The 2008 Ratings Predictions Thread

[Cindysphinx]

Make your predictions on who the computer will move up or down. Then we can see who had the best record in making these predictions.

I predict:

1. Our Superstar 4.0 Dude will move to 4.5.

2. I will stay mired in 3.0 for another year.

3. One female teammate who plays singles and doubles will move from 3.0 to 3.5.

4. One of my male mixed teammates will drop from 4.0 to 3.5 mixed exclusive.

5. Two other of my male mixed teammates (one current, one former) will drop from 3.5 to 3.0.

[edit: 6. One of my 3.0 female mixed teammates will move to 3.5, as she was on a team that went to sectionals. Another of my 3.0 female mixed teammates who went to sectionals will stay 3.0, as she played mostly No. 3 doubles.]

Everyone else on my teams will stay the same.

 

[Topaz]

Well, since a few of my friends went to Nationals, it will be pretty easy to predict what happens!

1. Me - I have no idea. I could go either way...move up to 3.5 (based on undefeated tournament (so far!) and districts play) or stay at 3.0 (based on losses that should have been wins during the regular season). I'm actually ok if I stay at 3.0. I will still play up at 3.5 (doubles and singles) and I've already been recruited for a 3.0 singles spot on a fairly strong team.
2. Regular doubles partner/outdoor team - will go from 3.0 to 3.5
3. My outdoor team's co-captain - will go 3.5 to 4.0
4. Regular doubles partner/indoor team - will stay 3.0 due to broken leg!!!

 

[bleach]

Not making any predictions here... after going 9-2 in the Spring plus 5-2 at Regionals (State) and "not" getting bumped for the early start (only 1 of 3 on my team that didn't), I'm convinced it's all rigged!

 

[Topaz]

I just did some digging around at the USTA site about year end ratings, and I never realized that tournaments and championship matches are weighted heavier. That is good news for me. However, I was already designated as a benchmark...I'm wondering if that will help the players who beat me more than it helps me! I'll have to go back and see if I beat any benchmarks. (I'm thinking there is at least two in there...hmmm.)

 

[raiden031]

I will stay at 3.0 due to mediocre performance, however 37% of the 3.0 males in my league got moved up in 2006, so who knows what will happen. I would expect to get moved up if they counted combo leagues toward the rating.

 

[kylebarendrick]

I expect to remain a 3.5. In this, my first 3.5 season, I had a 14-5 record on two teams. Between the fact that I only played doubles with strong partners and the fact that I was just bumped from 3.0 last year, I expect that this won't be enough to bump me again this year.

Although I've always promised myself I wouldn't appeal a rating increase, I'd be tempted this time. I'm not yet 4.0 material. It is my goal next year though!

 

[Cindysphinx]

I will stay at 3.0 due to mediocre performance, however 37% of the 3.0 males in my league got moved up in 2006, so who knows what will happen. I would expect to get moved up if they counted combo leagues toward the rating.

Raiden, how did you learn that?

I'd be shocked if 37% of the 3.0 women were moved up in my league. I'd guess fewer than 10%. . . .

 

[raiden031]

Raiden, how did you learn that?

I'd be shocked if 37% of the 3.0 women were moved up in my league. I'd guess fewer than 10%. . . .

I went through the tedious process of looking at the tennislink profile of every 3.0 male player in the league in 2006 and checked to see what their current rating is.

 

[JavierLW]

Make your predictions on who the computer will move up or down. Then we can see who had the best record in making these predictions.

I predict:

1. Our Superstar 4.0 Dude will move to 4.5.

2. I will stay mired in 3.0 for another year.

3. One female teammate who plays singles and doubles will move from 3.0 to 3.5.

4. One of my male mixed teammates will drop from 4.0 to 3.5 mixed exclusive.

5. Two other of my male mixed teammates (one current, one former) will drop from 3.5 to 3.0.

[edit: 6. One of my 3.0 female mixed teammates will move to 3.5, as she was on a team that went to sectionals. Another of my 3.0 female mixed teammates who went to sectionals will stay 3.0, as she played mostly No. 3 doubles.]

