Share your WTN - Crowd sourcing the NTRP to WTN mapping

11? sandbagging at ntrp3.5?!

NGL but everything I've seen in this thread so far either has me think that the UK implementation of WTN is horrifically off (underrating), or the US implementation is horrendously overrating.

Which is kind of worthless for an International system that's supposed to be comparable.

I currently play UK club tennis at what I would consider an okay standard (guesstimate 4.0ish? - yes yes, internet rating etc. etc.) and pretty much everyone in the top half of the league ranges between 15-20 (doubles) on average.

The best players that I've been able to find are a guy that's played a National Level event at 9.9 singles, 6.6 doubles; and a lady that played at Wimbledon in the 90s at 14.3 singles, 8.1 doubles.

At the same time, the League only used imported matches from last year and before to calculate the WTN and hasn't imported any of this years matches despite them saying they automatically would...
 
NGL but everything I've seen in this thread so far either has me think that the UK implementation of WTN is horrifically off (underrating), or the US implementation is horrendously overrating.

Which is kind of worthless for an International system that's supposed to be comparable.

I currently play UK club tennis at what I would consider an okay standard (guesstimate 4.0ish? - yes yes, internet rating etc. etc.) and pretty much everyone in the top half of the league ranges between 15-20 (doubles) on average.

The best players that I've been able to find are a guy that's played a National Level event at 9.9 singles, 6.6 doubles; and a lady that played at Wimbledon in the 90s at 14.3 singles, 8.1 doubles.

At the same time, the League only used imported matches from last year and before to calculate the WTN and hasn't imported any of this years matches despite them saying they automatically would...
Are they showing you which matches count for your rating?
 
4.5A M
Doubles 7.8
No singles rating because haven't played singles in years.

Will WTN be incorporated into the league ratings system (i.e. factored into whether a player is a 4.0/4.5/etc.)?
 
4.5A M
Doubles 7.8
No singles rating because haven't played singles in years.

Will WTN be incorporated into the league ratings system (i.e. factored into whether a player is a 4.0/4.5/etc.)?
Nope, this purely for fun right now. Might be used for seeding tournaments one day.
 
NGL but everything I've seen in this thread so far either has me think that the UK implementation of WTN is horrifically off (underrating), or the US implementation is horrendously overrating.

Which is kind of worthless for an International system that's supposed to be comparable.

I currently play UK club tennis at what I would consider an okay standard (guesstimate 4.0ish? - yes yes, internet rating etc. etc.) and pretty much everyone in the top half of the league ranges between 15-20 (doubles) on average.

The best players that I've been able to find are a guy that's played a National Level event at 9.9 singles, 6.6 doubles; and a lady that played at Wimbledon in the 90s at 14.3 singles, 8.1 doubles.

At the same time, the League only used imported matches from last year and before to calculate the WTN and hasn't imported any of this years matches despite them saying they automatically would...
yeah agreed, based on the descriptions of an 11, i'm def not an 11...
but like utr, i suspect it's a system that is constantly calculating and correcting itself as it gets more data (kinda like how ai's are trained to "learn")
 
From what I'm seeing so far, the singles ratings look pretty reasonable as long as you limit to those marked reliable (blue tick). Here are the singles ratings and game zone ranges for all the 4.0C singles players with a blue tick mark from a recently completed 7-team flight:

15.2 (13.4-17.0)
15.4 (13.6-17.2)
16.0 (14.2-17.7)
16.4 (14.7-18.2)
16.6 (14.9-18.4)
17.0 (15.2-18.7)
17.5 (15.7-19.2)
17.9 (16.1-19.7)
19.1 (17.3-20.9)
19.4 (17.7-21.2)
20.2 (18.5-22.0)
20.3 (18.5-22.0)
20.3 (18.6-22.1)

All the game zone ranges overlap except for the best few guys with the worst few guys, which I suppose is what you'd expect for a NTRP level range.
 
Are they showing you which matches count for your rating?

Afaik the initial calculation for mine seems based off of pretty much every league match (doubles) I've played (both Men's and Mixed) back to 2017, for a total of 95 matches.

Can't see what matches count for other people as I can only look them up on the League page, but that records every League match; matches that count towards your WTN are on your own profile page, which is how I noticed I was missing all of this year.

yeah agreed, based on the descriptions of an 11, i'm def not an 11...
but like utr, i suspect it's a system that is constantly calculating and correcting itself as it gets more data (kinda like how ai's are trained to "learn")

Yeah, the description on the USTA website seems to be reflect what i would say the UK ratings seen to follow.

