TennisRecord com Appreciation Thread

travlerajm

Talk Tennis Guru
TR has many haters, complaining that it doesn’t accurately predict usta bumps, so-and-so sandbagger had low TR rating but kicked my ass, yada yada yada

But this thread is to appreciate TR for what it is:

1. A free publicly available resource.
2. A database of rec player playing records.
3. Excellent user interface.
4. Logical and virtually transparent algorithm.
5. Separate ratings for mixed and same gender with cool color coding.
6. Provides match performance ratings for every match.
7. Rating updates frequently.
8. Rating is stable and doesn’t fluctuate between matches.
7. Singles and Doubles ratings can be easily separated by reviewing recent match record.
8. Useful scouting tool.
9. Easy to find a player with minimal info even if full name not known.
10. Continued source of TTW forum debate.
11. Builder of egos.
12. Killer of egos.
13. Neato dial graphic showing how bad you are compared to your level.
14. Accurate to 4 decimal places.
 
Last edited:

happyandbob

Legend
TR has many haters, complaining that it doesn’t accurately predict usta bumps, so-and-so sandbagger had high TR rating but kicked my ass, yada yada yada

But this thread is to appreciate TR for what it is:

1. A free publicly available resource.
2. A database of rec player playing records.
3. Excellent user interface.
4. Logical and virtually transparent algorithm.
5. Separate ratings for mixed and same gender with cool color coding.
6. Provides match performance ratings for every match.
7. Rating updates frequently.
8. Rating is stable and doesn’t fluctuate between matches.
7. Singles and Doubles ratings can be easily separated by reviewing recent match record.
8. Useful scouting tool.
9. Easy to find a player with minimal info even if full name not known.
10. Continued source of TTW forum debate.
11. Builder of egos.
12. Killer of egos.
13. Neato dial graphic showing how bad you are compared to your level.
14. Accurate to 4 decimal places.
I appreciate your ability to come up with 14 reasons and not one mention of UTR
 

travlerajm

Talk Tennis Guru
It’s been like that for about a year, hasn’t it? I think they changed it when they made updates to the site last year.
Not sure, but if you’re reading this, TR guy, please bring back the dial and add a little emoji describing the heat index
 

LuckyAC

New User
Yup, just using it today to look up if my old college teammates are still active. USTA website itself would be almost worthless for people in various unknown sections. Seems to have missed half my matches this season, but still very useful.
 

McLovin

Legend
What is “accurate” in this context? Tracking the USTA rating? Or predicting who will beat who?
By definition, a USTA rating is accurate if published by the USTA. Since their algorithm is proprietary, any other source is inaccurate. TR has 4 decimals of “precision”, but they can have 12 decimals of precision if they want. That doesn’t make it accurate.

If using it as a resource to see how people have done in past matches, then it’s a good site. Much better than tennislink.com. But, that’s not a very high bar to reach…

Hell, give me access to the data and I could do up a nice site using Vue in about 6-8 hours, and it would be worlds better than tennislink.com.
 

Roforot

Hall of Fame
When I've spectated some of my teammates play, I've found the team w/ the higher TR rating wins nearly all of the time.
I have found maybe two occasions where there may have been an upset; obviously people can be injured or have a really bad day.
Perhaps, UTR for those who pay their membership also provides similar predictive results for doubles?

Thanks to some videos and the spreadsheet explaining their calculations, I was able to go and gauge the effect of a singles tournament where I split sets w/ a 3.0, and two 3.5 players (losing the last match)
Adding that information would have lowered my rating to 3.55 which may explain why I was not bumped up as predicted. I'm not clear on why TR didn't have this tournament matches?

When I'm playing a match I avoid looking at the TR of my opponents till afterwards. It may be a con, but I have to believe I can win even we find ourselves going against 2 teaching pros :)
Despite saying how accurate I've found TR's predictability I do not believe in predetermined outcomes.
 

