ITF Tennis Number on USTA Profile (NTRP to stay)

ChaelAZ

ChaelAZ

G.O.A.T.

1a765dc1e1efbde650dad3c03c17bdd173d0404ff992373afeffd7e74883ccec.jpeg
Get Excited! Your Number is Coming!
The ITF World Tennis Number is a free, 40-1 global rating scale. For now, players with recorded matchplay in the USTA and ITF database will be given a number – one for singles and one for doubles. Beginners start at 40 and professional tennis players will be closer to 1. In the coming weeks, you will be able to view your ITF World Tennis number on your usta.com profile.
Your number will also feature a GameZone, which is a range for evaluating upcoming matches as well as determining players within your skill level. If you play against opponents within your GameZone, you could expect to win the match between 35-65% of the time!*
b7301a01d7967655d0a3c1077af0c5cc3b3e9afd7010e7c84cb715c419ba2530.gif
How will my number be used? What about USTA NTRP?
As more data is collected on player results across a variety of play, and as more tools are released, we believe the ITF World Tennis Number will be the gold standard rating system. If you are curious about USTA NTRP, it is NOT planned to go away as it is a key tool designed specifically for your USTA League experience.

For tournament play in 2022, while tournament directors may look at your ITF World Tennis Number, current methods for seeding and selection (like rankings) are still an integral part of USTA tournaments and will be used in parallel. Introducing the ITF World Tennis Number won’t be done all at once, as we are committed to ensuring it meets tennis player needs.

For more information on where and how USTA Tournaments are using the ITF World Tennis Number, we recommend you contact the tournament directors for events in which you are interested in playing.




LEARN MORE
 
Yeah, USTA is sending info out rapid fire. And e-mail a month ago, another yesterday saying webinars were coming, then this one today where you can sign-up for the webinars.

More on my blog, but it appears this is targeted primary at juniors to start, I wrote and the pasted text above even says NTRP is not going away, but in theory league players should still have a WTN calculated and have it show up.
 
schmke said:
Yeah, USTA is sending info out rapid fire. And e-mail a month ago, another yesterday saying webinars were coming, then this one today where you can sign-up for the webinars.

More on my blog, but it appears this is targeted primary at juniors to start, I wrote and the pasted text above even says NTRP is not going away, but in theory league players should still have a WTN calculated and have it show up.
Click to expand...


Yeah, I keep getting all that too. Will check out your post too!
 
jmnk said:
yeah, because nothing helps to have a reasonably standardized ranking as having _yet another ranking number based on yet another formula_ :rolleyes:
Click to expand...

Thing is, UTR is not just a number. It is a company which runs leagues and tournaments. It is a threat to the USTA. The USPTA has already partnered with UTR.
 
Cindysphinx said:
Well, I am very confused. Someone needs to spoon feed me.
Click to expand...

The ITF is making a new rating system and going to publish it.

It'll be based on USTA and ITF matches. It'll be separate for singles and doubles. It'll go from 1 to 40, with 40 being beginners and 1 being pros. It'll be updated weekly.

USTA is going to keep using NTRP for leagues and tournaments for now.
 
Cindysphinx said:
Well, I am very confused.
Click to expand...
Not sure what you are confused about. The ITF/WTN number is just like USTA number - only different. You know, with USTA - the higher the number the better the player. Just like with ITF number - only the absolute opposite. And it will update _daily_. Daily! Very important for folks that play 3 matches a year. If USTA number suggests you should be playing in bracket X, but USTA number says you are in bracket Y - then you use UTR number that will put you in bracket Z. Or flip a coin. Or start playing pickleball. And with ITF/WTN you will have one ranking for singles, another for doubles, and a third for mixed. Also very useful - since judging from average frequency of play of adults each one will be based on approximately 1 match/per year. So it will backed by a solid and reliable data.
I mean it will make things so much easier!
 
Last edited:
jmnk said:
Not sure what you are confused about. The ITF/WTN number is just like USTA number - only different. You know, with USTA - the higher the number the better the player. Just like with ITF number - only the absolute opposite. And it will update _daily_. Daily! Very important for folks that play 3 matches a year. If USTA number suggest you should be playing in bracket X, but USTA number says you are in bracket Y - then you use UTR number that will put you in bracket Z. Or flip a coin. Or start playing pickleball. And with ITF/WTN you will have one ranking for singles, another for doubles, and a third for mixed. Also very useful - since judging from average frequency of play of adults each one will be based on approximately 1 match/per year. So it will backed by a solid and reliable data.
I mean it will make things so much easier!
Click to expand...
This makes sense actually.
 
