Bumped up to 4.0 for the 3rd time...

[schmke]

Originally Posted by Ronaldo

Level play may get muddled a bit when a team is allowed to add two players of a higher level to their team.

But this is only for the top-level available in a given league. So if 5.0 is the top-level available, a 5.0 team can field two 5.5 players, but a 4.5 team cannot field 2 5.0 players. So it won't really muddle things at all in the 3.5 to 4.5 levels.

But to the other point about introducing quarter point levels, that should in theory make competition a bit more even and avoid players being stuck in the lower half of a level where they lose more than they win, but it would also result in players being bumped up or down a lot more often.

[OrangePower]

Originally Posted by schmke

But this is only for the top-level available in a given league. So if 5.0 is the top-level available, a 5.0 team can field two 5.5 players, but a 4.5 team cannot field 2 5.0 players. So it won't really muddle things at all in the 3.5 to 4.5 levels.

In most areas, the top-level available for the new 40 and over league is 4.5+, with 2 5.0s allowed on the team. And 40 and over counts towards NTRP AFAIK. So potential for some muddle impact at 4.5, although I don't think it will be significant. No different to 4.5 playing up at 5.0 from a rating calculations perspective.

[Ronaldo]

Really need another age level, 70 and above. Unfair for them to compete with youngsters 55 and up. Super seniors must play during the day though. Evening matches are past their curfew at the Home.

[ronray43]

Originally Posted by Ronaldo

Any thoughts of just blowing up the whole season to get rated down again? Our singles player got lucky and bumped down agin.

Frankly, that was the real goal. Ended up going 1-8 with losses to 4.0s of 0-2; 0-2; 1-4; 2-2; 2-3; 2-2; and a three setter that wend 6-4, 3-6, 0-1. Only win was 7-5, 1-6, 1-0 against a 3.5 who was playing up. Also lost to a 3.5 who was playing up 5-7, 6-3, 0-1. My guess is the single set win against the lower end 4.0 is what kept me from moving down. Both of the 3.5s who were playing up moved to 4.0. Will see what happens this year, but if I get whacked good in the first three matches, I'll stop there!

[leech]

I, too, have found the jump in skill level from 3.5 to 4.0 to be vast. In singles matches vs. other 3.5 players or in doubles matches where the other team's best player was 3.5, I went undefeated. In doubles matches where the other team's best player was a 4.0 (in 7.5 combo league), I went winless. I guess I'll know for sure in the Adult league this spring, but I suspect I'll have difficulty winning matches at the 4.0 level.

[Govnor]

The idea of 3.5, 3.75, 4.0, 4.25, 4.5 is actually very valid given that is where most people are.

[Govnor]

Originally Posted by leech

I, too, have found the jump in skill level from 3.5 to 4.0 to be vast. In singles matches vs. other 3.5 players or in doubles matches where the other team's best player was 3.5, I went undefeated. In doubles matches where the other team's best player was a 4.0 (in 7.5 combo league), I went winless. I guess I'll know for sure in the Adult league this spring, but I suspect I'll have difficulty winning matches at the 4.0 level.

It can be. I've heard the same for every other 0.5 jump as well!

[schmke]

Originally Posted by Govnor

It can be. I've heard the same for every other 0.5 jump as well!

This is the nature of a half point rating system where in theory a half point difference between opponents should result in a 6-0,6-0 match. This is the reason why a quarter point rating system seems attractive.

For example, someone right on the 3.5/4.0 boundary is going to be expected to beat the average 3.5 6-3,6-3 and lose to the average 4.0 6-3,6-3. In fact, they'd be the "pick" in any match against a 3.5 (that isn't themselves about to be moved up) and picked to lose in any 4.0 match (unless they get a 3.5 playing up).

Now, not every player is average and from what I've seen estimating dynamic ratings, an individual player's rating can vary a full half point from match to match so this hypothetical player won't necessarily lose every 4.0 match and win every 3.5 one, but that will be the result more often than not unless the player improves their game.

[maggmaster]

I don't understand, you will improve just by playing at that level. Just play and let the ratings fall as they do.

[schmke]

Originally Posted by maggmaster

I don't understand, you will improve just by playing at that level. Just play and let the ratings fall as they do.

