The USTA rating system and self-rated players

[ohplease]

Originally Posted by JavierLW

The visual verification was really only done for self rating and for play in the playoffs. Unless you are talking about how it may of been before the 90's (I didnt play back then).

Of course you cant go purely by skill ratings, that is impossible, but you would have to admit that games are better than whole matches, wouldnt you???

It's generally accepted that if you lose to someone 6-0, 6-0 that means you are a lot worse than if you are losing 7-6, 7-6 to them.

I have a friend that Ive probally lost 70 match to but the far majority of them are very very close, it certainly doesnt mean he's a whole diffrent level then I am. (we both do about the same in League play and I do better against certain types of opponents)

So it sounds like you agree with me. If you are going to claim that going by games (which is really how it was ALWAYS done in the core rating scheme anyway) is somehow wrong, then you cant somehow claim that whole matches is somehow better.

I agree that score differentials and whole matches are both the same thing - result metrics, and not somehow in any way indicative of "skill levels," as you originally claimed.

That neither of those things cares about skill or style is exactly my point. Make the results matter. Not just for who gets the trophy at the end. If you make the playoffs, even locally, you get bumped. Fail to compete well, you get dropped.

Right now, the USTA errs on the side of not bumping up/down enough. Err on the side of bumping up/down too much and it becomes MUCH harder for teams to work the system.

Heck, there's even an easy way to fix the sandbaggers who go from champs to goats every other year, working that border. First, you bump everyone who makes the playoffs. Then, for every progressive level of competition, you add a year's duration to that bump. Districts? 1 year at the next higher level. Sectional? 2 years. Regional? 3 years. Nationals? 4 years. Champs? 5 years.

There's even a way to fix the dudes who hide out during the regular season, only to come out during the playoffs (a direct result, I might add, of the blatantly wrong headed score differential metric) - you're only allowed to play in the playoffs as often as you played during the regular season. No more hiding, no more score or match management, no more playing to some mystery algorithm.

The point is that there's tons of things the USTA could be doing to fix this, but they're not. Heck, even changing the rules every year would help shake things up. Eventually, they'll stumble on something that works well. As it is right now, it's just a broken system that does actual harm in terms of growing the sport.

[Ace]

I actually have to caveat everything I've said by adding that I think the USTA rating system is actually pretty good.

Its just people with no common sense that are abusing the system.

Also, I think a lot of people call people "sandbaggers" just because they lost a match pretty bad, but that can also happen when someones bad day meets someones good day. Some people don't bother to check someones record or take into consideration that maybe they are in an easy flight (compared nationally) before they start throwing "sandbagger" around.

Sometimes the top team in a flight goes to districts every year, and gets their butts kicked. Well, in their little area, they are good, in a bigger area, they aren't....its a NATIONAL rating system.

Occassionally there is the all out "sandbagger" playing two levels down...but we don't see that too much, every now and then maybe.

[lostinamerica]

Originally Posted by Ace

Occassionally there is the all out "sandbagger" playing two levels down...but we don't see that too much, every now and then maybe.

At nationals, that is about all you see. Everyone is at least one level down and about half of the players are two levels down.

[Cindysphinx]

Ace,

In our league, there used to be a rule to discourage teams from playing up. You couldn't have more than some percentage of your team playing up. Maybe it was 40%, I can't remember.

They changed it for the 2006 season. So now you will see teams where over half the team is playing up. They get totally killed, of course. I mean, it's horrible. One such team had nine 2.5s out of 14 players. They lost all twelves of their matches, winning just 4 individual matches (one by default).

Still, it's OK. I imagine the weaker teams are relieved they have someone they can beat. And the stronger teams focus on destroying these teams to maximize their number of sets and games won for the playoff race. Or the captain can give her weaker players a chance without a real risk of taking a team loss.

On balance, I think the rule change was a good thing. So, uh, unless you're whipping these folks at love, there is a challenge there for you. If I were to encounter a weak player playing up, I would most definitely try to win the first set at love and then work on hitting slice or serve and volley, telling myself that if I actually dropped a game then I would resume playing the regular way.

Oh, one more thing. It is not easy to find players a level up who are willing to play you to give you experience in "playing up," so that's not much of an option for people like me. Besides, even if I lose when I play up, this might help my rating if I don't lose too badly. In my two singles losses at 3.5, I won a total of 10 games. That probably did my rating some good.

[Cindysphinx]

Originally Posted by Ace

I actually have to caveat everything I've said by adding that I think the USTA rating system is actually pretty good.

Ditto!!

