Self-Rates - Sandbaggers or optimistic on average?

[schmke]

This is the time of year when the complaints about the self-rated players at Nationals come out. So I thought I'd do some analysis on whether self-rated players generally do a good job, or if they tend to over/under rate on average.

See http://computerratings.blogspot.com/2016/10/do-usta-league-self-rated-players-self.html for the full blog, but it seems that men are more likely to under-rate or sandbag resulting in more self-rated players being bumped up than down, particularly at lower levels, while women are more likely to over-rate and have more self-rated players get bumped down.

 

[G A S]

Interesting analysis, it must be really be sad to be a sandbagger, unable to win on a higher level, they just stay on a lower level to win anything... that is just sad.

 

[NTRPolice]

I wonder if there is a correlation between women overrating and the use of makeup. Shots fired.

On a more serous note:

Thank you for your effort. I love reading conclusions based on actual sets of data. The only challenge I have is that the DNTRP of opponents in unknown. I'm not saying that I believe this, but there could be more reasons why men are bumped than simply sandbagging. If, for example, mens ratings are more highly distributed through the ranks, winning even once could mean more ratings increase on avg. If the ranks arnt distributed the same in womens, winning once as a woman could mean less DNTRP impact. We cannot assume that DNTRP is distributed in a bell curve throughout the ranks and across the genders.

Your analysis is definitely believable. However, correlation =/= causation.

 

[Moveforwardalways]

I think self rates just make honest mistakes sometimes, and sometimes "do what they have to do". I think self rates who come in intentiqlly low in order to win nationals or whatever are uncommon. The NTRP system is not intuitive and it is sometimes hard to tell where you should be until you play for a while.

 

[schmke]

I think self rates just make honest mistakes sometimes, and sometimes "do what they have to do". I think self rates who come in intentiqlly low in order to win nationals or whatever are uncommon. The NTRP system is not intuitive and it is sometimes hard to tell where you should be until you play for a while.

This is certainly true. Some, even perhaps much, of what the data shows could be honest mistakes. But why then do men and women seem to make honest mistakes in different ways?

 

[Startzel]

I think self rates just make honest mistakes sometimes, and sometimes "do what they have to do". I think self rates who come in intentiqlly low in order to win nationals or whatever are uncommon. The NTRP system is not intuitive and it is sometimes hard to tell where you should be until you play for a while.

I just don't understand how it can be an honest mistake. These players have to have played other guys with usta ratings before joining a league, right?

 

[Moveforwardalways]

This is certainly true. Some, even perhaps much, of what the data shows could be honest mistakes. But why then do men and women seem to make honest mistakes in different ways?

I am not totally sure, but I think it is a social factor. As a male (and I have seen this happen multiple times), if you self rate too high you will be kicking rocks rather than playing tennis matches on a league team. Then you will be stuck in no mans land, because you can't get a chance to play enough matches to lose enough to get bumped down. You will almost certainly have to file an appeal to get bumped down. In this scenario, if there is even a shred of doubt about your skill level, you have to "round down" when self rating. This is especially true if you do not know anyone, new to town, etc. By contrast, there must be some social factor for women that gives them more margin for error in self rating, so they feel confident enough to rate more aggressively.

 

[Moveforwardalways]

I just don't understand how it can be an honest mistake. These players have to have played other guys with usta ratings before joining a league, right?

Maybe so. But even that is not fool proof. Hitting around is different than match play. Everybody looks good just hitting around. And even then, players often get feedback like "I'd say you are a high 3.5 or low 4.0" or a captain telling them "I can't use you at 3.5 but at 3.0 you'll play". What are they supposed to do with that? Then you read the website and it talks about 3.5s being able to rally moderately paced balls with dependability and directional control. Sounds good, but when was the last time you saw that in a 3.5 singles match? And what is moderate pace anyway. You truly never know until you step on the court in a match play situation and see the final scores.

 

[NTRPolice]

This is certainly true. Some, even perhaps much, of what the data shows could be honest mistakes. But why then do men and women seem to make honest mistakes in different ways?

What if the DNTRP avg is different for men and women?

If the avg DNRTP for men is higher than women, then that could be a reason why more men get bumped when they self rate.

How active are men in league vs. women?

