Rating floors for RHP

Originally posted by gezza
The trouble is that it is now a continually moving goalpost. You won't know what level a "master" is - it moves year on year. Someone who clearly is that good (i.e. reached the level before inflation), could find themselves surrounded by weaker players who have the same "title", and who in essence just devalue it.

1) It's not going to move that much in just one year.
2) If you are worried about local rating inflation [selective inflation of just a few players] instead of general inflation [rise of the whole pool], you have greater threats than rating floors to worry about. Ratings manipulation is possible. There was a former USCF board member who had to resign because he pushed his rating to 2300 by only playing a select group of players.

Originally posted by gezza
"Proves" is hardly the correct word. But a 200 point improvement in rating *is* an indication of some progress.

Rating floors may *seem* a reasonable solution, but there is a disadvantage.

How do you know that the overall effect would be trivial?

Because the number of players who would be effected by the rating floor would be small. The number of players who mass resign games is small. Given that there are thousands of active players playing tens of thousands of games at any one time, the effect would be trivial.

There is no "disadvantage"; your complaint is petty and unreasonable and absolutely refuses to address the main issues. The idea that someone who mass resigns games has suffered a massive diminution in their chess playing ability is weird; the idea that a system shouldn't take into account that they haven't suffered a true drop in playing ability is bizarre.

Originally posted by gezza
[b]Shrug. It is not necessary to know the exact formula to realize that sandbagging and vastly underrated players are a serious problem. It is also easy to see that losing 1000 points due to mass resignation has zero correlation to chess skill.

If you don't know whether the system in use is Glicko, Glicko II, or ELO, then you have little basis on which of keeping the score. I am not keen for them to be broken or devalued.[/b]

If you don't know whether the system in use is Glicko, Glicko II, or ELO, then you have little basis on which to comment.

Guess what I found? FAQ Yes, the new system is not much different than the traditional one. [All with K=32] You still get 16 points for beating someone of the same rating. In the old system, you got 32 points for beating someone 400 points higher rated, and lost 32 points for losing to someone 400 points lower. Under the new one, you get 29 points for beating someone 32 points higher, and lose 29 points for losing to someone 400 points lower. That's a grand total of THREE points difference in each case. This conclusively dismisses this non-point you keep bringing up.

Is too. So I say the clan leader decides, and takes the flack either way.

Is too?! Nice argument. It's patently obvious that you just don't give a crap about clan challenges.

"ruin" open invites? How?

Because when your formerly 2000, now 1000 player comes back, after mass-resigning all previous games, his win expectancy is now three-tenths of a percent against a 2000 player. We both know this is obviously false. I would love to bet $$ on that match, but I'd feel too guilty about taking your money so easily.

Players against whom he plays get a game against a strong player, helping them improve, just like the chap rated 1200 who has currently got an open invite open for >2000 points.

And that chap isn't going to get to play the 2000-strength player if we let him drop to 1000, now is he? 🙄

Did too - you repeat it as a problem.

No, I accused you of focusing only on banded tourneys when the problem is larger in scope.

Eh? Unless that player is new to the site, then the Tournament Entry Rating says which tourney band he gets to play in.

And under the 'solution' you propose, the mass-resigner will not be able to play in ANY banded tourneys until he gets his rating almost all the way back up! Tough luck for paying customers who want to enter more tourneys. But I guess that's a small sacrifice to make so that you can keep your personal pride intact.

So why should it stop? You still lose a game.

You love to ask the same questions over and over again, no matter how many times they get answered.

So resignations should be a special case then?? Not like a standard loss? Laughs.

Mass-resignations, not resignations. You are either doing a piss-poor job of reading what I write, or you're too sloppy a writer to even summarize a position properly, or being deliberately deceitful.

Given that he last moved a few minutes ago, he may wish to comment himself. Maybe it is a co-incidence that he moved when you mentioned his name. We will indeed see. But note that as his rating was falling, he also won a few blocks of games in a row. I contend that he is just as likely to stop playing again, as he has done in the past.

I don't want RHP ratings to measure, say, how many finals someone has to take in college. The number of mass-resignations/timeouts is irrelevant to chess ability.

Oh. Your theoretical, has not happened, example.

Actually, your theoretical example, that I tweaked to show that you can indeed win games without merit on your part.