I consider myself at the high end of 3.0. I took two doubles losses this season. One was to the computer-rated No. 1 doubles team that is going to sectionals, and we lost 4-6, 5-5 (timed). The other loss was to a different self-rated No. 1 doubles team that is 12-0 and just won a tournament.

I think it speaks well of the rating system that these ladies didn't get DQ'd, despite beating everyone in sight.

[JavierLW]

Originally Posted by lostinamerica

At nationals, that is about all you see. Everyone is at least one level down and about half of the players are two levels down.

Here's the thing. 95% of us dont really care what happens in Nationals.

The teams that happen to win every single year to go there care more which is why they may sandbag in their own local league.

Although I still believe that sandbagging at the local level is not happening as much as ohplease wants to believe (even though he's basing his whole therory on potential sandbagging).

You have to be a good deal better than someone (probally more than one level in a lot of cases) to get away with sandbagging without risking losing, especially at 3.0 / 3.5. Sandbagging means you are switching the momentum to your opponent, and you are making yourself extra tired. Even a 3.0 player can look fantastic on any given day if you give him enough confidence, and 3.5 players cant necessarily just "turn it on" when they need to.

So to base a whole system of believe on potential sandbagging in my mind is silly, unless you are only focused on what happens in the playoffs and not the local leagues. (which is the whole mistake in my mind, teams only care about what happens in a few weekends a year rather than most of us who are signing up for the whole season)

[Topaz]

Originally Posted by Cindysphinx

Ditto!!

I consider myself at the high end of 3.0. I took two doubles losses this season. One was to the computer-rated No. 1 doubles team that is going to sectionals...

Hey, I know who's team that is! She was my roomie at districts!

And yes, Cindy, it used to be a percentage rule, but it was done away with two years ago. My NOVA 3.5 team had quite a bit of 3.0s on it, and we actually played pretty well, sometimes getting wins when the 3.5s didn't. We were all strong 3.0s (well, most of us). My 3.5 team had a bigger problem of not getting enough people to show for all the matches, thus lots of defaults. But for me, at least I got lots of playing time.

And, even though I was a 3.0 playing up, there was only ONE match that I would say was not competitive, and as Javier (I think it was him) said, it was more a matter of our opponents having a great day, and me and my partner having a horrible day. It was one of those matches where afterward, you just shake it off and keep on going.

[kevhen]

Originally Posted by ferocious4hand

how is it the computer's fault if they don't play up? You're the one that played 4.5 and won, they didn't. I don't see how they should be 4.5s simply b/c you can't beat them

This guy was 4.5 before he asked for medical exemption due to surgery. He is my age in his 30's and is playing like a 4.5 again. My serve is my strength but when the match score gets close, he puts hits his returns right back at my feet. He plays doubles with a weak 4.0 guy and they still win like 90% of their matches. This has gone on for 3 years now and he still hasn't gotten bumped up.

[SB]

I think most of the solutions given here to "fix" the NTRP system have far more pitfalls than the NTRP system does itself.

The only thing I'd like to see is wins taken into account, a little bit. If you win 25 of 28 matches, you should move up, even if they were "close."

[Ace]

Originally Posted by SB

I think most of the solutions given here to "fix" the NTRP system have far more pitfalls than the NTRP system does itself.

The only thing I'd like to see is wins taken into account, a little bit. If you win 25 of 28 matches, you should move up, even if they were "close."

If all your matches were "close", you are obviously playing the right level. Why should you move up? Isn't the goal really to have a league full of competitive matches? If you get too good, yeah, you should move up, but "close" matches mean you aren't there yet.

[Ace]

Originally Posted by Cindysphinx

Ditto!!

I consider myself at the high end of 3.0. I took two doubles losses this season. One was to the computer-rated No. 1 doubles team that is going to sectionals

...ok, but if that team gets their butts kicked playing other 3.0's at sectionals, you are not at the "high end" of 3.0.
if they compete well at sectionals, you may be a "high" 3.0.....
...or you may just be "high".

But looks like they did well at districts..... maybe you ain't high after all...

[ohplease]

Originally Posted by Ace

If all your matches were "close", you are obviously playing the right level. Why should you move up? Isn't the goal really to have a league full of competitive matches? If you get too good, yeah, you should move up, but "close" matches mean you aren't there yet.

If your matches are close, but you're still winning in excess of 70% of the time, then you're not at that level. I'm talking over many matches here, say 8-15. Fluke wins that could have gone either way should get you a winning percentage around 40-60, MAYBE. Not 30, not 70. If you always push someone to tiebreaks, but never win, guess what - you're not at their level. I've been on both sides of that equation, and on the side that always wins, guys who push me but can't close don't worry me at all.