If men play more matches on avg, then there is more opportunity to earn DNTRP. This, combined with a higher avg DNTRP could be a mixture biased for more bumps on avg.

I have no idea if either of that is true, but those are possible reasons why more men get bumped than women. I dont think we have reached the conclusion that "men are more likely to sandbag than women" quite yet.

EDIT: Just thought of something else... What DNTRP does someone receive when they self-rate? Is this number always the same? Does it change as DNTRP medians/avg change? If the starting DNTRP of self-rating is not the same between men and women, that could also alter YER's.

 

[schmke]

What if the DNTRP avg is different for men and women?

If the avg DNRTP for men is higher than women, then that could be a reason why more men get bumped when they self rate.

Not sure I follow. When looking at thousands of players in each case, you still expect a relatively even distribution within each level. More players being bumped up than down simply means those players apparently self-rated too low on average. Unless you are suggesting all self-rated men tend to be in the upper part of the range regardless of what level they self-rate at and more likely to be bumped up with a little improvement while self-rated women tend to be in the lower part of the range and more easily get bumped down?

How active are men in league vs. women?

If men play more matches on avg, then there is more opportunity to earn DNTRP. This, combined with a higher avg DNTRP could be a mixture biased for more bumps on avg.

I think it is quite the opposite. Women tend to play more matches than men.

For our self-rated population, the men played the following number of matches (total and average per self-rated player):

2.5 - 7,386 - 10.1
3.0 - 37,244 - 10.9
3.5 - 55,630 - 10.6
4.0 - 40,199 - 10.3
4.5 - 13,444 - 9.1
5.0 - 2,696 - 7.5

For the women, it was:

2.5 - 45,577 - 10.8
3.0 - 67,680 - 11.8
3.5 - 49,891 - 11.2
4.0 - 20,164 - 10.4
4.5 - 4,977 - 8.8
5.0 - 887 - 7.3

At the levels in question where the bulk of the self-rating goes on, the women play on average 0.6-1.1 more matches than the men. Only at 4.0 and higher do they become near equal or the men play slightly more. These are match counts from 18+, 40+, and 55+.

I have no idea if either of that is true, but those are possible reasons why more men get bumped than women. I dont think we have reached the conclusion that "men are more likely to sandbag than women" quite yet.

I agree one can't conclude men are more likely to sandbag than women, but the data certainly supports that more than not.

 

[S&V-not_dead_yet]

This is certainly true. Some, even perhaps much, of what the data shows could be honest mistakes. But why then do men and women seem to make honest mistakes in different ways?

There's a fair amount of research that the mistake is not random: men, for example, will embellish their resume more and apply for jobs for which they are not qualified more than women. Men will more aggressively ask for the raise/promotion and be more willing to "pull the trigger" if their demands are not met.

Translated to self-rating, one would think that men would over-rate and women would under-rate. Maybe men under-rate because they are hungrier for the win and the status that brings [moving up in the social hierarchy] whereas women over-rate because the rating itself is part of the social power?

if women using makeup is overrating, then so are men buying expensive cars.

[S&V retreats to the BL, ready for the incoming fire.]

 

[OrangePower]

Before drawing conclusions, a few things to point out:
- Comparing numbers between men and women, about the same number of women self-rate at 4.0 as men at 4.5, at 3.5 as men at 4.0, at 3.0 as men at 3.5, and (roughly) at 2.5 as men at 3.0. So perhaps this suggests that the distribution of men and women players is such that relatively speaking, level n for women self-rates is the same as level n+0.5 for men.
- For men, it's only up to the 3.5 level that more self-rates are bumped up than down, whereas at 4.0 and above, this is not the case.
- If we consider the differences in self-rates distribution between men and women (w=m+0.5), this suggest that we should expect women self-rates to be bumped up more so than down at the 2.5 and 3.0 levels, but not at 3.5 and above. And in fact this is the case.
So perhaps it's not due to any inherent behavioral difference between men and women, but rather due to the fact that at lower/mid levels it is easier to underestimate how quickly you will improve over the first few months of playing competitive tennis, and the definition of lower/mid levels is just different for men and women (for men up to 3.5, for women up to 3.0, based on the distribution of self-rates.