One thing I don't understand in the FAQ's about the World Tennis Number is the description about how the algorithm calculates your results at "set level." The USTA FAQ says,

"Match results are analyzed at set level; the algorithm takes into account each individual set as its own result."

and the FAQ on worldtennisnumber.com says,

"Match results are analysed at set level, meaning our algorithm takes into account each individual set as its own result. Simply, if your match ends 2 sets to 1 in your favour, then the system will update your Number with two set ‘wins’ and one set ‘loss’. Even if you don’t win the overall match any sets you have won will count towards your Number."

So does this mean the score of the set doesn't matter, i.e. a 7-6 set win is the same a 6-0 set win? And for a USTA league match, would winning a match-tiebreaker 3rd set count the same as winning a "normal" set?
 
Hmmmm, is it coincidence there was a Steve Johnson credit with the YA moniker, and there is a Steve Johnson in tennis?
The mind...wobbles.

its-its-always-sunny.gif
 
taylor15 said:
I’m amazed at how hard they are pushing this…all a coming soon feature
Click to expand...
Especially when it appears it will be focused on juniors to start, but are sending it to league players repeatedly, but are also saying NTRP isn't going away.
 
schmke said:
Especially when it appears it will be focused on juniors to start, but are sending it to league players repeatedly, but are also saying NTRP isn't going away.
Click to expand...
Well, this morning they sent a handy guide to understand the relationship. It’s so handy!!

edit: so I just read it, and if it’s not going to be used for league play then what is the point? Will is be used in USTA tournaments? There aren’t many ITF tournaments in Georgia, so struggling to find the value of this next rating system
 
Last edited:
taylor15 said:
Well, this morning they sent a handy guide to understand the relationship. It’s so handy!!

edit: so I just read it, and if it’s not going to be used for league play then what is the point? Will is be used in USTA tournaments? There aren’t many ITF tournaments in Georgia, so struggling to find the value of this next rating system
Click to expand...
Since it will be much more granular than NTRP, the primary purpose is for rating management, duh.
 
taylor15 said:
Well, this morning they sent a handy guide to understand the relationship. It’s so handy!!

edit: so I just read it, and if it’s not going to be used for league play then what is the point? Will is be used in USTA tournaments? There aren’t many ITF tournaments in Georgia, so struggling to find the value of this next rating system
Click to expand...
It appears, for League players, they are touting that it is a way for player's to track their progress on a weekly basis. Perhaps they think having it published weekly will do away with third party sites (and my ratings/reports! :eek: ) that estimate one's dynamic rating. The challenge with this is, WTN is a different rating using a different algorithm and it won't necessarily correlate with NTRP. Some folks will be disappointed to see their WTN improve, and then get the exact same NTRP at year-end.
 
I think there are some good use cases for it for adults. Tournament seeding is one, right now it's based on points and those don't mean much. Another one is a better self rating guide. Something like if you were a WTN 15 and didn't play for 10 years then no less than 4.5.
 
BallBag said:
Another one is a better self rating guide. Something like if you were a WTN 15 and didn't play for 10 years then no less than 4.5.
Click to expand...
Yep, and I think we will eventually see this (noted in my blog post yesterday).

Doing this would require some sort of mapping between the algorithms, and to spur discussion, I just published a blog post with a hypothetical mapping. Go read the details on my blog, and I'm sure it is "wrong", but here is what I came up with as I think it may be in the ball park.

NTPR2WTN.png


As I noted in my blog, I've done a discreet mapping between the algorithms, and given they are different reality won't match that. There will be NTRP 4.0s that happen to be WTN 24s or 19s, but you get the idea.
 
schmke said:
It appears, for League players, they are touting that it is a way for player's to track their progress on a weekly basis. Perhaps they think having it published weekly will do away with third party sites (and my ratings/reports! :eek: ) that estimate one's dynamic rating. The challenge with this is, WTN is a different rating using a different algorithm and it won't necessarily correlate with NTRP. Some folks will be disappointed to see their WTN improve, and then get the exact same NTRP at year-end.
Click to expand...

What's the median # of years playing tennis? 3? 5? 10? Skills improve rapidly usually in the earlier stages; the more years one has been playing and the older one gets, the less likely a big jump will occur [although big declines can always happen due to health reasons].

For people who have been at a given level for years, their skills/results are probably not changing much season to season. Those people will have little need for such a # and they make up the majority of the population.

The ones who would be most interested are those beginning competitive play, and juniors/collegiates, who can improve rapidly in a short period. UTR already accomplishes this but the USTA doesn't own UTR so I guess that's why it came up with WTN.

