Playing a UTR 12, can my UTR go up from a few games?

[HunterST]

I am a UTR 7 and entered an open tournament. A lot of the field is closer to my level, but one guy is a UTR 12, and, of course, I have him first round.

This is actually okay with me, because I have a goal of getting my UTR to a higher level. From what I've seen with others, the best way to increase your UTR is to play high level opponents. Now, I'm not going to be able to win this match unless the guy gets injured or something. My goal is to get a game or two and even that is fairly lofty.

What I wonder is this: would getting a game off of a UTR 12 make my UTR go up? That would not be an expected result for a 7, but it's not competitive either.

Anyone have a guess based on how their algorithms work?

 

[OnTheLine]

I thought that anytime the UTR difference was greater than 2 (3?) that they didn't score it as a valid match result for ratings purposes .... someone can correct me if I am wrong ...

 

[BallBag]

I thought that anytime the UTR difference was greater than 2 (3?) that they didn't score it as a valid match result for ratings purposes .... someone can correct me if I am wrong ...

Yep, 2.5 difference or higher won't count.

 

[HunterST]

I thought that anytime the UTR difference was greater than 2 (3?) that they didn't score it as a valid match result for ratings purposes .... someone can correct me if I am wrong ...

Yep, 2.5 difference or higher won't count.

Wow, that's stupid. What if I was somehow crazily misrated and I got 7 games off of him or something? I could just stay a UTR 7?

I guess I should also mention that I'm not an official 7. That is my projected rating.

 

[BallBag]

Wow, that's stupid. What if I was somehow crazily misrated and I got 7 games off of him or something? I could just stay a UTR 7?

I guess I should also mention that I'm not an official 7. That is my projected rating.

Not sure if it's the same for a projected rating. If you get your rating up to 10, then the match against the 12 will count.

 

[R1FF]

Im a fan of the UTR system. But this seems like flaw in the algorithm they should try to address.

At least there is a chance of fixing it. If this was USTA fat chance.

 

[HunterST]

Im a fan of the UTR system. But this seems like flaw in the algorithm they should try to address.

At least there is a chance of fixing it. If this was USTA fat chance.

Agreed. I could see not counting 1 or 2 games,as that could be a fluke. If someone has an even semi competitive match with someone more than 2.5 UTR above them, that should definitely be taken into account.

 

[TimeToPlaySets]

I thought that anytime the UTR difference was greater than 2 (3?) that they didn't score it as a valid match result for ratings purposes .... someone can correct me if I am wrong ...

https://blog.universaltennis.com/2019/03/05/understanding-utr-frequently-asked-questions/

 

[BallBag]

Im a fan of the UTR system. But this seems like flaw in the algorithm they should try to address.

At least there is a chance of fixing it. If this was USTA fat chance.

I dropped a few games when playing much lower rates players in the back draw. I knew the score didn't count and I don't think I would have played differently if it did. A real 2.5 difference is massive. Like a 4.0 playing a 5.0. If the ratings are accurate, there's no useful information in that score line. If the ratings are not accurate then they will adjust soon enough.

 

[HunterST]

https://blog.universaltennis.com/2019/03/05/understanding-utr-frequently-asked-questions/

I dropped a few games when playing much lower rates players in the back draw. I knew the score didn't count and I don't think I would have played differently if it did. A real 2.5 difference is massive. Like a 4.0 playing a 5.0. If the ratings are accurate, there's no useful information in that score line. If the ratings are not accurate then they will adjust soon enough.

I'm not sure their logic makes sense to me. They say they don't count it because it's "almost certain to be a blow out."
What? Of course it's almost certain, but it's NOT certain. That's the whole point.

They also say the blowout is not reflective or either player's skill?
Again, what?? If I lost 0 and 0 to a player 2.5 above me, that is definitely reflective of both of our skills. It says the algorithm has rightfully identified us as being pretty disparate in skill level.

On the other hand, If I win a few games per set off of a player 2.5 UTR points above me, that's also definitely saying something about our playing abilities. Either I'm a hell of a scrappy player and am probably rated too low, or the other guy might not be rock solid with all styles of play and is probably a little too highly rated.

 

[esgee48]

You would be a statistical outlier, out there at the 4-6 std deviation level of confidence. That means your results are a fluke. If you play other 4.5's and beat them, then the algorithms would yield meaningful projections. Your UTR would go up in a meaningful manner. If someone who had no tennis match history started to play and instantly started to beat 4.5's handily, the 1st few matches would not count. They probably would start to count after 5 or 6 matches when the results conclusively say the player belongs at 4.5 or higher.

 

[Max G.]

They also say the blowout is not reflective or either player's skill?
Again, what?? If I lost 0 and 0 to a player 2.5 above me, that is definitely reflective of both of our skills. It says the algorithm has rightfully identified us as being pretty disparate in skill level.