 

[SeeItHitIt]

Maybe so. But even that is not fool proof. Hitting around is different than match play. Everybody looks good just hitting around. And even then, players often get feedback like "I'd say you are a high 3.5 or low 4.0" or a captain telling them "I can't use you at 3.5 but at 3.0 you'll play". What are they supposed to do with that? Then you read the website and it talks about 3.5s being able to rally moderately paced balls with dependability and directional control. Sounds good, but when was the last time you saw that in a 3.5 singles match? And what is moderate pace anyway. You truly never know until you step on the court in a match play situation and see the final scores.

I agree, until matches are played and results are in, a finger in the wind to self rate is as good as any method. I have always asked that 'moderate pace' question...must be like obscenity/pornography (I can't describe it, but know it when I see it).

The NTRP guideline rating descriptions themselves don't mention a word about what you're expected to do on the court (for win/loss) to obtain/maintain a rating. It does say that your win/loss will ultimately determine your rating, but to self rate? The guidelines include (pretty vague) descriptions about skill sets. I find the descriptions are a half rating over what 'is'. the following is the 4.5 guideline, but reads more like the 4.0:

This player has begun to vary the use of pace and spins, has good movement, can control distance and depth of shots, and is beginning to develop game plans according to strengths and weaknesses. This player can hit the first serve with power and accuracy and can place the second serve. This player tends to overhit on difficult shots. Aggressive net play is common in doubles.

Heck, 3.5's do 2/3rds of this, and 4.0's certainly do most/all of these things.

I know of computer rated 3.5 guys who only care about playing (8.0) mixed. They play men's 3.5 to lose so they don't get bumped to 4.0 as they aren't 'desirable' to the mixed teams as 4.0 guys (true story). Sandbagging the NTRP system knows no bounds and not restricted to the self rates.

 

[Startzel]

Maybe so. But even that is not fool proof. Hitting around is different than match play. Everybody looks good just hitting around. And even then, players often get feedback like "I'd say you are a high 3.5 or low 4.0" or a captain telling them "I can't use you at 3.5 but at 3.0 you'll play". What are they supposed to do with that? Then you read the website and it talks about 3.5s being able to rally moderately paced balls with dependability and directional control. Sounds good, but when was the last time you saw that in a 3.5 singles match? And what is moderate pace anyway. You truly never know until you step on the court in a match play situation and see the final scores.

If a captain is telling them to self rate at 3.0 instead of 3.5 to win that isn't an honest mistake.

All you're doing is justifying blatant cheating. That's why cheating is so prevalent.

 

[Moveforwardalways]

If a captain is telling them to self rate at 3.0 instead of 3.5 to win that isn't an honest mistake.

All you're doing is justifying blatant cheating. That's why cheating is so prevalent.

Is not an honest mistake by the captain, but what is the player supposed to do? Rate high and not get asked to play?

 

[Startzel]

Is not an honest mistake by the captain, but what is the player supposed to do? Rate high and not get asked to play?

So you're arguing unless he cheats he can't play?

Seriously?

 

[beltsman]

Sandbagging really annoys me because it turns people off from tennis. It trickles down. New players get discouraged when they play at 2.5 or 3.0 and get obliterated 6-0 6-0. Sandbagging really sickens me.

 

[S&V-not_dead_yet]

Sandbagging really annoys me because it turns people off from tennis. It trickles down. New players get discouraged when they play at 2.5 or 3.0 and get obliterated 6-0 6-0. Sandbagging really sickens me.

If they are losing 0 & 0 every time I can see how they'd get discouraged. An occasional double bagel though? That's going to happen even in a perfect world of no sandbagging.

How many people do you know have lost like this to self-rated players and then quit? I don't know any.

The only solution that seems to stand a chance is the "no post-season play for self-rated players"; some kind of probationary period. Then the sandbagging will shift to rated players taking matches. Then maybe you put in some sort of 2-speed system whereby you can improve at any rate possible but you can only drop in level at a slower rate.

 

[beltsman]

If they are losing 0 & 0 every time I can see how they'd get discouraged. An occasional double bagel though? That's going to happen even in a perfect world of no sandbagging.

How many people do you know have lost like this to self-rated players and then quit? I don't know any.