As far as "potential sandbagging," not only do I have anecdotal evidence from both myself and others, there's statistical evidence as well. From this thread:

http://tt.tennis-warehouse.com/showthread.php?t=134649

Originally Posted by ohplease

That link confirms what I've long suspected. It used to be (and it's still generally held) that 4.0 was where the peak of the bell curve was in USTA play. In other words, 4.0s were considered "average" players. That peak has now been pushed back to span 3.0 and 3.5. Seems like ample evidence of rating inflation/deflation sandbagging effects to me.

I mean really, 4.0 doesn't sound very good, right? And yet, according to the link, only 1 in 4 players who play seriously enough to be involved in USTA leagues are 4.0 or better. Only 1 in 20 is good enough for 4.5 or better. 5.0? 1 in 100.

Since bell curves traditionally define "average, give or take" as between the 25-75% marks, then demographically, 4.0 is currently where players can start to be described as genuinely better than average, with 4.5 and up as significantly better than average.

In other words, sandbagging has crammed just about everyone into 3.0 and 3.5, which means most club players would do well to improve to 3.5, and very few will ever improve to 4.0 or beyond. That's two strata in a rating system with more than 10.

So if there's no sandbagging effect, how is it that the rating that used to represent the middle of the bell curve now represents the top 25% of tennis players? How is it that teams that dominate regular seasons, and even playoff matches, somehow go down with 6-1 and 6-0 sets to district regulars across all five lines, across multiple team meets?

How is it that what used to be considered "good club player" (ie 4.5), is now in the top 5% of tennis players. If you've got 100 people at your club, that's the FIVE best guys there - except those 5 dudes probably aren't good enough to play 4.5 USTA, because the tennis playing population at large is without a doubt at a lower level at each level compared to alleged USTA ratings.

Most people might not care about nationals, but those who do really are ruining it for everyone. People who are starting out in the game have to hang around for years before they get any positive feedback in the form of wins at 2.5 or 3.0. What motivation is there for them to stick around at that point? How long will they have to play, and how much better will they have to get, to play at the top of the 3.5 range? The answer, nowadays, is pretty much a lifetime. Didn't used to be that way.

In fact, there's such a disparity between USTA bizarro world and the regular tennis world, that when I'm talking to people and they want to talk levels - nothing else matters other than whether or not they've played USTA. I don't care what you think you are, or what you think your buddies are, because in truth most people who think they're pretty good will go down, badly, to somebody on a 3.0 or 3.5 team built for nonlocal competition.

[Topaz]

Originally Posted by Ace

If all your matches were "close", you are obviously playing the right level. Why should you move up? Isn't the goal really to have a league full of competitive matches? If you get too good, yeah, you should move up, but "close" matches mean you aren't there yet.

Exactly, if the wins are close, you are at the correct level, and shouldn't move up.

[AP328]

Originally Posted by Matt21

The USTA should just scrap the numeric rating system and go with a more simple, 4-tiered system: Beginner, Novice, Intermediate, Advanced. The larger percentage of players would rest in the 2nd and 3rd tiers and you'd no longer (ever) be eligible for "beginner" after having played 2 years of league tennis. Novice level would be where your traditional 2.5/3.0 players would play; intermediate would be where your 3.5/4.0 players would play and advanced would be 4.5 and up.

I understand why the USTA adopted the NTRP program and I'm not saying scrap it; just condense it. As I understood it, the whole idea behind NTRP was to make tennis competitive and fun for more people. The problem is that a half-point graduated system that (competitively) begins @ 2.5 and ends @ 5.5 has far too many levels. IMO, condensing the ratings into 4 groups total this would better control the blatant sandbagging and self-ego stroking that occurs while at the same time, increasing the potential for creating better tennis players.

Let's face it, if you're a "traditional" 4.5 playing "advanced" under the new system, you're in there with 5.0 and 5.5 players in which case, if you want to be competitive, you'll have to practice and "bring it" each time you play a league or tourney match. If you're serious about your game, you'll take your beatings early on but you'll get better (just like guys playing ATP who start out ranked in the 500s after leaving juniors, but over the course of 4+ years, get good enough to crack the top 100, 50, 25 or even 10).

Final thought: Can you imagine how convoluted and confused the juniors program would be if they used this same NTRP rating system?

I disagree. Condensing the ratings to four levels would only make the problem worse. I have seen it in recreational softball and volleyball.