J_R_B

Hall of Fame
TR has many haters, complaining that it doesn’t accurately predict usta bumps, so-and-so sandbagger had low TR rating but kicked my ass, yada yada yada

But this thread is to appreciate TR for what it is:

1. A free publicly available resource.
2. A database of rec player playing records.
3. Excellent user interface.
4. Logical and virtually transparent algorithm.
5. Separate ratings for mixed and same gender with cool color coding.
6. Provides match performance ratings for every match.
7. Rating updates frequently.
8. Rating is stable and doesn’t fluctuate between matches.
7. Singles and Doubles ratings can be easily separated by reviewing recent match record.
8. Useful scouting tool.
9. Easy to find a player with minimal info even if full name not known.
10. Continued source of TTW forum debate.
11. Builder of egos.
12. Killer of egos.
13. Neato dial graphic showing how bad you are compared to your level.
14. Accurate to 4 decimal places.
TR publishes a ton of really interesting and useful information for free. I truly appreciate what they do and I am grateful for the info. You just have to caveat it and use the data in the right way. It's best used for scouting, and the most accurate part of the data is the relative ranking of players within a team. Typically, when I look at TR ratings, the best players on the team are rated near the top and the weaker players near the bottom. This is an invaluable tool for scouting teams at districts or sectionals where you're not typically as familiar with the players on opposing teams. What it's not really accurate with is the absolute rating levels, which would be used to predict bumps or to look at favorites/underdogs at those playoff matches.

If you understand it's limitations and use TR the right way, it is one of the best resources out there, especially free resources (you can get much more accurate rating estimates from Schmke, but you get what you pay for).
 

mpnv1990

Semi-Pro
TR publishes a ton of really interesting and useful information for free. I truly appreciate what they do and I am grateful for the info. You just have to caveat it and use the data in the right way. It's best used for scouting, and the most accurate part of the data is the relative ranking of players within a team. Typically, when I look at TR ratings, the best players on the team are rated near the top and the weaker players near the bottom. This is an invaluable tool for scouting teams at districts or sectionals where you're not typically as familiar with the players on opposing teams. What it's not really accurate with is the absolute rating levels, which would be used to predict bumps or to look at favorites/underdogs at those playoff matches.

If you understand it's limitations and use TR the right way, it is one of the best resources out there, especially free resources (you can get much more accurate rating estimates from Schmke, but you get what you pay for).
UTR is the best IMO.
 

MGArchitect

New User
+1 for tennisrecord showing league teams and their specific results. Not only ranking by court rating, but showing how teams typically stack their lineups by reviewing past matchups. UTR does not have that information.
 

travlerajm

Talk Tennis Guru
UTR is the best at what? Model for measurement of actual tennis skills? Maybe, that's a different debate. Proxy for replicating USTA ratings? Definitely not, and that's what TR's aim is.
I don’t believe that TR’s creator cares about whether or not its algorithm accurately predicts usta bumps.

It does recalibrate every season to match the usta rating band for each player, that’s a different than trying to match the usta algo.

I would actually prefer it if TR would decouple itself from usta ratings. This would create more discrepancies between TR and Usta, but likely result in a more useful dynamic number that is not artificially adjusted at the beginning of each season by trying to match it to another algo.
 

E.T.

Rookie
I don’t believe that TR’s creator cares about whether or not its algorithm accurately predicts usta bumps.

It does recalibrate every season to match the usta rating band for each player, that’s a different than trying to match the usta algo.

I would actually prefer it if TR would decouple itself from usta ratings. This would create more discrepancies between TR and Usta, but likely result in a more useful dynamic number that is not artificially adjusted at the beginning of each season by trying to match it to another algo.
I think most people like it because it seems to mimic NTRP (even though we know it’s not perfect) and it’s based solely on USTA match data. Trying to decouple it from USTA ratings would be tough when it’s based on the same data. I think people like to see where they are in the spectrum of their NTRP level and compare to their opponents etc. I think the anxiety/speculation about whether or not it’s accurate makes it fun, at least for the people I know. Everyone likes to wonder whether or not they are getting bumped at year end, and in our region it’s a pretty good indicator of whether or not there is a chance of it happening.
 

Moon Shooter

Hall of Fame
TR is actually the best tennis in many ways. For example I can see that I have played 1012 doubles games and won 48.7% of them.

I played 140 singles games and won 33.6% of them.

Now that alone won't tell me if I play better singles or doubles. But then I can see under advanced that my average opponent in doubles was 3.2109 and my average opponent in singles was 3.2658 and then we get the idea that I am relatively worse at singles then at doubles. It would be good to the average rating of my partner in doubles to get an even better idea of my relative strength in doubles or singles.
 