I'm a numbers guy and I can geek out on such matters but as someone who has been playing 20+ years, I doubt I'm going to improve so much to make a significant difference in my WTN. I guess if it proves reliable, I'll stop looking at UTR and maybe that's the point.
 
S&V-not_dead_yet said:
What's the median # of years playing tennis? 3? 5? 10? Skills improve rapidly usually in the earlier stages; the more years one has been playing and the older one gets, the less likely a big jump will occur [although big declines can always happen due to health reasons].

For people who have been at a given level for years, their skills/results are probably not changing much season to season. Those people will have little need for such a # and they make up the majority of the population.
Click to expand...
And those (League players at least) that would care really care about getting bumped up, and like I noted earlier, since the algorithms are different there is no guarantee an improvement in one's WTN will result in that bump up. So ultimately even these players will grown disillusioned with it and not care.

S&V-not_dead_yet said:
The ones who would be most interested are those beginning competitive play, and juniors/collegiates, who can improve rapidly in a short period. UTR already accomplishes this but the USTA doesn't own UTR so I guess that's why it came up with WTN.
Click to expand...
Ding ding ding ding ding! The ITF (and USTA and LTA and FFT) came up with WTN in response to UTR as they didn't control or have influence over UTR. They say UTR becoming the defacto standard for juniors/collegiates and they couldn't stand for that, and that is why juniors is where WTN is starting with the USTA.

S&V-not_dead_yet said:
I'm a numbers guy and I can geek out on such matters but as someone who has been playing 20+ years, I doubt I'm going to improve so much to make a significant difference in my WTN. I guess if it proves reliable, I'll stop looking at UTR and maybe that's the point.
Click to expand...
This too. UTR was making the USTA less relevant when players and coaches are paying attention to UTR and not their USTA ranking, and as a result are choosing to play UTR events as a way to improve their UTR.
 
schmke said:
Yep, and I think we will eventually see this (noted in my blog post yesterday).

Doing this would require some sort of mapping between the algorithms, and to spur discussion, I just published a blog post with a hypothetical mapping. Go read the details on my blog, and I'm sure it is "wrong", but here is what I came up with as I think it may be in the ball park.

NTPR2WTN.png


As I noted in my blog, I've done a discreet mapping between the algorithms, and given they are different reality won't match that. There will be NTRP 4.0s that happen to be WTN 24s or 19s, but you get the idea.
Click to expand...

Nice blog post! Regarding this part from your post describing your mapping table:

"I've also taken the liberty of having NTRP levels from 3.0-4.5 cover 4 WTN levels each since that is where the bulk of the players are, and the fringes on the end of the NTRP scale cover 3 WTN levels each which is really just a guess on my part. Perhaps more differentiation is needed at the pro level and those ranges are larger."

That's interesting, but I'd be surprised if WTN varies the ranges for different levels. For NTRP I think it's clear that the expected score of a match is based only on the differential, e.g. the expected score of a 5.35 vs. 5.05 is the same as 2.35 vs. 2.05. It's possible that WTN uses a formula where the expected result depends on both the differential AND the absolute levels, but that seems unnecessarily complicated.

Also, what do you think about my earlier question about WTN being calculated at the "set level" (according to their FAQ)? Is it possible that they'll be treating every USTA match as a 2-0 or 2-1 result and ignoring the game scores? That could make for very large disparities between WTN and NTRP, especially for low-data players.
 
TennisOTM said:
Nice blog post!
Click to expand...
Thanks!
TennisOTM said:
Regarding this part from your post describing your mapping table:

"I've also taken the liberty of having NTRP levels from 3.0-4.5 cover 4 WTN levels each since that is where the bulk of the players are, and the fringes on the end of the NTRP scale cover 3 WTN levels each which is really just a guess on my part. Perhaps more differentiation is needed at the pro level and those ranges are larger."

That's interesting, but I'd be surprised if WTN varies the ranges for different levels. For NTRP I think it's clear that the expected score of a match is based only on the differential, e.g. the expected score of a 5.35 vs. 5.05 is the same as 2.35 vs. 2.05. It's possible that WTN uses a formula where the expected result depends on both the differential AND the absolute levels, but that seems unnecessarily complicated.
Click to expand...
I agree with you, I don't think there is any specific varying of ranges. I was simply addressing how to fit 12 NTRP levels into 40 WTN numbers and how that might look. And for simplicity did a discrete mapping where it was 1:N at each level, but I know in reality that alignment won't be there. But in doing a discrete mapping, I had to map some NTRP level to 4 WTN numbers and elected to do it for the main NTRP levels as that is where NTRP has the most data. Frankly, I don't think we really know much about levels below 2.5, nor if 1.0 is really the floor and anyone even exists there, and we don't have much data above 5.5 to inform whether a top pro really would be 7.0 or not. It could very well be that 7.0 is artificial and a WTN 1 would really need to be NTRP 8 or 9 if NTRP were calculated for all levels of players.