And so what if it is theoretical? Do you have some kind of strange issue with using hypotheticals in an argument? Do I really need to cull through your games and dig up some blunder-filled abomination that you nevertheless won?

I claim timeouts if I feel like it, but rarely in an interesting game.

If you feel like it?! By your earlier logic, you are deliberately skewing your own rating by gaining points through a means other than pure chess skill. What blatant hypocrisy!

So why do you insist that mass resignation because you cannot cope with the current game load for whatever reason should not affect your rating if you hit some "floor".

Because mass resignations may be deliberate for the sake of sandbagging. Since RHP's rating calculator cannot go and ask the player if he is trying to sandbag or not, or run an investigation on him to determine the purity of his motives, I kinda think we should throw a floor in to prevent him from doing so. 🙄

I did not state the order.

A sign of a bad writer.

What about that example of "spiteful resigning"? Is that just another theoretical possibility to attempt to strengthen your arguments.

Yes, you do indeed have some strange hangup with hypotheticals. Maybe I should quit using them and come down more to your level, with something like "Is too!" or "I know you are, but what am I?" 😛

Past performance is used to give an indication of future performance.

Key difference: You want chess ratings to specifically measure all life vs. chess prioritization, and I want to minimize the impact of life-driven hiatuses on the system.

This is sure what it sounds like - you want rating points if your opponent resigns, even if he resigned 200 games before resigning yours.

Yes, if he is of the proper rating class - but in the interests of accuracy, not pride.

The only thing you care about is getting more rating points. Rating points which the "fix" you want to add to RHP will devalue - but it does not matter - you get more points and more bragging rights.

Yes, you've uncovered my grand conspiracy! First, rating floors; tomorrow, I will be rated 3200 before any of you fools know what hit you! 🙄🙄🙄

I want to know if I progress from one year to the next - I care about how I progress.

Then you'd better check the rating history of everyone you play. If they used to be 2200, and are now 1000, they are going to rape you for 32 easy points.

Originally posted by SwissGambit

...The only thing you care about is getting more rating points. Rating points which the "fix" you want to add to RHP will devalue - but it does not matter - you get more points and more bragging rights.

Yes, you've uncovered my grand conspiracy! First, rating floors; tomorrow, I will be rated 3200 before any of you fools know what hit you! 🙄🙄🙄...[/b]

Busted!

He's on to us, meet me back at the ratings hideout!

P-

Originally posted by SwissGambit
snip ...

Guess what I found? FAQ Yes, the new system is not much different than the traditional one. [All with K=32] You still get 16 points for beating someone of the same rating. In the old system, you got 32 points for beating someone 400 points higher rated, and lost 32 points for losing to someone 400 points lower. Under the new one, you get 29 points for beating someone 32 points higher, and lose 29 points for losing to someone 400 points lower. That's a grand total of THREE points difference in each case. This conclusively dismisses this non-point you keep bringing up.

Dismisses? You keep getting points for a greater difference in ratings. How do yo dismiss that? Congratulations on finding the FAQ. Perhaps now you will have an idea what you are talking about. I look forward to more informed arguments...


Is too?! Nice argument. It's patently obvious that you just don't give a crap about clan challenges.

Your argument reminds me of pantomime. I respond in character. I have already said that I am not qualified to comment on clan challenges, beyond hearsay, as I am not in any clan. What more do want? The the pope to comment on contraception?

Because when your formerly 2000, now 1000 player comes back, after mass-resigning all previous games, his win expectancy is now three-tenths of a percent against a 2000 player. We both know this is obviously false. I would love to bet $$ on that match, but I'd feel too guilty about taking your money so easily. And that chap isn't going to get to play the 2000-strength player if we let him drop to 1000, now is he? 🙄

Someone wants a game against a strong player. If a strong player, currently rated 1000 for whatever reason, were to challenge the lower rated player who has an open invite for >2000, would it be accepted? If not, then the lower rated guy is clearly putting far too much emphasis on rating. If yes, he may learn something, even if the stronger, but lower rated guy needs to leave the site again for whatever reason.

Me:
Did too - you repeat it as a problem.

You:
No, I accused you of focusing only on banded tourneys when the problem is larger in scope.