On the other hand, If I win a few games per set off of a player 2.5 UTR points above me, that's also definitely saying something about our playing abilities. Either I'm a hell of a scrappy player and am probably rated too low, or the other guy might not be rock solid with all styles of play and is probably a little too highly rated.

Or, he's bored and is messing around in a match he knows he could win in his sleep.

 

[HunterST]

Or, he's bored and is messing around in a match he knows he could win in his sleep.

Could be true, but I don’t think UTR can or should start accounting for mental elements affecting performance. A player who was only one utr point ahead could also be someone who “gets bored” playing someone lower than them, and maybe the match ends up closer than it should. Where does it stop? What about skilled players who get tight and drop games they shouldn’t?

I just think it makes no sense to say nothing can be interpreted from the results in a match that is a mismatch. People can still over or underperform in a meaningful way.

 

[Max G.]

Where does it stop?

Well, there's an obvious answer to that - it stops when the match has a chance of being close, so that the score actually matters. I guess UTR chose the threshold of 2.5 - I’ve got no specific opinion on whether that threshold is too low, too high, or just right.

I just think it makes no sense to say nothing can be interpreted from the results in a match that is a mismatch. People can still over or underperform in a meaningful way.

Of course they CAN overperform in a meaningful way. But people also CAN over/under perform in meaningless ways that do not reflect skill.

The question the ratings designers have to figure out is which matches are more likely to give misleading information than useful information. It seems to me that they concluded that “matches with too high ratings differences” are in that category.

 

[HunterST]

Well, there's an obvious answer to that - it stops when the match has a chance of being close, so that the score actually matters. I guess UTR chose the threshold of 2.5 - I’ve got no specific opinion on whether that threshold is too low, too high, or just right.


Of course they CAN overperform in a meaningful way. But people also CAN over/under perform in meaningless ways that do not reflect skill.

The question the ratings designers have to figure out is which matches are more likely to give misleading information than useful information. It seems to me that they concluded that “matches with too high ratings differences” are in that category.

"Has a chance to be close" is an assumption, though. What if it WAS close, yet their algorithm said it shouldn't be? The algorithm takes priority over reality?

I know a guy who is rated a UTR 2. That is based on several results. However, I know him to be a moderate to low 4.0. If we played, I would beat him something like 2 and 2 or 1 and 2. That would suggest he's more like a 5 and would be much more reflective of his level, but it would not count because I'm rated too highly above him. In fact, I wonder if he has remained at a 2 because most of his matches don't count. That's a flaw in the system.

 

[LJ92]

I've seen them say both -- that results with a greater than 2.50 difference are always excluded, and that they're only excluded if the match is a blowout. In the second case, if your match against the 2.50+ player was actually competitive, then the algorithm would, in fact, take it into consideration for both players.

 

[HunterST]

I've seen them say both -- that results with a greater than 2.50 difference are always excluded, and that they're only excluded if the match is a blowout. In the second case, if your match against the 2.50+ player was actually competitive, then the algorithm would, in fact, take it into consideration for both players.

Now that would make sense.

 

[LJ92]

it would, indeed.

 

[schmke]

Now that would make sense.

Unless the higher rated player is intentionally taking it easy and giving up games to either A) lower is rating for some reason or B) help his buddy raise his rating.

I'm not saying matches between players rated far apart are meaningless, but there is way more opportunity for huge swings in a rating due to how "unexpected" the lower rated player even winning or game or two is. Any rating system I would concoct would give some weight to such results, but not as much as matches played against players closer in rating. With less weight, if those were the only matches played and they consistently did better than expected, eventually the player would get to where they belong. But if they had other results against similarly rated players that rated lower, the unexpected results having less weight would not impact their rating as much.

 

[LJ92]

Yeah, utr gives matches between players who are rated closely more weight than those between players with wider rating gaps.

 

[NortheastTennis]

Where is your open tournament? Would love to hear the final result and your opinion on how accurate you feel your “projected” rating is as well as your opinion on your opponents true rating. Must be a big event to have such large differences in UTR levels present

 

[LJ92]

To our earlier discussion, found this at https://support.myutr.com/support/s...-understanding-the-algorithm-complete-summary:

“The algorithm excludes matches in which . . . c) there is a UTR difference of more than 2.50 and the higher-rated player wins as expected (i.e., in a blowout).”

(Emphasis is my own.)

 

[R1FF]

I dropped a few games when playing much lower rates players in the back draw. I knew the score didn't count and I don't think I would have played differently if it did. A real 2.5 difference is massive. Like a 4.0 playing a 5.0. If the ratings are accurate, there's no useful information in that score line. If the ratings are not accurate then they will adjust soon enough.

Interesting. I see the argument you have made and it makes sense.

You’ve changed my opinion.

 

[power_backhand]

People shouldn’t worry about their rating. Just play. Play the ball.