Which brings up an interesting question with WTN. They've introduced it with artificial boundaries of 1 and 40, and as calculations are done, what happens when a 1.0 regularly beats another 1.0. Do they venture into 0.x range? Or does the losing 1.0 move to 1.2 and correspondingly there is a mass adjustment required as that filters down? Or what happens when a 40 loses to another 40? Same issue the other way.

The result could be, and I hope this isn't the case, that the weekly updates that are published could have strange changes for players as their rating is affected by these trickle down adjustments as much or more than the matches they actually play. We wild swings with UTR for some players, we might see the same with WTN.

TennisOTM said:
Also, what do you think about my earlier question about WTN being calculated at the "set level" (according to their FAQ)? Is it possible that they'll be treating every USTA match as a 2-0 or 2-1 result and ignoring the game scores? That could make for very large disparities between WTN and NTRP, especially for low-data players.
Click to expand...
It is possible the score within a set may have limited to no factor. But I wouldn't necessarily assume that is the case as calculating at the set level could mean looking at each set as its own rateable thing, but still using the score in the set in calculating that rating.

But there are different camps with rating systems, and I've been meaning to try to write a summary on this but haven't. The TLDR; is that some systems look strictly at just win/loss and ignore scores, while other look at just the score with no credence being given to who won. In tennis a strict "score" approach can lead to a 0-6, 7-6, 1-0 win where the winner "loses" by winning just 8 games to the opponents 12. But a strict "win" approach would look at a 6-0,6-0 the same as 7-6,6-7,1-0 and that seems to be a problem. If the WTN takes the approach of looking at set winners but not the score, they'd be taking a middle ground on the subject.
 
schmke said:
Which brings up an interesting question with WTN. They've introduced it with artificial boundaries of 1 and 40, and as calculations are done, what happens when a 1.0 regularly beats another 1.0. Do they venture into 0.x range? Or does the losing 1.0 move to 1.2 and correspondingly there is a mass adjustment required as that filters down? Or what happens when a 40 loses to another 40? Same issue the other way.

The result could be, and I hope this isn't the case, that the weekly updates that are published could have strange changes for players as their rating is affected by these trickle down adjustments as much or more than the matches they actually play. We wild swings with UTR for some players, we might see the same with WTN.
Click to expand...

Cool, yeah I can see how dealing with the boundary cases could end up skewing things via trickle-down. Interesting that NTRP seems to be better than UTR and probably WTN in this aspect, where they don't have the artificial boundary issue. Or if they do have boundaries they are far enough away from the players the system was designed for so that they don't cause problems.
 
TennisOTM said:
Cool, yeah I can see how dealing with the boundary cases could end up skewing things via trickle-down. Interesting that NTRP seems to be better than UTR and probably WTN in this aspect, where they don't have the artificial boundary issue. Or if they do have boundaries they are far enough away from the players the system was designed for so that they don't cause problems.
Click to expand...
Correct. Until they decide to do manual adjustments or redistribution which they have done periodically for some sections or across the board it seemed one year. But that is the USTA messing with things, not the algorithm.
 
schmke said:
And those (League players at least) that would care really care about getting bumped up, and like I noted earlier, since the algorithms are different there is no guarantee an improvement in one's WTN will result in that bump up. So ultimately even these players will grown disillusioned with it and not care.
Click to expand...

With enough data, you'll be able to figure how A) how well WTN correlates with bumps; and B) how well WTN correlates with your model.

The problem I see: the simplest thing to do would have been for the USTA to release people's DNTRP so we can see the under the hood #s and changes. But they probably didn't want to do that because that would lead to more ratings management.

So they came up with WTN which isn't as good as DNTRP to hopefully avoid the ratings management. But the less well WTN correlates to DNTRP, the less relevant it will be.

it seems like they are stuck between a rock and a hard place. I'm betting they will have to water it down enough such that your model will still be superior.
 
The fact that players will have a separate WTN for singles & doubles will make it especially difficult to relate it to NTRP. It will be interesting to see how much certain players differ in singles vs. doubles WTN when they play both in leagues. It'll start to be more obvious which players might be out-of-level at one and in-level at the other.