Everyone else on my teams will stay the same.

I expect I will remain a 3.5. I went 2-4, but that doesnt matter, I think all the matches were close enough related to whoever I was playing.

At least I hope so. A personal rival of mine and me had a bet that one of us would get bumped up to 3.5 before the other. I won last year (he stayed at 3.0, I got moved up to 3.5). It would suck if I got rated DOWN, the one year that he may actually move up to 3.5.

(if I did, I will never play 3.0 again anyway though, it's just more or less about appearance)

 

[aldekeuk]

Anyone know a definitive timeline for when the new 2008 ratings will be posted on TennisLink?

I seem to remember that last year they came out sometime around Thanksgiving...

 

[Topaz]

Anyone know a definitive timeline for when the new 2008 ratings will be posted on TennisLink?

I seem to remember that last year they came out sometime around Thanksgiving...

Yes, it is usually the last week of November, after Turkey Day.

 

[johnkidd]

I get bumped down to 4.0 since I took a year off.

 

[johnkidd]

I just did some digging around at the USTA site about year end ratings, and I never realized that tournaments and championship matches are weighted heavier. That is good news for me. However, I was already designated as a benchmark...I'm wondering if that will help the players who beat me more than it helps me! I'll have to go back and see if I beat any benchmarks. (I'm thinking there is at least two in there...hmmm.)

Topaz-You become a benchmark player when you advance to playoffs of some type. I think you can also become a benchmark player when you have played at the same level for a number of years.

Being a Benchmark player really does not impact you. But say someone self rates and beats benchmark player 1 & 1 that will probably be a strike on the computer for the person that won.

 

[cghipp]

I predict I will not be bumped to 4.0 and my opponents will b!tch and moan about it for another year.

 

[JavierLW]

I just did some digging around at the USTA site about year end ratings, and I never realized that tournaments and championship matches are weighted heavier. That is good news for me. However, I was already designated as a benchmark...I'm wondering if that will help the players who beat me more than it helps me! I'll have to go back and see if I beat any benchmarks. (I'm thinking there is at least two in there...hmmm.)

You're right, it does help them go up, more than it helps you to go up.

Because their entire rating is based on their local league play (where they might of done very well, even against you benchmark players), where your rating is about 50% based on championship play where your results may not necessarily look so hot because you are playing tougher competion.

It was worse in the old days (pre-DNTRP), back then your entire rating was based on the highest level of competition. So you would always have a phenomenon where the best teams won every year and didnt lose anyone, but your 2nd or 3rd place team that just happened to beat someone on that team would get bumped up.

 

[JavierLW]

I get bumped down to 4.0 since I took a year off.

That's not going to happen. (you'll keep your rating)

 

[Ace]

I dont think "Benchmark" has exactly the effect you are describing.

If someones rating is "benchmark" (from last years championship play) and you beat them this year, it doesn't count any more or less than anyone else.

If you beat a person this year, who THIS year goes on to championship play and BECOMES a benchmark THIS year (in November), THEN you are compared to them and it is considered more accurate.

So, its not if you beat last years benchmark, its if you beat the person who is going to be the benchmark THIS year. So, how did you do against the people who moved on to higher championship play THIS year?

"Benchmark" means that you were used to calculate other peoples ratings when ratings were put out that year. The "B's" you are looking at now are already factored into the rating you got at the end of last year. You need THIS years "B's".

 

[JavierLW]

I dont think "Benchmark" has exactly the effect you are describing.

If someones rating is "benchmark" (from last years championship play) and you beat them this year, it doesn't count any more or less than anyone else.

If you beat a person this year, who THIS year goes on to championship play and BECOMES a benchmark THIS year (in November), THEN you are compared to them and it is considered more accurate.

So, its not if you beat last years benchmark, its if you beat the person who is going to be the benchmark THIS year. So, how did you do against the people who moved on to higher championship play THIS year?