I think you said, "it is not that they leave, it is that they come back and unfairly win clan challenges and banded tournaments. " Therefore I re-affirm that you repeat a problem which I have already responded to. You do not manage to come up with a reason why my response is unacceptable, you just re-iterate your vision of a problem.

And under the 'solution' you propose, the mass-resigner will not be able to play in ANY banded tourneys until he gets his rating almost all the way back up! Tough luck for paying customers who want to enter more tourneys. But I guess that's a small sacrifice to make so that you can keep your personal pride intact.

Yawn. Which part of "Tournament Entry Rating" do you have trouble with? As it stands, the tournament entry rating says which section of a banded tournament you can enter. Paying members can still enter banded tourneys at their TER. If you do not understand the system, you have little right to criticise. Perhaps you should look at the FAQ a little harder. Failing that, try Russ' and Chris' posts, especially those from March 2007.


You love to ask the same questions over and over again, no matter how many times they get answered.

Answer? Clearly I missed it, again. How unobservant of me. Perhaps it was not clear what the advantage over the current system was.

Mass-resignations, not resignations. You are either doing a piss-poor job of reading what I write, or you're too sloppy a writer to even summarize a position properly, or being deliberately deceitful.

So when you are losing an argument you accuse me of deceit? Go away and do your homework.

I don't want RHP ratings to measure, say, how many finals someone has to take in college. The number of mass-resignations/timeouts is irrelevant to chess ability.

You argue that time management is part of chess, yet inability to see that you will not have time to play in the future due to college finals is somehow not part of time management. Start wriggling. No - carry on wriggling, it is funny.

Actually, your theoretical example, that I tweaked to show that you can indeed win games without merit on your part.

And so what if it is theoretical? Do you have some kind of strange issue with using hypotheticals in an argument? Do I really need to cull through your games and dig up some blunder-filled abomination that you nevertheless won?

Oh, I am sure there is one example of a game where neither I nor my opponent has played well. Let's make it interesting: Find 5. And DO NOT COMMENT ON GAMES IN PROGRESS!! Sorry to shout, but you have failed to understand some basic points in the past.

If you feel like it?! By your earlier logic, you are deliberately skewing your own rating by gaining points through a means other than pure chess skill. What blatant hypocrisy!

What is hypocritical about stating in ones profile that one will claim timeouts on a whim, and then doing so? It seems you have trouble understanding more than just ratings.

Because mass resignations may be deliberate for the sake of sandbagging.

But it "may not" be deliberate. Why do you continue to think so ill of people?

A sign of a bad writer.

My unreserved apologies. I was writing under the assumption that those reading were willing and able to understand implication. If you cannot, it is hardly worth continuing. Did you manage to understand this simpler version I wrote afterwards:", because your opponent's rating drops correctly because he resigns a whole bunch of games before resigning against you"? How simple do I need to write?


Yes, you do indeed have some strange hangup with hypotheticals. Maybe I should quit using them and come down more to your level, with something like "Is too!" or "I know you are, but what am I?" 😛

Or perhaps you could give the example you seem to be focusing on. I state when I use extreme examples. You appear to have trouble with them. So is your example hypothetical, or is it real?

Key difference: You want chess ratings to specifically measure all life vs. chess prioritization, and I want to minimize the impact of life-driven hiatuses on the system.

In the absence of a better way of indicating likelihood of losing games, rating will have to do. What alternative have you proposed? What you have suggested to date is that players who can be strong if they concentrate have a rating which corresponds to their ability if they concentrate all the time. But they don't, so I think their rating should reflect how they actually play.

Yes, if he is of the proper rating class - but in the interests of accuracy, not pride.

Clearly accuracy is dear to you, as long as the accuracy rounds upwards.

Yes, you've uncovered my grand conspiracy! First, rating floors; tomorrow, I will be rated 3200 before any of you fools know what hit you! 🙄🙄🙄

It has been done before. Having seen a player rated 3000, I will no longer be amazed at how desperate people are to increase their chess rating.

Then you'd better check the rating history of everyone you play. If they used to be 2200, and are now 1000, they are going to rape you for 32 easy points.

You jest. No-one gets easy points. Even 2200 rated players have to give me a game to win rating points. If I consent to play someone who has been higher rated in the past, it is hardly "rape". If you have understood anything I have written, you will understand that I am more than willing to play against stronger players. I may have reservations about playing someone who leaves the site at their convenience, but that is for me to discuss with such a player at an appropriate time.