 

[justinmadison]

A UTR 12 is a USTA 5.5 player. A UTR 7 is somewhere around 4.0 It is hard to imagine a match that would yield any meaningful results.

I doubt you could rally with this guy, much less return a 1st or 2nd serve.

 

[Aestheticsaboveallelse]

A UTR 12 is a USTA 5.5 player. A UTR 7 is somewhere around 4.0 It is hard to imagine a match that would yield any meaningful results.

I doubt you could rally with this guy, much less return a 1st or 2nd serve.

Its hard to believe UTR 12 is still only 5.5 ntrp

 

[FuzzyYellowBalls]

Its hard to believe UTR 12 is still only 5.5 ntrp

He wrote that 4 years ago.

 

[Aestheticsaboveallelse]

He wrote that 4 years ago.

Oh I'd have thought the rating conversion always stays the same. So 12 utr = weak 6.0 now?

 

[Moveforwardalways]

Federer beat me 6-0 6-0. So did my buddy at the club. That means they must be rated the same right?

You can only beat someone 6-0 6-0. You can’t beat them 8-0 8-0. That’s why these matches don’t count.

 

[FuzzyYellowBalls]

Oh I'd have thought the rating conversion always stays the same. So 12 utr = weak 6.0 now?

Not sure lol, I just meant whoever wrote that and commented on this thread heck, might not even playe tennis anymore or read this.

 

[tennis3]

Its hard to believe UTR 12 is still only 5.5 ntrp

That's because NTRP isn't made for a UTR that high. 5.5 is really as high as NTRP goes for all practical purposes. Really, 5.0 NTRP is the top for all practical purposes.

D1 college players will generally range from UTR 12 - 14.

 

[jhick]

Anyone know if HunterST played this UTR 12 guy in the tournament and what were his scores?

 

[nyta2]

4y old thread... but answer is yes
i was a utr7+, played a utr13+
got 3 games (shanks, treeing, <hard to keep focus against someone so bad>, etc...) 1-6, 2-6
utr went up to 11
played more singles (mostly against my level), eventually came back down to reality.

 

[tennis3]

4y old thread... but answer is yes
i was a utr7+, played a utr13+
got 3 games (shanks, treeing, <hard to keep focus against someone so bad>, etc...) 1-6, 2-6
utr went up to 11
played more singles (mostly against my level), eventually came back down to reality.

That's one of the very weird things about UTR. Any idea what happened to the UTR of the 13+ you played?

UTR is a pretty good tool for competitive Juniors that are looking to play in college. That's what it's for. It's got it's problems. But it's still pretty good as one of several tools college teams can use for recruiting purposes.

 

[Aestheticsaboveallelse]

4y old thread... but answer is yes
i was a utr7+, played a utr13+
got 3 games (shanks, treeing, <hard to keep focus against someone so bad>, etc...) 1-6, 2-6
utr went up to 11
played more singles (mostly against my level), eventually came back down to reality.

Utr went up to 11? Lmfao. Also im extremely surprised a 7 utr can win that many games

 

[jhick]

Utr went up to 11? Lmfao. Also im extremely surprised a 7 utr can win that many games

I don't think 7 was an accurate reflection of his UTR. A player with an accurate UTR of 7 would likely get double bageled.

 

[Aestheticsaboveallelse]

I don't think 7 was an accurate reflection of his UTR. A player with an accurate UTR of 7 would likely get double bageled.

Yea must be that I'd think not even one game

 

[tennis3]

I don't think 7 was an accurate reflection of his UTR. A player with an accurate UTR of 7 would likely get double bageled.

That's the weird thing about UTR. The 13+ might just not care enough in that particular match to win 0-0. But UTR will normally punish this guy for giving up 3 games.

That's why a high UTR doesn't play a low UTR. Nothing good can happen.

 

[nyta2]

Utr went up to 11? Lmfao. Also im extremely surprised a 7 utr can win that many games

trust me, i was surprised i even won one... i did have one outstanding service game... sun was an issue, so at some point in our service games we were serving underhand or whatever just to get the ball toss out of the sun...
could also be that his rating was boosted... at the time, it was a utr13...

just checked my match results... i lost 0,3,... doesn't show a point-in-time rating... but his "current" rating was utr12+ (based on his last match from 2021, i played him in 2018)

 

[nyta2]

That's the weird thing about UTR. The 13+ might just not care enough in that particular match to win 0-0. But UTR will normally punish this guy for giving up 3 games.

That's why a high UTR doesn't play a low UTR. Nothing good can happen.

definitely could be an issue of <i can't keep my focus against someone this bad> as i mentioned above, but at the same time, after i thanked him for gifting me 3 games, he insisted he was not giving away anything because he was trying to keep his utr up (because it affected his seedings in future tournaments).
when i play other coaches that know my game (eg. utr9+), they bagel me too... so doesn't take much... but i guess anything can happen in tennis