"Benchmark" means that you were used to calculate other peoples ratings when ratings were put out that year. The "B's" you are looking at now are already factored into the rating you got at the end of last year. You need THIS years "B's".

If you were talking to me, I was talking about a situation that used to happen a lot where the benchmarks were always benchmarks. (the same team wins every year)

Otherwise it sounds like we are talking about the same thing, you just made it sound better.

 

[Ace]

No, I wasn't talking specifically about you. But a lot of people think, "oh, i beat someone who is a benchmark, my rating is going up"....and then in the meantime, get their butts kicked by their "real" benchmark for this year (the one who WILL BE going to districts), and then wonder why they didn't get bumped up.

if you don't win your flight, you need to beat the person who is on the team that wins your flight this year (assuming that person will actually play in districts), not beat the team that got benchmarked for going to districts last year.

 

[Topaz]

Topaz-You become a benchmark player when you advance to playoffs of some type. I think you can also become a benchmark player when you have played at the same level for a number of years.

Being a Benchmark player really does not impact you. But say someone self rates and beats benchmark player 1 & 1 that will probably be a strike on the computer for the person that won.

But *I* was a benchmark player this last season (I went to districts last season as well as this season)...so if someone beat me who wasn't a benchmark, that 'helps' them, right? Also, if I beat another benchmark, good for me.

If someone who self rates beats a benchmark, it factors into their Dynamic Rating...it doesn't generate a strike. Right?

I'm not sure if I'm right on any of this, sooo...feel free to correct me! Either way, I still think I'm borderline!

 

[Ace]

if you were benchmarked for last year (so this year you are 3.0B), and didn't go to districts this year... then no, it isn't any more special to beat you or some other non-benchmark player. its just a regular calculation of dynamic rating.

after all the league play is over, the people who went to championship play are "benchmarked". other peoples ratings are then compared to theirs.

ratings are then calculated based on 1/2 your dynamic rating and 1/2 your comparison to this years benchmarks.

however, in your case, you will become a Benchmark this year too, so yes, peoples calculations against you are weighed more heavily this year.

Also, "helps" or "hurts" is all relative...you can be a low-3.0 player, but still be a "Benchmark". it just means your rating is "more accurate", not necessarily better or "high-3.0".

 

[JavierLW]

But *I* was a benchmark player this last season (I went to districts last season as well as this season)...so if someone beat me who wasn't a benchmark, that 'helps' them, right? Also, if I beat another benchmark, good for me.

If someone who self rates beats a benchmark, it factors into their Dynamic Rating...it doesn't generate a strike. Right?

I'm not sure if I'm right on any of this, sooo...feel free to correct me! Either way, I still think I'm borderline!

No, in local season play, your rating is affected by their rating, it doesnt matter if it's benchmark or computer generated. (it's not weighed any diffrently)

benchmark just indicates that part of your rating is based on post season play, it doesnt really do anything special once you have that rating. (except for that it doesnt go away for X amount of time, etc....)

 

[XkwizitOvrhed]

Maybe you all can predict what will happen with my rating:

I started the 2007 USTA season a computer rated 3.5. My 3.5 Women's league partner started the season as a self-rated 3.5. She had been a 4.0 some 5 years ago and when she re-applied to play this season she appealed down to a 3.5 (This is what she told me because I did not know her prior to being paired with her by the captain of our team)

Anyway, we played 8 matches, won all of them handily (usually 6-2, 6-3) with the exception of one match that we won against a Benchmark team. (Yes both players were benchmarks because they'd both played in Districts and Sectionals last year) We beat this Benchmark team 3-6, 6-2, 1-0.

Now my team ended up in 3rd place at the end of the season and the team with the Benchmarks my partner and I beat went on to Districts, Sectionals and then Nationals. That team lost in the first round of Nationals.

But rewind back to August when the Early Start Ratings came out and my self-rated partner is listed on the ESR as being bumped up to 4.0. I on the other hand am not listed at all so I guess I won't be getting bumped up.

What really sucks is that next season she'll have to play up on a 4.0 team and I won't get to play with her on that 4.0 team because the only 4.0 team at my club only picks up 4.0's.