Admittedly my 20.8 doubles rating isn't verified, I'm guessing because it doesn't have any recent matches, but i don't see it getting anywhere near the low 10s.

There was an algorithm update for the UK that potentially moved people "no more than 1 or 2 points", but i don't think it's a good look for the US system if people are going to get massively adjusted (i mean honestly could see these WTNs doubled and think they'd be more in line with the UK).

Also looking up some random people, Felix of the Tennis Brothers youtube channel is currently rated 6.8 Singles, 10.8 doubles.
 
4.5A M
Doubles 7.8
No singles rating because haven't played singles in years.

Will WTN be incorporated into the league ratings system (i.e. factored into whether a player is a 4.0/4.5/etc.)?
I don't believe so. The USTA even has come out and said NTRP isn't going away. For league players, WTN is just a curiosity, for now at least.
 
All self-kidding aside.....

I have a bunch of results in tennislink but the "USTA" site shows no results for me and then shows my grey WTN. I am not sure why my results are not pulled over into USTA from tennislink and used to compute an actual number for me.
 
All self-kidding aside.....

I have a bunch of results in tennislink but the "USTA" site shows no results for me and then shows my grey WTN. I am not sure why my results are not pulled over into USTA from tennislink and used to compute an actual number for me.
Strange. DM me your info if you want me to see if I can find an explanation.
 
On the subject of the mapping, I have done some initial analysis of 4.0 and 4.5 men and it is up on my blog. Some pretty large ranges of WTN's for a given level, even when you factor in the confidence level throw out outliers by looking at the standard deviation.
 
4.5 C
WTN singles 13.4
WTN doubles 12.1
What are the check marks people are referencing ?
 
Last edited:
All self-kidding aside.....

I have a bunch of results in tennislink but the "USTA" site shows no results for me and then shows my grey WTN. I am not sure why my results are not pulled over into USTA from tennislink and used to compute an actual number for me.
The USTA site only shows tournament results. WTN should include your league results from tennislink but I don't know how many results are needed for the blue checkmark. I remember reading that they include upto 4 years of results.
4.5 C
WTN singles 15.1
WTN doubles 11.6
What are the check marks people are referencing ?
There's a small circle with a checkmark to the upper right of your number. Blue means high confidence and grey is low confidence. The Game Zone range is also an indicator of confidence with a narrow Game Zone band indicating higher confidence in the number. Though it looks like everyone is between 3.5 and 4.0 so maybe it's not the rating deviation from the Glicko 2 algorithm.
 
The USTA site only shows tournament results. WTN should include your league results from tennislink but I don't know how many results are needed for the blue checkmark. I remember reading that they include upto 4 years of results.

There's a small circle with a checkmark to the upper right of your number. Blue means high confidence and grey is low confidence. The Game Zone range is also an indicator of confidence with a narrow Game Zone band indicating higher confidence in the number. Though it looks like everyone is between 3.5 and 4.0 so maybe it's not the rating deviation from the Glicko 2 algorithm.
The vast majority of ranges for the Game zONe I've seen are between 3.5 and 4.25. There are a very few outliers below 1.0 and the highest I've seen is 4.36.
 
The vast majority of ranges for the Game zONe I've seen are between 3.5 and 4.25. There are a very few outliers below 1.0 and the highest I've seen is 4.36.
Maybe it is RD then. Once they get their website going we should be able to see history and maybe try to reverse engineer the algorithm.
 
USTA 3.0

My single rating is odd with 26.5 and my double rating non available despite one double won in May 2022. I am not sure if the have loaded all my results...
I have contacted them twice in two days and the USTA can't see how many games were loaded to calculate my WTN. Not that I am mad about it but this lack of precision is very puzzling haha.

In addition, I have won two games by a pretty large margin and they have better rating than mine.

(Won 6-2; 6-3 on 3 May 2022 against a WTN 25.6)
(Won 6-3; 6-1 on 31 May 2022 against a WTN 25.1)

This confirms that something is odd, the USTA said they will escalate and come back to me...
 
USTA 3.0

My single rating is odd with 26.5 and my double rating non available despite one double won in May 2022. I am not sure if the have loaded all my results...
I have contacted them twice in two days and the USTA can't see how many games were loaded to calculate my WTN. Not that I am mad about it but this lack of precision is very puzzling haha.

In addition, I have won two games by a pretty large margin and they have better rating than mine.