Originally posted by gezza
[b]Guess what I found? FAQ Yes, the new system is not much different than the traditional one. [All with K=32] You still get 16 points for beating someone of the same rating. In the old system, you got 32 points for beating someone 400 points higher rated, and lost 32 points for losing to someone 400 points lower. Under the new one, you get 29 point t is for me to discuss with such a player at an appropriate time.[/b]

Dismisses?

Yep. I already informed you that I knew the exact formula for the traditional ELO system. The one for RHP leads to roughly the same results. [Just as I knew it would, since I have actually played more rated games here than you have, and the rating results have conformed to my expectations.]

I have already said that I am not qualified to comment on clan challenges, beyond hearsay, as I am not in any clan.

Riiiight...you're just qualified enough to tell a clan leader that sandbagging is HIS problem, and not RHP's.

If a strong player, currently rated 1000 for whatever reason, were to challenge the lower rated player who has an open invite for >2000, would it be accepted?

Why should they have to do the extra legwork? Mr. 2000 should stay up near his proper rating class.

Notice how doing things your way causes people to second-guess the rating system just to get a single open invite game started. Now imagine extending that headache to organizing a big clan match [Oops, sorry, nevermind; you're not qualified].

I think you said, "it is not that they leave, it is that they come back and unfairly win clan challenges and banded tournaments. "

You do not manage to come up with a reason why my response is unacceptable...

You did not manage to come up with ANY solution for the clan leader other than "deal with it". That's the glaring reason your response is unacceptable.

Which part of "Tournament Entry Rating" do you have trouble with?

Bravo. It's about time you actually scored in this debate. Yes, I got this wrong - I added the edit too hastily. What I meant to say was: 1000-rated players can now use a high TER floor to enter [say] 1700+ tournaments. Adding only a TER floor does not fully solve the banded tourney problem.

Perhaps it was not clear what the advantage over the current system was.

It has been clear to every other poster but you in this thread.

So when you are losing an argument you accuse me of deceit?

You think you're actually winning this?! LMAO. The one guy you actually had on your side earlier defected after a post or two pointing out the obvious flaws in your inflation argument.

You argue that time management is part of chess, yet inability to see that you will not have time to play in the future due to college finals is somehow not part of time management.

Sorry, I prefer to measure time management when someone is actually trying to play the game. I know you'd prefer to have as many non-chess factors as possible included in a chess rating, but you're thankfully in the minority.

Oh, I am sure there is one example of a game where neither I nor my opponent has played well. Let's make it interesting: Find 5.

I only need one: Game 2277051. This is exactly what I was talking about in my example!

And DO NOT COMMENT ON GAMES IN PROGRESS!! Sorry to shout, but you have failed to understand some basic points in the past.

If you actually comprehended the example, you'd know that it NEEDS A FINISHED GAME, by DEFINITION. Sorry to shout, but you're acting like a fekkin' moron.

What is hypocritical about stating in ones profile that one will claim timeouts on a whim, and then doing so? It seems you have trouble understanding more than just ratings.

According to you:
1) If your opponent times YOU out, it sullies the conditions required for rating to reflect ability.
2) The integrity of the rating formula is preserved by strictly following the mathematical formula at all times - no exceptions!
3) You allow yourself to time out your OPPONENT at whim. So much for the conditions required for rating to reflect ability.

Get it now? 🙄

My unreserved apologies. I was writing under the assumption that those reading were willing and able to understand implication.

Implication has nothing to do with it. If you write a sequence of sentences about an event, or events, it is normal for the reader to think that you are placing events in chronological order. If you want to suddenly shift back into the past, you need to give a clear indication that you are doing so.

How simple do I need to write?

Just make your writing more clear. That will do nicely. And stop blaming your readership for failing to decode your murky prose.

Or perhaps you could give the example you seem to be focusing on. I state when I use extreme examples. You appear to have trouble with them. So is your example hypothetical, or is it real?

Great. You're just trying to match criticism with criticism, without regard to whether you make an ounce of sense or not. You need to follow your own advice. Post less. You're not as thick-skinned as you though you were.

In the absence of a better way of indicating likelihood of losing games, rating will have to do. What alternative have you proposed? What you have suggested to date is that players who can be strong if they concentrate have a rating which corresponds to their ability if they concentrate all the time. But they don't, so I think their rating should reflect how they actually play.