Oh and one last bit of info that may be useful: that Benchmark team that my partner and I beat back in May that went on to Districts, Sectionals and Nationals? Well one of the 2 players blew her knees out and couldn't play post season while her partner went on to the playoffs without her. However, she lost both matches she played in one of which was Sectionals and one in Nationals.


Anyway, I think I get why my self-rated partner got bumped up (I'm guessing she generated a couple of strikes) but I'm wondering if I still have a shot at getting bumped up when the 2008 ratings come out later this month?

 

[Ace]

Anyway, we played 8 matches, won all of them handily (usually 6-2, 6-3) with the exception of one match that we won against a Benchmark team. (Yes both players were benchmarks because they'd both played in Districts and Sectionals last year) We beat this Benchmark team 3-6, 6-2, 1-0.

My original point is that just because they were benchmarks LAST year, they don't count as the "benchmarks that your scores will be compared to" THIS year......
....however, then you do go on to say that this same team went to districts THIS year, so they will be Benchmarks upon which your rating is based.
I believe the ESRs are based solely upon the "Dynamic" portion of your rating. Since the dynamic portion of your rating was not 4.0, it will be all up to your comparison to that "Benchmark". If it looks like those people get bumped up, chances are good that you will to, because you had a close match against them.
Hard to tell in this case.... Did the player that lost in the Sectional and Nationals at least put up a good fight?

 

[Topaz]

if you were benchmarked for last year (so this year you are 3.0B), and didn't go to districts this year... then no, it isn't any more special to beat you or some other non-benchmark player. its just a regular calculation of dynamic rating.

after all the league play is over, the people who went to championship play are "benchmarked". other peoples ratings are then compared to theirs.

ratings are then calculated based on 1/2 your dynamic rating and 1/2 your comparison to this years benchmarks.

however, in your case, you will become a Benchmark this year too, so yes, peoples calculations against you are weighed more heavily this year.

Also, "helps" or "hurts" is all relative...you can be a low-3.0 player, but still be a "Benchmark". it just means your rating is "more accurate", not necessarily better or "high-3.0".

Just to clarify...I went to districts at the end of 2006, became a benchmark player for 2007 season, and went to districts again this year, in 2007. So, I will be a benchmark again next year, correct? That is my understanding...however, if I get bumped to 3.5, will I be a benchmark then?

 

[XkwizitOvrhed]

The benchmark designation means a player had gone on to a playoff the previous year so yes if you get bumped up later this month to a 3.5 you'll be a 3.5 benchmark for next season. (That same scenario happened to a friend of mine last year)

And to answer your question Ace, that benchmark I beat who played 2 matches and lost at Regionals and Nationals? Yes they were close matches, Regionals she lost 3-6, 5-7 and Nationals she lost 6-7, 6-2, 0-1.

 

[JavierLW]

Just to clarify...I went to districts at the end of 2006, became a benchmark player for 2007 season, and went to districts again this year, in 2007. So, I will be a benchmark again next year, correct? That is my understanding...however, if I get bumped to 3.5, will I be a benchmark then?

No matter what rating you get it will be a benchmark rating.

(but being a benchmark doesnt mean anything special for anyone during the next season anyway, it's just indicating that your rating is partially based on post season play)

 

[Topaz]

Ahh, ok, thanks Javier and Xkwizit, I always did get confused with all that benchmark stuff. One of my captains had made a big deal out of it, so I was never sure quite what it meant, though I knew how I had ended up that way.

 

[Cindysphinx]

Am I correct that a benchmark rating at a particular level (say, 3.5) can be high or low? In other words, if a team of 20 players goes to sectionals and every player plays a match at some point during the playoffs, they all become benchmark players but the really good ones have higher ratings than the others?

 

[XkwizitOvrhed]

A benchmark designation is nothing more than a sign that a player has gone on to post season playoffs. It's neither high nor low, it's just a sign.