(Won 6-2; 6-3 on 3 May 2022 against a WTN 25.6)
(Won 6-3; 6-1 on 31 May 2022 against a WTN 25.1)

This confirms that something is odd, the USTA said they will escalate and come back to me...
Wow, you tried contacting them and got a response?
 
On the subject of the mapping, I have done some initial analysis of 4.0 and 4.5 men and it is up on my blog. Some pretty large ranges of WTN's for a given level, even when you factor in the confidence level throw out outliers by looking at the standard deviation.

@schmke , where are you getting the confidence number from? I'm not seeing it on my profile.

I have the same question - your blog says there's a WTN confidence number that goes up to 100. Where do you find that, and can you tell at what confidence level the blue tick mark gets triggered?
 
Can't see what matches count for other people as I can only look them up on the League page, but that records every League match; matches that count towards your WTN are on your own profile page, which is how I noticed I was missing all of this year.

The WTN system might have a major data issue, how come many games are not included?
I have also checked with my USTA league team mates and it seems that a lot of matches are not included???
 
Some of these WTN doubles ratings are absolutely bonkers. I checked a pair of 4.0 guys who regularly partner together in league, but also occasionally play with other partners in both adult and mixed leagues. I have played them and it would be hard to guess which one is higher rated. On Tennisrecord one is 3.83 adult / 3.87 mixed and the other is 3.73 adult / 3.77 mixed. They are both 6.xx on UTR with over 20 doubles matches each in the past year.

Yet on WTN, the first guy is 7.1 and the second is 20.9, each with high confidence (blue tick). To me that suggests their doubles rating calculations have some severe instability. Very disappointing.
 
Some of these WTN doubles ratings are absolutely bonkers. I checked a pair of 4.0 guys who regularly partner together in league, but also occasionally play with other partners in both adult and mixed leagues. I have played them and it would be hard to guess which one is higher rated. On Tennisrecord one is 3.83 adult / 3.87 mixed and the other is 3.73 adult / 3.77 mixed. They are both 6.xx on UTR with over 20 doubles matches each in the past year.

Yet on WTN, the first guy is 7.1 and the second is 20.9, each with high confidence (blue tick). To me that suggests their doubles rating calculations have some severe instability. Very disappointing.

they have a huge data collection issue, just checked one of my team mate, USTA league NY, 3 games (2 singles and 1 double) in May and June 2022, so very recent... and no WTN... so crazy
 
they have a huge data collection issue, just checked one of my team mate, USTA league NY, 3 games (2 singles and 1 double) in May and June 2022, so very recent... and no WTN... so crazy
Is he self-rated and these are his only matches?
 
they have a huge data collection issue, just checked one of my team mate, USTA league NY, 3 games (2 singles and 1 double) in May and June 2022, so very recent... and no WTN... so crazy

It seems they have a huge issue even when they are using enough data, because the ratings (for doubles at least) can be nonsensical even when they think they have plenty of data (blue tick mark).
 
I should have added having folks include the "Game zONe" ranges too. What is odd is it appears for my WTNs but not my WTNd. What are others seeing?

If you can include your GZ, I can track if it is more or less a fixed range around one's WTN or if it varies beyond that.

NTRP: 4.5C
WTNs: 12.6 with GZ of 10.6-14.7
WTNd: 21.0 with no GZ
I think the check mark color (grey or blue) is useful too. I don't trust "grey" ratings.
 
I don't get the WTN widget. It's such an inefficient gimmick. Why don't they show a grid with WTN, tick color & GameZone for both singles and doubles? Two rows and three columns, and no clicking and spinning required! They could even narrow it down to two columns by color coding the WTN rating. Geez.
 
Here are several ratings from the "starting" members of my 18 & Over M3.5 team:

PM
WTNs: 24.3/grey
WTNd: 28.7/blue

MH
WTNs: 29.7/grey
WTNd: 27.9/blue

DF
WTNs: No data
WTNd: 30.5s/blue

MB
WTNs: 17.1/grey //quite an outlier
WYNd: 27.8/blue

RN:
WTNs: 28.9/blue
WTNd: 31.3/grey

MD:
WTNs: 25.5/grey
WTNd: 29.5/grey

JJ
WTNs: 28.3/blue
WTNd: 31.0/grey

TP
WTNs: 26.9/grey
WTNd: 24.1/blue


So, given the eight M3.5 players above, the range is:
WTNs: 29.7 - 17.1 Mean: 25.81
WTNd: 31.3 - 24.1 Mean: 28.85
 
Last edited:
4.0C (male)
20.2 singles (game zone 18.5-22.0); 20.7 doubles - both are blue.