I agree, I was just pointing out why it is "normal" for someone to win all (or nearly all) their matches at 3.5 and then turn around and lose all (or nearly all) their matches at 4.0. I noted this will happen unless they improve, and like you say, playing at a higher level is one of the better ways to improve so they likely will unless they have reached their peak or aren't motivated to do so or any number of other reasons.

[Angle Queen]

I'm very likely one of those players "on the cusp," so to speak. Few 3.5s trouble me, yet I trouble few 4.0s. My 4.0 matches have been competitive (certainly in actual "play" although perhaps not in score) and I don't feel any of the 4.0s felt "cheated" in having to play me when I was playing "up."

But going to a .25 rating system seems just a bit too much. We're blessed in our area to have lots of players and lots of leagues. A finer rating differentiation might work here, but in other places, where they sometimes have trouble fielding more than 2 or 3 teams per league/flight....eh? Besides, team/league play is where I've had the best chance to meet more people/players outside of my (playing) comfort zone....both higher and lower....in both age and skill level.

I wish the algorithm more appropriately rewarded an actual "win" and/or de-emphasized the so-called predicted result. I really don't think a top-end (say) 3.5 should have to destroy a low-end 3.5 -1 and -1. That process, in and of itself, sets up such vast differences within each NTRP.

[schmke]

Originally Posted by Angle Queen

But going to a .25 rating system seems just a bit too much. We're blessed in our area to have lots of players and lots of leagues. A finer rating differentiation might work here, but in other places, where they sometimes have trouble fielding more than 2 or 3 teams per league/flight....eh? Besides, team/league play is where I've had the best chance to meet more people/players outside of my (playing) comfort zone....both higher and lower....in both age and skill level.

Excellent point and one I thought of as well. It appears that they do run quarter-point levels in Atlanta, but they have a huge number of participants so they can do it. Your point is valid for a few, and perhaps more than that, areas of the country where they don't have high levels of participation. To field enough teams, they may be forced to bypass the "lower" quarter-point level and have folks play up.

Originally Posted by Angle Queen

I wish the algorithm more appropriately rewarded an actual "win" and/or de-emphasized the so-called predicted result. I really don't think a top-end (say) 3.5 should have to destroy a low-end 3.5 -1 and -1. That process, in and of itself, sets up such vast differences within each NTRP.

Another excellent point. IMHO, accurate rating systems give more weight to "meaningful" matches/games and whether you blow out a much weaker opponent by enough or not should not be a significant factor as matches against these much weaker (or much stronger) opponents are likely not as meaningful. However, that also may make it harder for the lower level player to improve their rating appropriately when they have a competitive loss against a much higher rated player. It is hard to have a perfect system

[J_R_B]

Originally Posted by schmke

This is the nature of a half point rating system where in theory a half point difference between opponents should result in a 6-0,6-0 match. This is the reason why a quarter point rating system seems attractive.

This is a myth due to one misinterpreted statement by the USTA that is now perpetuated on here endlessly. The fact is that the expected outcome of a match between two players 0.5 apart is not 6-0 6-0. What the statement said is that a 6-0 6-0 match is not unexpected for players with a 0.5 difference. But matches between any two players will have a whole distribution of possible outcomes. If the expected outcome is 6-0 6-0 only 5% of the time (or 1 on 20), then it's still not unexpected, and since there are thousands of matches in the system between players with this ratings difference, seeing some 6-0 6-0 outcomes is normal and expected. They were writing in the context of people who see a 6-0 6-0 outcome and start jumping up and down that that person is automatically misrated or worse, sandbagging. The fact is that the expected outcome (in terms of a mathematical expected value or average) of a match between two players one level apart is probably about 6-2 6-2 or something like that with 6-0 6-0 having enough positive probability that it's not unexpected or unusual to see some in the real results.