Now, ratings themselves can be either high or low. For example a 3.5 rated player can in reality be anywhere from a 3.01 to 3.50, which is why you hear people talk about a low rated 3.5 player, a medium rated 3.5 player and a high rated 3.5 player. The exact formula that the USTA computer uses to calculate whether a player is a low, medium or high end rated player is, from what I've read, proprietary, so I can't tell you how it's calculated but it is calculated every time you play a match and the result of that calculation is what's known as your Dynamic NTRP. Now say you're a low end 3.5 rated player and you play a high end rated 3.5 and blow them off the court then your "true" rating is going to jump up depending on the margin of the scores as well as the "true" rating of your opponent. If your "true" rating jumps up past your current rating (3.5) then you're going to get a strike; 3 strikes and you're dq'd for play at that level. The rub is that no one except higher ups in the USTA knows what a person's "true" rating is and the higher ups are only made aware of a player's "true" rating when a strike or strike(s) occur, at which point (I'm told), the player's Local League Coordinator contacts the player's team captain with the news of a DQ.

The whole process is complicated by the fact that the USTA doesn't divulge the exact formula they use nor do they allow the players to know their "true" ratings. (I understand that the reason they don't allow that information to go public is because then players could manipulate scores and other things to their advantage, which I find completely believable).


To wrap up, I'm told that a player's rating for the upcoming season is based half on their Dynamic NTRP for the season they just played and half on the scores any Benchmarked players had in post season play. See how it gets way complicated? Anywhoo, beating a Benchmarked player does not guarantee a rating bump up, there are too many other factors involved in the calculation that we players aren't privy to.


Hope this makes some sense.

 

[raiden031]

I have been told that benchmark ratings are determined by human verifiers, not by the computer. I don't know if this is true or not, but it would certainly make sense to me.

The reason why players in the playoffs are special (aka. benchmark), is because they have competed against players in far away places so all the league players' ratings are based on how they do against the benchmark players since their ratings are more meaningful. This is simply a way to keep ratings on the same scale across the entire country. Of course in theory it sounds good but doesn't really work in practice for whatever reasons.

So if a guy gets to Nationals at the 3.0 level, and the verifiers (or computer) think that he is a 4.0 based on how performs at 3.0 Nationals, then everyone that played well against him during the regular season will all get bumped up to 3.5 or 4.0 as a result of him being bumped up. If he proves at Nationals that he sucks, then everyone that played competitively against him in regular season will likely stay at the same level. This algorithm then trickles down to all the players, since everyone that played a match will be connected at least indirectly to the benchmark players. There are other factors but this is just a simplified example.

 

[Cindysphinx]

My understanding is that there is no such thing as a human "verifier" who goes out and watches players play.

True? False?

 

[Ace]

This is what the USTA website says:

"19. Will there still be verifiers at District/Section/National league championship events? What is their role there?

There will be a limited number of verifiers at National championship events. Their focus will be on insuring rating accuracy in same-partner situations and out-of-contention teams. Whether verifiers will be assigned to Sectional or District championship events has not yet been determined. If verifiers do attend, their role will be limited to observing the Dynamic NTRP process and providing feedback to improve that system. Verifiers will not gather or report visual ratings, nor will they initiate any disqualification proceedings."

 

[raiden031]

This is what the USTA website says:

"19. Will there still be verifiers at District/Section/National league championship events? What is their role there?

There will be a limited number of verifiers at National championship events. Their focus will be on insuring rating accuracy in same-partner situations and out-of-contention teams. Whether verifiers will be assigned to Sectional or District championship events has not yet been determined. If verifiers do attend, their role will be limited to observing the Dynamic NTRP process and providing feedback to improve that system. Verifiers will not gather or report visual ratings, nor will they initiate any disqualification proceedings."

So the verifiers have no effect on ratings other than helping find ways to tweak the algorithm for future events?

 

[DANMAN]

So the verifiers have no effect on ratings other than helping find ways to tweak the algorithm for future events?

Verifiers determine a certain percentage of your rating if you advance to nationals only. I think it may be 50%. This was the case when I went to Nationals in 2005.