The only solution that seems to stand a chance is the "no post-season play for self-rated players"; some kind of probationary period. Then the sandbagging will shift to rated players taking matches. Then maybe you put in some sort of 2-speed system whereby you can improve at any rate possible but you can only drop in level at a slower rate.

I have friends that picked up tennis and joined some leagues, and they get discouraged when they aren't even having competitive matches. Bagels happen, sure. But it's stupid if there is no beginner floor for beginners to play on. A 2.5 should be playing another 2.5 aka beginner.

 

[Startzel]

Sandbagging really annoys me because it turns people off from tennis. It trickles down. New players get discouraged when they play at 2.5 or 3.0 and get obliterated 6-0 6-0. Sandbagging really sickens me.

Precisely why I think it's such a big deal. It's amazing how many people choose not to play league in my community solely because of it.

 

[beltsman]

Precisely why I think it's such a big deal. It's amazing how many people choose not to play league in my community solely because of it.

Yep. I mean some guy sandbags 5.0 to 4.5, it's not the end of the world. But it trickles down because of the downward force it exerts and it hurts the sport as a whole because it turns new players away. Beginners leagues need to really be strictly enforced for beginner-only.

 

[Booger]

Beginners leagues need to really be strictly enforced for beginner-only.

Who is going to enforce it and how do you propose they do it?

 

[S&V-not_dead_yet]

I have friends that picked up tennis and joined some leagues, and they get discouraged when they aren't even having competitive matches. Bagels happen, sure. But it's stupid if there is no beginner floor for beginners to play on. A 2.5 should be playing another 2.5 aka beginner.

That is lame, no doubt about it. The only remedy that they control is to focus on the game and the process, not the rating and the winning. They will get over the hump quickly. Have you advised them about their sandbagging opponents? That could at least ameliorate the discouragement.

 

[beltsman]

Who is going to enforce it and how do you propose they do it?

Oh I don't know, I'm just talking generally.

 

[beltsman]

That is lame, no doubt about it. The only remedy that they control is to focus on the game and the process, not the rating and the winning. They will get over the hump quickly. Have you advised them about their sandbagging opponents? That could at least ameliorate the discouragement.

Yes, I do. Because sometimes I know the players, or sometimes I watch them play, and I'm like lol yeah that guy isn't a 3.0.

 

[S&V-not_dead_yet]

Yep. I mean some guy sandbags 5.0 to 4.5, it's not the end of the world. But it trickles down because of the downward force it exerts and it hurts the sport as a whole because it turns new players away. Beginners leagues need to really be strictly enforced for beginner-only.

What if they are a beginner but highly athletic? They'll still double bagel many opponents.

So you can't judge on tennis experience. So do you then start measuring people's athleticism? Then people will start sandbagging THAT. I see it spiralling into a never-ending series of measures and counter-measures which will take more away from the game than it adds ['the juice is not worth the squeeze"/"the cure is worse than the disease"].

 

[Moveforwardalways]

What is the alternative for new players? Have mandatory USTA tryouts with an official who assigns a fixed rating? We need more people playing not less. Barriers to entry do not help. Then people will just tank the tryout. You can't avoid sandbagging, and honestly the problem isn't not that bad unless you want to go to nationals. If you just want to play good tennis in your league on a Tuesday night, sandbagging probably isn't on your mind.

 

[beltsman]

What if they are a beginner but highly athletic? They'll still double bagel many opponents.

So you can't judge on tennis experience. So do you then start measuring people's athleticism? Then people will start sandbagging THAT. I see it spiralling into a never-ending series of measures and counter-measures which will take more away from the game than it adds ['the juice is not worth the squeeze"/"the cure is worse than the disease"].

What are you talking about?

 

[S&V-not_dead_yet]

What are you talking about?

I'm talking about your solution of having a beginners-only league would still result in double bagels by beginner tennis players who are highly athletic [vs beginner tennis players who are not athletic at all]. That's a mismatch just as big as someone sandbagging.

My point was that a beginners-only league would have problems as well; it would not be a panacea.

 

[Startzel]

What is the alternative for new players? Have mandatory USTA tryouts with an official who assigns a fixed rating? We need more people playing not less. Barriers to entry do not help. Then people will just tank the tryout. You can't avoid sandbagging, and honestly the problem isn't not that bad unless you want to go to nationals. If you just want to play good tennis in your league on a Tuesday night, sandbagging probably isn't on your mind.