I think it is ridiculous that there is no appeal process with self rated dq's. My team actually went to the head of the NTRP, Gail Marx and asked her to review our dq, because beating a computer rated 4.0, 7-5, 6-3 should not give you a strike.

She emailed us back and basically said that our singles player should never have even played 4.0. Yet, ironically, this is what the computer gave her. Our singles player is a division 3, non-scholarship player, at a small college in Wisconsin that isn't even big enought to have dorms. Gail continued to berate us and tell us that an NCAA college player shouldn't be playing 4.0.

And she wonders why we think a mistake has been made. Our player plays NAIA, Division 111, not NCAA. She obviously didn't look at her record in college to see that she was playing other d3 schools that are not even in the UW system.

We now just laugh at the situation, and will probably pull out of the USTA, because no one thinks that a mistake has been made.

Still very confuses.

[AP328]

Originally Posted by Ace

I actually have to caveat everything I've said by adding that I think the USTA rating system is actually pretty good.

Its just people with no common sense that are abusing the system.

Also, I think a lot of people call people "sandbaggers" just because they lost a match pretty bad, but that can also happen when someones bad day meets someones good day. Some people don't bother to check someones record or take into consideration that maybe they are in an easy flight (compared nationally) before they start throwing "sandbagger" around.

Sometimes the top team in a flight goes to districts every year, and gets their butts kicked. Well, in their little area, they are good, in a bigger area, they aren't....its a NATIONAL rating system.

Occassionally there is the all out "sandbagger" playing two levels down...but we don't see that too much, every now and then maybe.

So Ace, as I understand it... when a team does go to sectionals, wins and then goes to nationals, those players should be moved up right?

There is a team we are scheduled to play next week. One of their players went 4-1 last year in sectionals (won 6-2,6-2; 6-4,6-2; two 3rd set tiebreaks). He lost two of three at nationals (6-3,7-5; 6-2,6-4). I don't understand how he didn't get moved up!?!?

Thoughts?

(They also have three self-rated players, added four weeks into the season who have gone 9-2 combined...rarely close)

[Cindysphinx]

Originally Posted by cosmo23

I think it is ridiculous that there is no appeal process with self rated dq's. My team actually went to the head of the NTRP, Gail Marx and asked her to review our dq, because beating a computer rated 4.0, 7-5, 6-3 should not give you a strike.

She emailed us back and basically said that our singles player should never have even played 4.0. Yet, ironically, this is what the computer gave her. Our singles player is a division 3, non-scholarship player, at a small college in Wisconsin that isn't even big enought to have dorms. Gail continued to berate us and tell us that an NCAA college player shouldn't be playing 4.0.

And she wonders why we think a mistake has been made. Our player plays NAIA, Division 111, not NCAA. She obviously didn't look at her record in college to see that she was playing other d3 schools that are not even in the UW system.

We now just laugh at the situation, and will probably pull out of the USTA, because no one thinks that a mistake has been made.

Still very confuses.

I added bold to your post.

I don't think I understand your complaint. I think I recall that when I self-rated, the computer asked me what level I wanted. Sure, if I had a certain background the computer would deny me the requested rating. But I always assumed the responsibility not to self-rate (and play) too low was entirely mine.

Second, your player is the best judge of whether she is a 4.0. Why should this USTA official have to scour the player's record, and how meaningful could this review be, anyway? Adult league is for amateurs, so your player could be a middling college player but still destroy the 40-somethings encountered at 4.0.

And, erm . . . what were the other two strikes?

No, I don't think there should be an appeal. That's just yet another way people would try to game the system. It sounds like your player self-rated too low. Gotta just suck it up, methinks.

As for you, Ace . . .

:stamps foot:

I am *TOO* at the high end of 3.0! I will cry my eyes out if anyone suggests otherwise.

[Cindysphinx]

Originally Posted by Topaz

Hey, I know who's team that is! She was my roomie at districts!

I don't think so. Our districts were a different weekend than yours.

In a different state!

[aidenous]

I have seen some players go undefeated the entire season and not get bumped up. I know of two that were 12-0 and 10-0 and are back for another season. The previous year one of my guys was 7-6 and got bumped so wins & losses doesn't appear to be in the equation and needs to be.

[Topaz]

Originally Posted by Cindysphinx

I don't think so. Our districts were a different weekend than yours.

In a different state!

No, I know...she played VA districts the weekend after your districts. Her 'other' team won their district and is headed to sectionals, and her VA team (my team) is not.

Think....she plays on a bazillion teams in four different counties, if that helps!