I play 4.0. I'm probably near the top of the 4.0 range (I was 14-2 last year but wasn't bumped). I won a match against a 4.0 rated player 6-0 6-0. It was a kid who wasn't bad, but I just happened to get all the games. I also won a match 6-0 6-2 against a 3.5 playing up. The 4.0 kid was a lot better than this guy, but in the course of this match, I happened to lose a couple games. The 6-0 6-0 kid also lost 6-0 6-3 to another college kid who was bumped at the end of the year who can kick my *** all over the court. If we all played a hundred matches, I would probably win 2 or 3 by double bagel against the kid and the other kid would probably win 15-20 by double bagel. That's not how the actual record is, though, and seeing my 6-0 6-0 win in the record is not a sure sign that I am under rated or that my dynamic rating is or should be lower than the other kid who got bumped and who can kick my ***.

You have to be careful about how you interpret statements and expectations keeping in mind that individual match results are a random draw from a distribution and not a fixed measurement of the difference in ratings.

[schmke]

Originally Posted by J_R_B

This is a myth due to one misinterpreted statement by the USTA that is now perpetuated on here endlessly. The fact is that the expected outcome of a match between two players 0.5 apart is not 6-0 6-0. What the statement said is that a 6-0 6-0 match is not unexpected for players with a 0.5 difference. But matches between any two players will have a whole distribution of possible outcomes. If the expected outcome is 6-0 6-0 only 5% of the time (or 1 on 20), then it's still not unexpected, and since there are thousands of matches in the system between players with this ratings difference, seeing some 6-0 6-0 outcomes is normal and expected. They were writing in the context of people who see a 6-0 6-0 outcome and start jumping up and down that that person is automatically misrated or worse, sandbagging. The fact is that the expected outcome (in terms of a mathematical expected value or average) of a match between two players one level apart is probably about 6-2 6-2 or something like that with 6-0 6-0 having enough positive probability that it's not unexpected or unusual to see some in the real results.

I play 4.0. I'm probably near the top of the 4.0 range (I was 14-2 last year but wasn't bumped). I won a match against a 4.0 rated player 6-0 6-0. It was a kid who wasn't bad, but I just happened to get all the games. I also won a match 6-0 6-2 against a 3.5 playing up. The 4.0 kid was a lot better than this guy, but in the course of this match, I happened to lose a couple games. The 6-0 6-0 kid also lost 6-0 6-3 to another college kid who was bumped at the end of the year who can kick my *** all over the court. If we all played a hundred matches, I would probably win 2 or 3 by double bagel against the kid and the other kid would probably win 15-20 by double bagel. That's not how the actual record is, though, and seeing my 6-0 6-0 win in the record is not a sure sign that I am under rated or that my dynamic rating is or should be lower than the other kid who got bumped and who can kick my ***.

You have to be careful about how you interpret statements and expectations keeping in mind that individual match results are a random draw from a distribution and not a fixed measurement of the difference in ratings.

So, perhaps a half point difference doesn't have an expected result of 6-0,6-0, perhaps it is 6-1,6-1. You don't disagree that the USTA does have a table of expected results based on rating differential do you? Is that a myth too? At some differential, a 6-0,6-0 result is what is expected and a result short of that will improve the losing players rating and hurt the winning players rating.

You chose not to quote it but I also said a players performance for a given match can vary as much as half a point. This is because you play better/worse on any given day and how you individually match up with an opponent carries far more weight than what some hypothetical rating say should happen.

That said, if your rating is right at the threshold between levels, that means your performance on average indicates you probably beat the lower level players, but lose to the higher level players, more often than not. Of course, if you beat some higher level players and lose to some lower level players, you can arrive at the same point so this is possible.

This also means that if your performance goes up or down a bit from year to year, you will find yourself yo-yo-ing between the levels like the OP cited.

[ronray43]

Originally Posted by J_R_B

This is a myth due to one misinterpreted statement by the USTA that is now perpetuated on here endlessly. The fact is that the expected outcome of a match between two players 0.5 apart is not 6-0 6-0. What the statement said is that a 6-0 6-0 match is not unexpected for players with a 0.5 difference. But matches between any two players will have a whole distribution of possible outcomes. If the expected outcome is 6-0 6-0 only 5% of the time (or 1 on 20), then it's still not unexpected, and since there are thousands of matches in the system between players with this ratings difference, seeing some 6-0 6-0 outcomes is normal and expected. They were writing in the context of people who see a 6-0 6-0 outcome and start jumping up and down that that person is automatically misrated or worse, sandbagging. The fact is that the expected outcome (in terms of a mathematical expected value or average) of a match between two players one level apart is probably about 6-2 6-2 or something like that with 6-0 6-0 having enough positive probability that it's not unexpected or unusual to see some in the real results.

