible matches, they predicted 11 out of 19 (58%) correct. That's 3/6 correct for singles and 8/13 for doubles. I had 4 TR-upset wins and 4 TR-upset losses.

Shockingly, for 11 mixed doubles matches, TR predicted every single one of them correctly via their mixed-specific (red) pre-match rating numbers. 5 wins and 6 losses, all predicted correctly. Just a lucky fluke, maybe??

It'd be interesting to test how well UTR did for the same matches, though I don't think it's possible to tell by looking back, even with a paid UTR membership? They update the ratings across the board every night and you can't see what everyone's rating was before a past match was played, I don't think.


I was thinking of doing the same thing for my matches. With UTR even if you used the current rating for your opponents and partners it might be interesting to see what their record was - assuming your partners and opponents have a decent number of games in utr. I don't think my own opponents or partners had enough games in UTR to make it worth the time.
 
For USTA, I bet the most challenging matches are ones for players with very few matches. Like, a guy with 4 doubles matches every year year plays a guy in singles with 2 singles matches last year and two more four years ago and no more, good luck with THAT prediction lol


Yes that does make it challenging. But that is largely self inflicted because USTA has a separate men's and women's rating system and none of the mixed matches are included. I think the lack of rated matches is the main reason ntrp lacks accuracy.
 
If UTR and Tennis Record remain forever broken, all this discussion is Moot. When are they going to fix it??
This is what the UTR service desk wrote to me nine days ago. It wasn't encouraging.

"We know as a player having all of your scores reflected in your profile is crucial, and we sincerely apologize for the delay as we are currently experiencing a roadblock with importing USTA League results. Our data team is aware these scores are missing and they're working to get the scores in as soon as possible! That said, we don't have a set timeframe for when you'll see them in your profile, but know that this is a top priority and we'll have them added in time."
 
Oh, that is who it is.

To be fair, if TR had him at 4.38, I had him no where close to that high. Solid 4.0 yes so 3.5 seems egregiously low, but if we are going to consider the USTA's year-end level to be accurate/correct, I was off less than half the amount TR was off. I'll still take my rating over TR's any day.
Your ratings are far superior to TR. State tested and approved.
 
This is what the UTR service desk wrote to me nine days ago. It wasn't encouraging.

"We know as a player having all of your scores reflected in your profile is crucial, and we sincerely apologize for the delay as we are currently experiencing a roadblock with importing USTA League results. Our data team is aware these scores are missing and they're working to get the scores in as soon as possible! That said, we don't have a set timeframe for when you'll see them in your profile, but know that this is a top priority and we'll have them added in time."

I suspect USTA turned off access data access for UTR and possibly TR in order to try to force everyone into the ITF rating universe. TR usually has a pause around this time of year anyway so with them it’s hard to judge.
 
I suspect USTA turned off access data access for UTR and possibly TR in order to try to force everyone into the ITF rating universe. TR usually has a pause around this time of year anyway so with them it’s hard to judge.

You would think they would get the WTN issues worked out before they tried to force everyone to pay attention to it.
 
You would think they would get the WTN issues worked out before they tried to force everyone to pay attention to it.
OMG, I just spot-checked WTN for some player records, and their DB is still useless. One of my teammates, who played four matches this fall and received a year-end of rating of 4.0C doesn't even appear in the WTN records. I searched his partners, and his name simply appears as "player" with a very poor WTN rating. His real rating, if one was generated, would certainly be much better.

In other cases, only one of our three-line matches got recorded. Of course, all of the data was recorded correctly and entered immediately into TennisLink. But, WTN failed to upload it, or record it.

So, I did a small spot-check and found several omissions or other problems. I don't know anything about the WTN algo. But, if the data input is garbage the output is garbage (GIGO as they used to say). USTA should be ashamed of themselves. Apparently, no one even checked the accuracy of the WTN DB! They spent millions on marketing a junk product.
 
Last edited:
OMG, I just spot-checked WTN for some player records, and their DB is still useless. One of my teammates, who played four matches this fall and received a year-end of rating of 4.0C doesn't even appear in the WTN records. I searched his partners, and his name simply appears as "player" with a very poor WTN rating. His real rating, if one was generated, would certainly be much better.

In other cases, only one of our three-line matches got recorded. Of course, all of the data was recorded correctly and entered immediately into TennisLink. But, WTN failed to upload it, or record it.

So, I did a small spot-check and found several omissions or other problems. I don't know anything about the WTN algo. But, if the data input is garbage the output is garbage (GIGO as they used to say). USTA should be ashamed of themselves. Apparently, no one even checked the accuracy of the WTN DB! They spent millions on marketing a junk product.

Yeah I've come across a few of those "player" guys as well. I think they were all guys who had become first-time USTA members pretty recently, but I could be remembering wrong.

GIGO may contribute to the problems but I'm pretty sure the algorithm also has major issues. It's not hard to find players with lots of data entered correctly but still end up with nonsensical ratings.
 