But even for players who play only singles or only doubles, if their opponents include those who play both (and are better at one or the other), their NTRP vs. WTN ratings could start to deviate pretty dramatically.
 
I am receiving email daily now, so wonder when this will hit USTA profiles. All email keep just saying 'coming very soon'.
 
It's sad that all of these tennis rating systems do so many meaningless things like:

1) We go from 1- 40 instead of just 1-7! (there are just as many real numbers between 1 and 2 as there are between 1-40 so this is completely irrelevant)
2) We go in reverse order to every other rating system known to man. In our system the best player has the lowest numbered rating. (Wow thanks for making this a bit more convoluted than it has to be!)

Instead of doing some real improvements that people should value like:
A) we don't put a ceiling on the ratings so your rating will actually have a meaning. For example if you are 1.00 higher than your opponent we predict you should win 6-3 6-3.
B) having the predictions out in the open means people understand the impact of the ratings and it makes the whole thing more transparent and significant. So yes instead of being upset that you lost maybe you should be happy because you lost 6-4 7-6 and the rating predicted you would lose 6-1 6-1.


At least they are publishing the ratings more frequently which hopefully will take some of the mystery out of ratings.
 
Moon Shooter said:
It's sad that all of these tennis rating systems do so many meaningless things like:

1) We go from 1- 40 instead of just 1-7! (there are just as many real numbers between 1 and 2 as there are between 1-40 so this is completely irrelevant)
2) We go in reverse order to every other rating system known to man. In our system the best player has the lowest numbered rating. (Wow thanks for making this a bit more convoluted than it has to be!)

Instead of doing some real improvements that people should value like:
A) we don't put a ceiling on the ratings so your rating will actually have a meaning. For example if you are 1.00 higher than your opponent we predict you should win 6-3 6-3.
B) having the predictions out in the open means people understand the impact of the ratings and it makes the whole thing more transparent and significant. So yes instead of being upset that you lost maybe you should be happy because you lost 6-4 7-6 and the rating predicted you would lose 6-1 6-1.


At least they are publishing the ratings more frequently which hopefully will take some of the mystery out of ratings.
Click to expand...
It is interesting that they are giving you an optimal competition range, roughly +-2 that the algorithm says you have a 35-65% chance of winning the match. That also tells me that each half level in current ranking equals about 5 but it's more dynamic.

Ex. a 3.5 should beat another 3.5 about 35-65% of the time probably, but a 3.49 dynamic and a 3.51 are probably also in that range. Since this has a larger and more dynamic range then you can end up in a more competitive range quicker. I know the majority of USTA players end up in the same bucket for years, but it's also not uncommon for someone to improve a lot and then be stuck at the same level all year b/c of the once per year update. We are seeing that locally now, a couple of players who have gotten coaching or are young that are winning all their matches, but struggling to play up b/c there are teams that don't want a lower ranked player on their team. It takes practice matches with these teams, and then after winning hoping egos don't get in the way to get accepted to play with them.

I don't think that means that the WTN is necessarily fixing anything, especially since NTRP will be the decider for leagues. Having the dynamic NTRP updated weekly could have solved the same problem I think. at least then players how are improving (or who over self rated and can't win) could get into the appropriate level quicker. I know someone who self rated at 5.0 and is losing every match almost 6-0 6-0, his ego is bruised and he's stuck playing that level all year unless he can appeal down - which I doubt he'll do.
 
Another email today. Maybe 5 or 6 so far.
And did I mention... I let my USTA membership lapse 4 years ago.
Boy am I excited to get my WTN number I can't wait!!! :unsure:
 
ChaelAZ said:
lol. IT IS ALMOST HERE!!!
Click to expand...
It's like Achilles and the Tortoise...
You are 99% of the way to getting your WTN!!
You are 99.5% of the way to getting your WTN!!
You are 99.75% of the way to getting your WTN!!
You are 99.875% of the way to getting your WTN!!
...
 
You must log in or register to reply here.

Similar threads

F
ITF World Tennis Number to NTRP Rating Conversion
Replies
9
Views
6K
Moon Shooter
M
schmke
ITF ups the ante in the "global" ratings game
Replies
26
Views
8K
Moon Shooter
M
V
Are big changes to USTA and NTRP on the horizon?
Replies
14
Views
2K
Moveforwardalways
Moveforwardalways
CiscoPC600
World Tennis Number Updates Now Live: Are High Confidence Ratings Improved?
Replies
24
Views
1K
Roforot
R
K
Opens, ITF, Adult league?
Replies
6
Views
1K
krnmc
K
Back
Top