What does any of this have to do with a hiatus from chess play? Please, look up the word "hiatus".

It has been done before. Having seen a player rated 3000, I will no longer be amazed at how desperate people are to increase their chess rating.

Right - on ICC and PlayChess, there are no rating floors, and yet people are rated over 3000 in bullet and blitz. Sounds like I was right earlier when I told you that there were other more tangible causes of inflation besides rating floors.

You jest. No-one gets easy points. Even 2200 rated players have to give me a game to win rating points.

You might think you're giving them a game, but, to be blunt, I doubt they find you a very challenging opponent. That tends to come naturally with being 400 points stronger.

I may have reservations about playing someone who leaves the site at their convenience, but that is for me to discuss with such a player at an appropriate time.

Why? What's wrong? Surely, their new, lower rating actually represents a decreased ability, as you've been insisting all along. What happened to your complete faith in the rating system as-is?

Originally posted by SwissGambit

So when you are losing an argument you accuse me of deceit?

You think you're actually winning this?! LMAO. The one guy you actually had on your side earlier defected after a post or two pointing out the obvious flaws in your inflation argument.

Good. You are still here.

Actually, I don't think I am winning. I think I have won!! Here's why:

Originally posted by Phlabibit
With ratings floors, the average RHP player's ratings won't go up 100 points. 15 perhaps... but no way will everything go up 100 because a fat fist full of users tank 1 or 200 games.

Phlabibit accepted that there will be rating inflation.

Originally posted by SwissGambit
1) It's not going to move that much in just one year.

You accept there will be rating inflation.

Originally posted by gezza
Can someone explain to me how this will not lead to grade inflation?

My original comment was about: rating (or grade, whatever you want to call it) inflation.


So we go from a position of you both saying there will be no rating inflation to you both accepting that it will happen, and arguing about how much it will be.

In the meantime, you have also managed to understand what the tournament entry rating is about, and to find the bit of the FAQ which deals with rating.

You've also grasped the concept that a TER floor could be helpful in solving the issues with players sandbagging to enter banded tournaments at a lower band than they should.

So, to summarise: We all agree that a rating floor would cause rating point inflation. We agree that the banded tournament problem could be solved by a floor on the TER. Where we disagree is that if someone drops their grade by resigning, you think a rating floor should be used to stop it falling (too far), whereas I think that although the rating system is not perfect, the inflation caused by a rating floor breaks it more than the problems in the current system.

Is the above an accurate reflection on where we have got to? If not, please indicate what I have not captured correctly.

Originally posted by gezza
Good. You are still here.

Actually, I don't think I am winning. I think I have won!! Here's why:

Originally posted by Phlabibit
[b]With ratings floors, the average RHP player's ratings won't go up 100 points. 15 perhaps... but no way will everything go up 100 because a fat fist full of users tank 1 or 200 games.

Phlabibit accepted tha ...[text shortened]... tion on where we have got to? If not, please indicate what I have not captured correctly.[/b]

If there is any inflation (speculation) it won't really be noticed. I threw a figure of 15 or so out, I thought you said hundreds someplace and that's crazy.

Also, IF there is any inflation it won't really make a difference. A rating is a rating is a rating and can only be compared to other ratings in the same system. You seem to think any inflation (if it happens at all) is bad.

So just because I admit there may be inflation, it doesn't really weigh in as a reason to NOT have ratings floors. As I've already said, ratings will seek there true level for each player. If I'm rated 1475... and everyone's ratings go up 10 or 20 we are still able to compare our ratings and know what to expect.

Where you've lost is thinking a 2000 player should be rated 800 because they tanked hundreds of games. 800 doesn't reflect their skill.

P-

I would like my personal ratings floor to be 2700.

Originally posted by gezza
Good. You are still here.

Actually, I don't think I am winning. I think I have won!! Here's why:

Originally posted by Phlabibit
[b]With ratings floors, the average RHP player's ratings won't go up 100 points. 15 perhaps... but no way will everything go up 100 because a fat fist full of users tank 1 or 200 games.

Phlabibit accepted tha tion on where we have got to? If not, please indicate what I have not captured correctly.[/b]

So we go from a position of you both saying there will be no rating inflation to you both accepting that it will happen, and arguing about how much it will be.