Yeah I've come across a few of those "player" guys as well. I think they were all guys who had become first-time USTA members pretty recently, but I could be remembering wrong.

GIGO may contribute to the problems but I'm pretty sure the algorithm also has major issues. It's not hard to find players with lots of data entered correctly but still end up with nonsensical ratings.


Yes a team loses more games then they win against a much worse rated team, but they still improve their rating. (I hate that I can't say "higher" or "lower" rated teams with WTN because they wanted to make their system special by making the higher rated team to be the worse rated team. That decision alone should disqualify them from being taken seriously)
 
  • Like
Reactions: PK6
Santa Claus left a bag full of missing scores at UTR this morning!

You made me look .... and in my case, a bunch of matches that had fallen off (no longer with gray ball thingy) are now back listed as used for UTR calculation.
I think 6-8 matches. Odd.
 
Sooo glad I don’t have to deal with this s—t anymore-used to be nervous wreck about end of the year rating as well others on team when I was captain. As I’ve said USTA is a waste of time! Along with all of you saying individual goals mean nothing playing on mixed/mens league teams. My Personal goals are more important then the USTA ****/sooo glad I’m done with this forever!!!
 
Sooo glad I don’t have to deal with this s—t anymore-used to be nervous wreck about end of the year rating as well others on team when I was captain. As I’ve said USTA is a waste of time! Along with all of you saying individual goals mean nothing playing on mixed/mens league teams. My Personal goals are more important then the USTA ****/sooo glad I’m done with this forever!!!

If you are done with this forever: Please.Go.Away

Your posts bring nothing to this board. You are not helpful, give no insight that may assist another ... and you don't even play tennis anymore.

Go Away!
 
UTR's integration with USTA league reporting is still broken with no timeline for being fixed per USTA customer service

This doesn't make any sense to me. USTA does not care about UTR, they have been pushing WTN. Do you mean UTR customer service? They have more incentive to integrate with USTA results.

All of my USTA results over the past 2 months have now been incorporated into UTR (was fixed about 2 weeks ago)./
 
This doesn't make any sense to me. USTA does not care about UTR, they have been pushing WTN. Do you mean UTR customer service? They have more incentive to integrate with USTA results.

All of my USTA results over the past 2 months have now been incorporated into UTR (was fixed about 2 weeks ago)./
Correct, clearly, USTA and UTR are independent of each other. They're also competitors in some cases.

But UTR hasn't updated all of the USTA Adult League Matches. They're still missing many matches beyond December 4th. That's only two weeks of data. But, power users pay for this service, and it's not unreasonable to expect timely updates.
 
Correct, clearly, USTA and UTR are independent of each other. They're also competitors in some cases.

But UTR hasn't updated all of the USTA Adult League Matches. They're still missing many matches beyond December 4th. That's only two weeks of data. But, power users pay for this service, and it's not unreasonable to expect timely updates.

it's interesting that your matches beyond Dec 4th are not included, but mine are!
 
I got an email from UTR saying matches were added to my profile. They added 2/3 that I have played since mid November.
 
I notice TR hasn’t updated since 11/12, which was the same time UTR originally broke. Does USTA offer a developer API, or are sites like UTR, TLS, and TR scraping the tennislink site with automated queries? It seems like USTA has changed or disabled whatever data access method these sites were using. I still suspect it’s part of the plan to get players using WTN instead. Too bad the WTN subsite is a total joke, requiring manual queries of player names one at a time.
 
UTR added 19 out of my last 20 matches now, almost there!
UTR hasn't added any matches from last weekend (Saturday and Sunday) in my section. They used to add them within 48 hours of entry into TennisLink. Based on what UTR has said recently, I assume they're still experiencing data problems with USTA.

TR hasn't calculated matches since October 31st!
 
Last edited:
TR hasn't updated since November 6th!
I know USTA hates TR because players are always using TR as an indicator of what their rating ”should be” and complaining to league coordinators about differences. I wouldn’t be surprised if they found a way to kill it, especially since they are rolling out the WTN system for continuous ratings. Personally I wouldn’t care if WTN replaced TR as a source of player ratings, but only if they give it a similarly convenient interface to scout teams and players.
 
I know USTA hates TR because players are always using TR as an indicator of what their rating ”should be” and complaining to league coordinators about differences. I wouldn’t be surprised if they found a way to kill it, especially since they are rolling out the WTN system for continuous ratings. Personally I wouldn’t care if WTN replaced TR as a source of player ratings, but only if they give it a similarly convenient interface to scout teams and players.
My WTN hasn't changed in 3 weeks even though I've played 8 matches in that time period. How often does it update?
 
I know USTA hates TR because players are always using TR as an indicator of what their rating ”should be” and complaining to league coordinators about differences. I wouldn’t be surprised if they found a way to kill it, especially since they are rolling out the WTN system for continuous ratings. Personally I wouldn’t care if WTN replaced TR as a source of player ratings, but only if they give it a similarly convenient interface to scout teams and players.
 