Other 4.0C men from my teams (blue checkmarks only):

25.8 doubles
20.0 doubles
20.5 doubles
20.3 singles (game zone 18.5-22.0); 21.3 doubles
13.3 doubles
24.1 doubles
22.6 doubles
17.4 doubles
19.4 singles (game zone 17.7-21.2); 22.6 doubles
16.1 doubles

These doubles number seems to be a mess. There's a bit of range in ability of these guys, but nowhere near this much.
Wow, the range for the WTNd/blue players is 25.8 - 13.3! The lower rating (25.8) makes sense based on the WTNs of my 4.0 teammates (30.5 - 17.1). But, 13.3 seems like an outlier. Is he playing like a 4.5?
 
Some other 4.0C players in our Southwest section. Actually knowing and having played these guys for years, it is interesting to see the ratings. Some are spot on feeling, but many seem off knowing how strong or weak a player they are and how they rate against everyone. Might be like when UTR hit there just needed to be more data points and history, and tweaks to the algo to get things set.


NTRPWTN sgl/dub
3.7816.8 / 16.8
3.94 (One of the strongest 4.0's we have 6-0 sgl, 4-0 dubs)16.9 / 22.3
3.96 (Interestingly, only played Dubs and was 9-1)14.0 / 14.4
3.67 (again, only dubs)13.8 / 13.8
3.7616.4 / 16.3
3.6517.7 / 17.1
4.0s (Dynamic looks to be around 3.77 - Was 3-3 sgl and 1-0 dubs)16.5 / 28


Anyone creating a correlating chart between NTRP and WTN?
 
Some other 4.0C players in our Southwest section. Actually knowing and having played these guys for years, it is interesting to see the ratings. Some are spot on feeling, but many seem off knowing how strong or weak a player they are and how they rate against everyone. Might be like when UTR hit there just needed to be more data points and history, and tweaks to the algo to get things set.


NTRPWTN sgl/dub
3.7816.8 / 16.8
3.94 (One of the strongest 4.0's we have 6-0 sgl, 4-0 dubs)16.9 / 22.3
3.96 (Interestingly, only played Dubs and was 9-1)14.0 / 14.4
3.67 (again, only dubs)13.8 / 13.8
3.7616.4 / 16.3
3.6517.7 / 17.1
4.0s (Dynamic looks to be around 3.77 - Was 3-3 sgl and 1-0 dubs)16.5 / 28


Anyone creating a correlating chart between NTRP and WTN?
Where are you obtaining this NTRP data from? USTA only publishes year-end data, and only to the tenths place!
 
All self-kidding aside.....

I have a bunch of results in tennislink but the "USTA" site shows no results for me and then shows my grey WTN. I am not sure why my results are not pulled over into USTA from tennislink and used to compute an actual number for me.
Same with me— tennislink shows my league matches but the USTA site that lists my WTN doesn’t display my matches.
 
It’s just another useless number no one will understand but happily obsess over. It appears to be based on tournament results. At least for the M1dwest.
 
Last edited:
Wow, the range for the WTNd/blue players is 25.8 - 13.3! The lower rating (25.8) makes sense based on the WTNs of my 4.0 teammates (30.5 - 17.1). But, 13.3 seems like an outlier. Is he playing like a 4.5?

Nope, the 13.3 WTNd guy is 3.63 on Tennisrecord and 5.xx doubles on UTR. He is has about a .500 W/L record in 4.0 the last few years. No clue why he'd be an outlier. Meanwhile on the other end the 25.8 WTNd guy is 3.60 on Tennisrecord. What!?
 
Nope, the 13.3 WTNd guy is 3.63 on Tennisrecord and 5.xx doubles on UTR. He is has about a .500 W/L record in 4.0 the last few years. No clue why he'd be an outlier. Meanwhile on the other end the 25.8 WTNd guy is 3.60 on Tennisrecord. What!?
Curious ... does one play a lot of mixed doubles and the other not at all?
 
I've been able to start doing a more thorough analysis and posted some observations for women's doubles WTN's. See my blog for all the details (I can't post links here, others can if they wish) but here is a chart I put together showing for each WTN, the count of players at each NTRP level.

20220612-wtndf.png


As you can see, there is a lot of overlap at a bunch of WTN levels, and the number of WTN levels you'll find within an NTRP level is well over 10 at every NTRP level.

Men's doubles chart coming tomorrow (Monday).
 
Last edited:
Back
Top