There need to be serious penalties to captains that allow cheating.

The answer is for players to better enforce themselves.

 

[Moveforwardalways]

There need to be serious penalties to captains that allow cheating.

The answer is for players to better enforce themselves.

So you are a new player. Would you rather play tennis or fall on your sword alone to prove a point?

 

[beltsman]

I'm talking about your solution of having a beginners-only league would still result in double bagels by beginner tennis players who are highly athletic [vs beginner tennis players who are not athletic at all]. That's a mismatch just as big as someone sandbagging.

My point was that a beginners-only league would have problems as well; it would not be a panacea.

You're missing the point. It has nothing to do with avoiding bagels.

 

[S&V-not_dead_yet]

You're missing the point. It has nothing to do with avoiding bagels.

OK; forget about the score. What I meant emphasize was not the score but the fact that a beginner's-only league still has some flaws in that a beginner who is very athletic will pummel a beginner who is not. This might lead to the second beginner getting discouraged just as if he had gotten pummeled by a sandbagger. If your goal is to eliminate such uneven matchups, a beginner's league will not accomplish it. It will theoretically reduce the # of such uneven matchups by eliminating the sandbaggers if you can come up with a metric that's enforceable.

The best short-term solution is to get them to realize that the enjoyment of the game and process of improvement are more important than winning and ratings. If they strongly disagree, maybe they ought to find a different endeavor.

I saw sandbagging in 2-day volleyball tournaments where a team would bring their B squad on Saturday, do mediocrely [is that a word?], and then show up with their A squad on Sunday and crush everyone. But this was the rare exception, not the rule. The unwritten code of conduct was that you just didn't do that and you'd be shunned if you did. It happened to my team a few times: that's life. We played, had fun, got our clocks cleaned, and then went out to dinner.

One year, my team reverse-sandbagged, in that we were supposed to come in 3rd in our pool of 4 at Nationals and we ended up winning, thus getting placed in the Gold division and getting creamed. We all agreed it was a thrill to play up.

 

[beltsman]

OK; forget about the score. What I meant emphasize was not the score but the fact that a beginner's-only league still has some flaws in that a beginner who is very athletic will pummel a beginner who is not. This might lead to the second beginner getting discouraged just as if he had gotten pummeled by a sandbagger. If your goal is to eliminate such uneven matchups, a beginner's league will not accomplish it. It will theoretically reduce the # of such uneven matchups by eliminating the sandbaggers if you can come up with a metric that's enforceable.

The best short-term solution is to get them to realize that the enjoyment of the game and process of improvement are more important than winning and ratings. If they strongly disagree, maybe they ought to find a different endeavor.

I saw sandbagging in 2-day volleyball tournaments where a team would bring their B squad on Saturday, do mediocrely [is that a word?], and then show up with their A squad on Sunday and crush everyone. But this was the rare exception, not the rule. The unwritten code of conduct was that you just didn't do that and you'd be shunned if you did. It happened to my team a few times: that's life. We played, had fun, got our clocks cleaned, and then went out to dinner.

One year, my team reverse-sandbagged, in that we were supposed to come in 3rd in our pool of 4 at Nationals and we ended up winning, thus getting placed in the Gold division and getting creamed. We all agreed it was a thrill to play up.

What is your point?

My goal is not to eliminate uneven matchups. It's to eliminate sandbagging.

You're writing thesis after thesis on non-existent points.

 

[Startzel]

What is your point?

My goal is not to eliminate uneven matchups. It's to eliminate sandbagging.

You're writing thesis after thesis on non-existent points.

That's how all of his arguments turn out. It's incredibly frustrating.

 

[S&V-not_dead_yet]

What is your point?

My goal is not to eliminate uneven matchups. It's to eliminate sandbagging.

You're writing thesis after thesis on non-existent points.

OK; I mis-interpreted what you said.

"Beginners leagues need to really be strictly enforced for beginner-only."

How would this be enforced?

 

[beltsman]

OK; I mis-interpreted what you said.

"Beginners leagues need to really be strictly enforced for beginner-only."

How would this be enforced?