This is wrong. I have never denied that floors would cause some small additional rating inflation. I just don't think it's enough to worry about.

We agree that the banded tournament problem could be solved by a floor on the TER.

No, I don't agree with this. A TER floor still permits an 1850 rated player to drop to 1000, then enter a banded tournament for 1600+ players. There should not be 1000-rated players in a 1600+ tournament.

Originally posted by SwissGambit

No, I don't agree with this. A TER floor still permits an 1850 rated player to drop to 1000, then enter a banded tournament for 1600+ players. There should not be 1000-rated players in a 1600+ tournament.

How annoying. I post my enitre post in the quote section and then it gets shortened. oops.

I don't have the energy to type it all again, but to summarise my thoughts -

The problem is not so black and white.

At the top- and bottom- ends you may be correct, but mid-rating positions (like mine) can fluctuate easily by 100s of points in no time.

Any solution would have to prevent deliberate rating abuse, whilst not affecting legitimately fluctuating users of the site.

Most solutions that I see tabled do not do this, they overarchingly affect everyone, which is unfair because the majority of tourny entrants are not deliberate sandbaggers or similar.

If you have actual real fluctuation, it's not right that a rule to stop some deliberate abuse should prevent you from entering your current band.

Originally posted by mrmist
How annoying. I post my enitre post in the quote section and then it gets shortened. oops.

I don't have the energy to type it all again, but to summarise my thoughts -

The problem is not so black and white.

At the top- and bottom- ends you may be correct, but mid-rating positions (like mine) can fluctuate easily by 100s of points in no time.

Any that a rule to stop some deliberate abuse should prevent you from entering your current band.

Under rating floors, you could still fluctuate quite a bit. [Up to 200 points or so.]

It does not matter if sandbagging is deliberate or not. Imagine if you entered an U1200 banded tourney at that one point on your graph when you were 1168. Can't you see why other career 1100 players might not want you to play in the tourney?

Originally posted by SwissGambit
This is wrong. I have never denied that floors would cause some small additional rating inflation.

I see a slight discrepancy here.

To quote the posts in order:

Originally posted by gezza
Overall, more points get added to the system - giving inflation.

Originally posted by Humph3
The point about rating inflation is valid, so why don't we just put a floor on the TER?

Originally posted by SwissGambit
And I cannot concede that the point about rating inflation is valid.


Looks like someone has a poor memory. Actually, I don't care whether you deny denying that rating floors would cause inflation. You now accept that they would. Time to move on to the next issue in the debate.

Summarising:
We all agree that there will be rating inflation.

Originally posted by gezza
I see a slight discrepancy here.

To quote the posts in order:

Originally posted by gezza
[b]Overall, more points get added to the system - giving inflation.

Originally posted by Humph3
The point about rating inflation is valid, so why don't we just put a floor on the TER?

Originally posted by SwissGambit
And I e next issue in the debate.

Summarising:
We all agree that there will be rating inflation.

[/b]SO WHAT?

A rating is to compare yourself to another user. If I'm rated 1100 or 1900 I know where I stand and where everyone else does.

What you don't want is a 800 rated player who is really more than a thousand points stronger.

Tell me, please.... what is the problem with 'inflation' when all we compare these numbers is to ourselves? You can't compare numbers from most systems to another.

P-

Originally posted by gezza
I see a slight discrepancy here.

To quote the posts in order:

Originally posted by gezza
[b]Overall, more points get added to the system - giving inflation.

Originally posted by Humph3
The point about rating inflation is valid, so why don't we just put a floor on the TER?

Originally posted by SwissGambit
And I ...[text shortened]... ext issue in the debate.

Summarising:
We all agree that there will be rating inflation.[/b]

Looks like someone has a poor memory.

That would be you. You have also thrown in your usual measure of [willful?] obtuseness.

One post above humph's post, I said:

I think [rating-floor driven inflation] is by far the lesser evil. Players won't be able to maintain an inflated rating for long if their play is not up to it.

I have consistently admitted that some small amount of inflation will occur with rating floors.

The 'point about rating inflation' was that rating floors would cause too much inflation. Of course I would deny this. No, it is not the same as denying rating floors cause inflation. You're just making yourself look more foolish by willfully refusing to read things in context.