I play 4.0. I'm probably near the top of the 4.0 range (I was 14-2 last year but wasn't bumped). I won a match against a 4.0 rated player 6-0 6-0. It was a kid who wasn't bad, but I just happened to get all the games. I also won a match 6-0 6-2 against a 3.5 playing up. The 4.0 kid was a lot better than this guy, but in the course of this match, I happened to lose a couple games. The 6-0 6-0 kid also lost 6-0 6-3 to another college kid who was bumped at the end of the year who can kick my *** all over the court. If we all played a hundred matches, I would probably win 2 or 3 by double bagel against the kid and the other kid would probably win 15-20 by double bagel. That's not how the actual record is, though, and seeing my 6-0 6-0 win in the record is not a sure sign that I am under rated or that my dynamic rating is or should be lower than the other kid who got bumped and who can kick my ***.

You have to be careful about how you interpret statements and expectations keeping in mind that individual match results are a random draw from a distribution and not a fixed measurement of the difference in ratings.

From the USTA website, "Are all players in a given NTRP level equal in ability?
No: The NTRP system identifies general levels of ability, but an individual will be rated within those levels at 50 different hundredths of a point. For example, a 3.5 player can fall anywhere between a 3.01 and a 3.50. That is the reason many people feel they are playing sandbaggers – they are closer to the bottom of that range while their opponents are closer to the top of the range.

A typical match result for a player, for example, with a 3.01 rating versus a 3.49 player, both of whom are 3.5s, would be 6-0, 6-0 in favor of the higher rated player.

[J_R_B]

Originally Posted by schmke

So, perhaps a half point difference doesn't have an expected result of 6-0,6-0, perhaps it is 6-1,6-1. You don't disagree that the USTA does have a table of expected results based on rating differential do you? Is that a myth too? At some differential, a 6-0,6-0 result is what is expected and a result short of that will improve the losing players rating and hurt the winning players rating.

You chose not to quote it but I also said a players performance for a given match can vary as much as half a point. This is because you play better/worse on any given day and how you individually match up with an opponent carries far more weight than what some hypothetical rating say should happen.

That said, if your rating is right at the threshold between levels, that means your performance on average indicates you probably beat the lower level players, but lose to the higher level players, more often than not. Of course, if you beat some higher level players and lose to some lower level players, you can arrive at the same point so this is possible.

This also means that if your performance goes up or down a bit from year to year, you will find yourself yo-yo-ing between the levels like the OP cited.

I have heard that the rating system is based on a table of expected results based on rating differential, but since it's not public, no one knows exactly how it works. What I've heard (basically here...) is that the table and the rating calculation are based on total game differentials, not match scores, so it would say something like rating difference 0.50, expected game differential = 8 or something like that. Again, no one but the keepers of the magic formula know for sure and they are sworn to secrecy upon penalty of death. this also means that you can lose matches on the court but win in the rating calculation (consider winning 1-6, 7-5, 1-0, your game score is 9-11).

Of course, you'll beat lower rated players and lose to higher rated players more often than not (as long as the dynamic ratings are accurate), but either predicting or evaluating actual scores is much trickier. Two guys can be a full level apart and play once and the score is 6-0 6-0 and the next time it's 6-4 6-3. The higher rated player will almost always win, but the lower player can get a substantial number of games in any given match. Case in point, again, I am 4.0. I played an open tournament 2 years ago against a 5.0 B rated league player who was a mid 20's teaching pro and former national junior college doubles champ. He is way better than I ever have been or ever will be, but I played the set of my life in the second set and lost 6-1 7-6. This is a match with a rating differential TWO levels apart, and the score still wasn't 6-0 6-0. Even despite the score, I was a long way from actually winning the match, but when it comes to scores, you just never know. Last year in that tournament, I played another mid-20s former college player (D3 ranked school) who would be 5.0 if he played league tennis, and I lost 6-0 6-1.