Don't know.

This thread is about sandbagging. Not eliminating bad losses.

 

[Moveforwardalways]

As an aside question in this thread, how does a sandbagger collect strikes for disqualification? The USTA website says this...

"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."

So you can win each match 6-0 6-0 and still be legitimately rated appropriately. So then how bad do you have to win to generate a strike? 7-0 7-0? 6 - (-1)? I don't get it. Seems like a "strike" would be impossible to get in that scenario.

 

[GlennK]

Sure, a 6-0 6-0 score can occur within a ranking without a strike, but if a self-rate defeats a highly ranked (say a 3.5 self rate versus a 3.4 computer ranked) by a score of 6-0 6-0, that would definitely be a strike. Heck, in that case, 6-3 6-3 might even be a strike.

 

[Moveforwardalways]

Sure, a 6-0 6-0 score can occur within a ranking without a strike, but if a self-rate defeats a highly ranked (say a 3.5 self rate versus a 3.4 computer ranked) by a score of 6-0 6-0, that would definitely be a strike. Heck, in that case, 6-3 6-3 might even be a strike.

So is a 6-0 6-0 win equivalent to 0.48 ranking points then? So then in your scenario, the self rate who won would be considered to have a rating of 3.4+0.48 = 3.88? I don't have enough insider USTA knowledge to know these things.

 

[Clintspin]

Woman will frequently drag their husbands into USTA league play and encourage them to start at the 2.5 level because that is what they did. However, their husbands who have played even less them they have, beat them with ease. I hear this from the ladies I coach all the time. Some of these guys should have rated as 3.5s and some as 3.0s but I have yet to see one that was a 2.5. League coordinators also encourage this because they can't come up with enough men to form 2.5 leagues.

 

[GlennK]

So is a 6-0 6-0 win equivalent to 0.48 ranking points then? So then in your scenario, the self rate who won would be considered to have a rating of 3.4+0.48 = 3.88? I don't have enough insider USTA knowledge to know these things.

I am not an expert and rarely pay attention to NTRP. I just wanted you to realize that the algorithm uses the rankings at the time of playing to determine the expected result. So if both players are 3.30, then expect a 1 game difference. Schmke and others have a better understanding of what the expected outcome in games is based on current ratings differences. And when strikes are generated.

Glenn

 

[MathGeek]

I just don't understand how it can be an honest mistake. These players have to have played other guys with usta ratings before joining a league, right?

Sure, but unless you ask a lot of questions and double check all your opponents' computer records, you can be honestly misled.

Suppose I play a few local guys who all tell me that they are 3.5s, and I take their word for it. But they are really sandbagging 4.0s or high level 3.5s. If I'm a medium or low 3.5 in real ability, I may well get beaten soundly enough by these guys to think I'm a 3.0.

There is also the issue of small sample sizes, good and bad days, and good and bad match-ups. Pushers overperform against one kind of player and underperform against others. So do junk ballers. All these issues produce much more variation in 3.0 and 3.5 than in 4.0 and higher where greater consistency provides more predictable outcomes based on ratings with small sample sizes. Unless a 3.0 or 3.5 plays 10 matches against players with known and verified computer ratings of the same sex, he may well have an inaccurate self-perception of his abilities.

At 3.0 and 3.5, at least 61% and 71% of men self-rate accurately. Since some bumps may result from improvement over the season, a bump may not indicate an inaccurate self-rate, so these percentages are really a lower bound. That's probably not too bad when assessing one's ability in a given sport. I know it's a lot more accurate than many amateurs are likely to do in minigolf, bowling, fencing, shooting, angling, and mountain biking. In most of those sports, it is extremely common to improve the equivalent of one (or more) ranking step over the course of a season, especially at the lower levels.

I also wouldn't draw too many conclusions about sex differences and sandbagging. How have alternate explanations been excluded, such as the hypothesis that many men are bumped because of improvement over the course of the season or that women are not bumped because they see it might be coming and throw a few matches late in the year to avoid it?

 

[Deleted member 23235]

There need to be serious penalties to captains that allow cheating.

The answer is for players to better enforce themselves.

i understand where you're coming from... but having been a player that's played in multiple regions (ie. was new to the region and didn't know anyone), I'll "cheat" every time... (ie. self rate lower just to get on a team, build my tennis network, etc...).