Purestriker

Legend
TR has many haters, complaining that it doesn’t accurately predict usta bumps, so-and-so sandbagger had low TR rating but kicked my ass, yada yada yada

But this thread is to appreciate TR for what it is:

1. A free publicly available resource.
2. A database of rec player playing records.
3. Excellent user interface.
4. Logical and virtually transparent algorithm.
5. Separate ratings for mixed and same gender with cool color coding.
6. Provides match performance ratings for every match.
7. Rating updates frequently.
8. Rating is stable and doesn’t fluctuate between matches.
7. Singles and Doubles ratings can be easily separated by reviewing recent match record.
8. Useful scouting tool.
9. Easy to find a player with minimal info even if full name not known.
10. Continued source of TTW forum debate.
11. Builder of egos.
12. Killer of egos.
13. Neato dial graphic showing how bad you are compared to your level.
14. Accurate to 4 decimal places.
TR is number 1 at number 2
 

MGArchitect

New User
Yes, the rec player data base is cool! Unfortunately, our area (phoenix) has moved it's singles flex league to using WTN and only reporting the results there so they are no longer picked up by TR or UTR. So I can no longer see the database in one location. I also have matches on TLN which of course don't show up anywhere else

FWIW, in singles since 2020, on TR, I have a 23-9 record, with 324-218 games W/L. Average opponent rating of 4.0098. (I have a further W/L record of 9-1 in the USTA/WTN singles league and 6-5 on tennis league network)
In doubles, on TR, I have a 51-50 record, with 991-949 games W/L. Average opponent rating of 3.4328.

I guess I am a better singles than doubles player, but this also includes a lot of mixed doubles (which seems to lower the average opponent rating) and doubles matches from when I first came back to tennis after a 20+ year break playing, as I tried to start slow.

Also, TR says I have dished out one double bagel, two bagel and breadsticks, and 3 double breadsticks, while getting double breadsticked myself once and bageled and breadsticked one time. (<-- no one else keeps this stat either!)

Anyway, all this is available for free on TR, so appreciation for sure!
 
Last edited:

E.T.

Rookie
I’ve noticed that I am getting some sort of security notice on tennislink now after a certain amount of time, maybe 20-30 minutes, that it has logged me out. I wonder if this is part of the reason tennisrecord hasn’t updated recently. Maybe they are preventing them from scraping the data.
 

jmnk

Hall of Fame
TR has many haters, complaining that it doesn’t accurately predict usta bumps, so-and-so sandbagger had low TR rating but kicked my ass, yada yada yada

But this thread is to appreciate TR for what it is:

1. A free publicly available resource.
2. A database of rec player playing records.
3. Excellent user interface.
4. Logical and virtually transparent algorithm.
5. Separate ratings for mixed and same gender with cool color coding.
6. Provides match performance ratings for every match.
7. Rating updates frequently.
8. Rating is stable and doesn’t fluctuate between matches.
7. Singles and Doubles ratings can be easily separated by reviewing recent match record.
8. Useful scouting tool.
9. Easy to find a player with minimal info even if full name not known.
10. Continued source of TTW forum debate.
11. Builder of egos.
12. Killer of egos.
13. Neato dial graphic showing how bad you are compared to your level.
14. Accurate to 4 decimal places.
I don't always agree with @travlerajm post but I do when he is right....
 

jmnk

Hall of Fame
[...]

Hell, give me access to the data and I could do up a nice site using Vue in about 6-8 hours, and it would be worlds better than tennislink.com.
first, in programming, just like in construction, _nothing ever worth anything_ gets done in 8 hours. So I'm going to highly doubt your claim.
second, the fact that the maintainer of tennisrecord is able to _get_ the data _is_ very much part of the tennisrecord value. Your statement is like saying 'give me two millions and I can get you ~100,000 annually' - well, yeah.....
 

McLovin

Legend
first, in programming, just like in construction, _nothing ever worth anything_ gets done in 8 hours. So I'm going to highly doubt your claim.
second, the fact that the maintainer of tennisrecord is able to _get_ the data _is_ very much part of the tennisrecord value. Your statement is like saying 'give me two millions and I can get you ~100,000 annually' - well, yeah.....
Sorry, but if all I’m doing is showing a table of data based off of a search, 8 hours is actually an overestimate. I’m talking just the pure UI portion. All queries I need are “open APIs”. Go take a look at PrimeVue. Simple and elegant.

Now, if I have to go get the data & host it? Sure, that’s much harder. But I’m critiquing the tennislink site’s UI, not tennis record, and they have ready access to all of the data.
 
Top