WTN's logic produces useless results. Curiously, WTN is recording matches every week, while UTR acknowledges data retrieval delays and updates slowly and sporadically, if at all.

TR has shut down since they have not calculated since October 30th. Something is up. Let's see if USTA data starts to flow again to 3rd parties.

USTA is in bed with WTN, so WTN is not a 3rd party; they're on the team.
 
Last edited:
Glad I checked here before making my own thread about missing matches.

On a different note, once you have a C rating, do you still need three matches for any chance to get bumped? For example, if I play 2 sectional matches only this year and not play for rest of year, can I still get bumped?
 
Glad I checked here before making my own thread about missing matches.

On a different note, once you have a C rating, do you still need three matches for any chance to get bumped? For example, if I play 2 sectional matches only this year and not play for rest of year, can I still get bumped?

How would you qualify to play sectionals without any regular season matches?
 
WTN's logic produces useless results. Curiously, WTN is recording matches every week, while UTR acknowledges data retrieval delays and updates slowly and sporadically, if at all.

TR has shut down since they have not calculated since October 30th. Something is up. Let's see if USTA data starts to flow again to 3rd parties.

USTA is in bed with WTN, so WTN is not a 3rd party; they're on the team.


I find WTN to be basically the best rating system there is for my area. I think if you play in an area where most people are playing over a dozen matches rated by UTR in both singles and doubles every 12 months then UTR may be better.

WTN still has a few outliers that are hard to understand, but I noticed those outliers are moving back into line as well. Maybe not as quickly as they should but still moving in the correct direction. So a USTA 3.5a player that used to have an insane 3.7 wtn doubles rating now has an 11.1 rating. So I think if there was a bug they may have found it. It will take a while to recover in doubles but it is moving in the right direction.

The main consistent inaccuracies still involve groups in different pools such as men and women. But this is not the fault of the WTN as NTRP is also off and even worse. I have heard the pools of adult and junior players have issues as well. Until these populations truly compete with each other in a statistically significant way the ratings will always be off.

WTN needs to switch around so higher rated players have a higher number. Instead of having higher rated players having a lower number. They should also remove the 40 ceiling.
 
Glad I checked here before making my own thread about missing matches.

On a different note, once you have a C rating, do you still need three matches for any chance to get bumped? For example, if I play 2 sectional matches only this year and not play for rest of year, can I still get bumped?
Your rating won't change if you play fewer than 3 matches in a year. You can keep your old C rating for 36 months before you become a self rate again.
 
I find WTN to be basically the best rating system there is for my area. I think if you play in an area where most people are playing over a dozen matches rated by UTR in both singles and doubles every 12 months then UTR may be better.

WTN still has a few outliers that are hard to understand, but I noticed those outliers are moving back into line as well. Maybe not as quickly as they should but still moving in the correct direction. So a USTA 3.5a player that used to have an insane 3.7 wtn doubles rating now has an 11.1 rating. So I think if there was a bug they may have found it. It will take a while to recover in doubles but it is moving in the right direction.

The main consistent inaccuracies still involve groups in different pools such as men and women. But this is not the fault of the WTN as NTRP is also off and even worse. I have heard the pools of adult and junior players have issues as well. Until these populations truly compete with each other in a statistically significant way the ratings will always be off.

WTN needs to switch around so higher rated players have a higher number. Instead of having higher rated players having a lower number. They should also remove the 40 ceiling.
I’m certainly not sold on WTN. My WTN is not aligned very well with utr or tr or the eye test for that matter. I’m not a borderline 5.0
 
I find WTN to be basically the best rating system there is for my area. I think if you play in an area where most people are playing over a dozen matches rated by UTR in both singles and doubles every 12 months then UTR may be better.

WTN still has a few outliers that are hard to understand, but I noticed those outliers are moving back into line as well. Maybe not as quickly as they should but still moving in the correct direction. So a USTA 3.5a player that used to have an insane 3.7 wtn doubles rating now has an 11.1 rating. So I think if there was a bug they may have found it. It will take a while to recover in doubles but it is moving in the right direction.

The main consistent inaccuracies still involve groups in different pools such as men and women. But this is not the fault of the WTN as NTRP is also off and even worse. I have heard the pools of adult and junior players have issues as well. Until these populations truly compete with each other in a statistically significant way the ratings will always be off.

WTN needs to switch around so higher rated players have a higher number. Instead of having higher rated players having a lower number. They should also remove the 40 ceiling.
Um, a 3.5 player should not have an eleven handle on their WTN. I track several dozen 3.5, 4.0 and 4.5 players. The 3.5s are normally mid 20s to very low 30s. Some are low 20s.

But, I've seen outliers. I know of two players with high 6 or low 7 UTRs (doubles) who have WTN_d ratings around 30! They've each played over twenty matches in 2022. WTN is garbage.
 
Last edited:
Back
Top