It's *hard* establishing a group (at least it is/was for me). everyone thinks you suck, until you prove otherwise. So there has always been a period of starting at the bottom, and working my way up to the proper level.

A common pattern of "proving myself" seems to be challenging the local 3.5 or 4.0 pusher, beat them, to be considered someone worthy of playing with... at the 4.0+ level.

Honestly I doubt I'd be able to get on a 4.5 team my first season if i was new to my area, unless i got lucky. I'd have to join one of the local clubs that has ladders (in the winter), where higher level folks would be forced to play me.

There was one guy recently that moved to the area... a few seasons back. he didn't know NTRP, and only played (high level) junior tourneys in his youth. this guy is a solid 5.0 player, beating and playing competitively with former ranked div1 guys. not knowing anyone, he just wanted to play. he was recruited by a local 4.0 team, where they managed his scores... playing singles in big matches, and hiding in doubles the rest of the time... the next season he played 4.5, and was managed similarly. finally he's playing at the 5.0 level, 3 seasons later. while the captains could arguably be held culpable, as the 5.0 player, i would have proably done the same, just excited to get on a team and start playing, making friends, making tennis connections, etc... (especially if i didn't have time to network myself).

 

[J_R_B]

This is the time of year when the complaints about the self-rated players at Nationals come out. So I thought I'd do some analysis on whether self-rated players generally do a good job, or if they tend to over/under rate on average.

See http://computerratings.blogspot.com/2016/10/do-usta-league-self-rated-players-self.html for the full blog, but it seems that men are more likely to under-rate or sandbag resulting in more self-rated players being bumped up than down, particularly at lower levels, while women are more likely to over-rate and have more self-rated players get bumped down.

This, as always, is interesting analysis, but for the issue of sandbaggers intentionally cheating to get an advantage at the national level, we're talking about such a small minority that they get lost in the margin of error of aggregate studies (unless you use the Startzel version of "cheating" where anyone who can win is a cheater).

 

[schmke]

This, as always, is interesting analysis, but for the issue of sandbaggers intentionally cheating to get an advantage at the national level, we're talking about such a small minority that they get lost in the margin of error of aggregate studies (unless you use the Startzel version of "cheating" where anyone who can win is a cheater).

I agree the true cheaters are a small minority. It is still interesting to see the difference by level and gender in the accuracy of self-rating and the bump up/down rates. There are certainly trends that go beyond what you would expect to see if it was only a small minority that were under self-rating for the men.

 

[Startzel]

i understand where you're coming from... but having been a player that's played in multiple regions (ie. was new to the region and didn't know anyone), I'll "cheat" every time... (ie. self rate lower just to get on a team, build my tennis network, etc...).

It's *hard* establishing a group (at least it is/was for me). everyone thinks you suck, until you prove otherwise. So there has always been a period of starting at the bottom, and working my way up to the proper level.

A common pattern of "proving myself" seems to be challenging the local 3.5 or 4.0 pusher, beat them, to be considered someone worthy of playing with... at the 4.0+ level.

Honestly I doubt I'd be able to get on a 4.5 team my first season if i was new to my area, unless i got lucky. I'd have to join one of the local clubs that has ladders (in the winter), where higher level folks would be forced to play me.

There was one guy recently that moved to the area... a few seasons back. he didn't know NTRP, and only played (high level) junior tourneys in his youth. this guy is a solid 5.0 player, beating and playing competitively with former ranked div1 guys. now knowing anyone, he just wanted to play. he was recruited by a local 4.0 team, where they managed his scores... playing singles in big matches, and hiding in doubles the rest of the time... the next season he played 4.5, and was managed similarly. finally he's playing at the 5.0 level, 3 seasons later. while the captains could arguably be held culpable, as the 5.0 player, i would have proably done the same, just excited to get on a team and start playing, making friends, making tennis connections, etc... (especially if i didn't have time to network myself).

I understand what you're trying to say, but at the end of the day how is this fair to the other players in the league that are playing by the rules?

That seems to be the big disconnect in these dicusssions.

Why is it fair for you to cheat just so you can play tennis?

 

[Deleted member 23235]

I understand what you're trying to say, but at the end of the day how is this fair to the other players in the league that are playing by the rules?

That seems to be the big disconnect in these dicusssions.

Why is it fair for you to cheat just so you can play tennis?

because the option is not to play at all (ie. i accidently self rate too high, and no coaches will take me on their team and play me).

"don't hate the playa, hate the game"... (ie. the ntrp rating system)

i have one guy on my team that self rated as a 5.0.... he was terrible.
he appealed down to 4.5, just to play on our team... and he was still terrible. we'd only use him when there was absolutely no else to play, and no one ever wanted to play with him because it was an automatic loss. if we fill a full squad, he won't play at all. if i were him, i'd have self rated at 3.5 or 4.0, and worked my way up to getting bumped to the appropriate level.
if you look at leagues as pseudo ladder, it's not cheating at all, just a natural way for folks to settle on the level they are intended to be at (working from the bottom up).

sure, folks who've been in the league for years, who end up getting clobbered by guys "working their way up", get annoyed because they can't win. but IMO, i should be grateful that I have a never ending stream of 5.0's that can beat me up (I learn from them). in the end, I am not playing league to win some stupid rec trophy (or "go to nationals"). I like the competition (win or lose... and even when I lose, I "win" - eg play someone better than me), not the shiny medal/hardware for winning.

if anything, the process for getting rated should change.
visual ratings doesn't work/not accurate.
maybe new folks would have play ntrp tournaments 2-3 times (or enough times to be able to to play 2-3 folks with computer ratings), to determine their correct rating.
or a flex ladder thing where matches count toward ratings
or tighter computer scrutinization on new players entering the system.
etc...

 

[Nacho]

This is the time of year when the complaints about the self-rated players at Nationals come out. So I thought I'd do some analysis on whether self-rated players generally do a good job, or if they tend to over/under rate on average.

See http://computerratings.blogspot.com/2016/10/do-usta-league-self-rated-players-self.html for the full blog, but it seems that men are more likely to under-rate or sandbag resulting in more self-rated players being bumped up than down, particularly at lower levels, while women are more likely to over-rate and have more self-rated players get bumped down.

My guess would be that women self rate higher because there are more recreational women playing then men, and thus more women in the beginning stages of understanding the game and ratings. This could also be caused by more womens teams looking for and recruiting players. For Men, many have played longer and understand where they are at better relative to the ratings. It is also more likely that former pro, teaching pro, and college playing men are in the playing fold then women. Again, my guess on the disparity

 

[Startzel]

because the option is not to play at all (ie. i accidently self rate too high, and no coaches will take me on their team and play me).

"don't hate the playa, hate the game"... (ie. the ntrp rating system)

i have one guy on my team that self rated as a 5.0.... he was terrible.
he appealed down to 4.5, just to play on our team... and he was still terrible. we'd only use him when there was absolutely no else to play, and no one ever wanted to play with him because it was an automatic loss. if we fill a full squad, he won't play at all. if i were him, i'd have self rated at 3.5 or 4.0, and worked my way up to getting bumped to the appropriate level.
if you look at leagues as pseudo ladder, it's not cheating at all, just a natural way for folks to settle on the level they are intended to be at (working from the bottom up).

sure, folks who've been in the league for years, who end up getting clobbered by guys "working their way up", get annoyed because they can't win. but IMO, i should be grateful that I have a never ending stream of 5.0's that can beat me up (I learn from them). in the end, I am not playing league to win some stupid rec trophy (or "go to nationals"). I like the competition (win or lose... and even when I lose, I "win" - eg play someone better than me), not the shiny medal/hardware for winning.

if anything, the process for getting rated should change.
visual ratings doesn't work/not accurate.
maybe new folks would have play ntrp tournaments 2-3 times (or enough times to be able to to play 2-3 folks with computer ratings), to determine their correct rating.
or a flex ladder thing where matches count toward ratings
or tighter computer scrutinization on new players entering the system.
etc...

Should this attitude also apply for people that are being bumped up?

I'm going to get bumped this year. So I will be playing at a new level and should be one of the weaker players. So it's unlikely that good teams will want me nor play me if they do.

Should I just throw my matches so I can get bumped down next year to a level I will have ample playing time?