grenv: I believe that that is in the works also

Vikings: You're right, they both will be here. I still would like to fix the ratings so they degrade over time without play.

Even something simple like subtracting 5% for each month without a game of that type finished or something.

grenv: yes but a true ladder can only be played one game at a time, so there is no advantage in playing 1000's of games. Ladder rating is just that, BKR is just that, I believe that there will be both

Vikings: Yes but I play 20-30 games at a time and not all of the same type. This is because I have other things to do. Some people seem to play 24 hours a day and have 1000 games going. These people have an advantage.

I would prefer that ratings degrade with little or no activity, that would solve the problem you correctly identified.

grenv: it stops people from getting a high rating and then sitting on it,and avoiding good players, how can you expect to go up if you don't play?

AbigailII: some people were creating multiple id's (pawns) for the purpose of causing trouble on the boards, and then editing or deleting them before they would get caught, by eliminating the edit button for pawns, this has decresed

Vikings: I have never seen a ladder system i liked. Too often rewards people who play a lot, rather than the best players.

ScarletRose: I have noticed an inverse relationship to the stridency of someone proclaiming they only play for fun and the level of their rating.

"The object of a game is to win, but the reason to play a game is to have fun."

It should go without saying that you play a game for fun. Why else?

The last thing I want to hear my opponent say is that he doesn't care who wins, and I feel sorry for someone elses game if their opponent says that to them.

This is why the people that care about their rating want the ratings to reflect some value that's easy to understand and is readily comparible to someone elses. This way when I look at your profile I can tell which games I might want to invite you to play me in. At least as a rough guide. As others have posted here if someone plays in a small pool of opponents and loses most of the games they might actually be a better player than is reflected by their rating because the people they have been losing to might be really good players. Is this the case for you?

I'm sure my Janus Chess rating is lower than people would expect from my play, but ol' Sumerian's Smirf beats me almost every game. Consequently I have a lower rating. Now that I've been playing a wider pool of players in Janus Chess, my rating has leveled to my ability. For awhile I was dragging along around 1400. I still don't have a high rating in the game, but this is probably because I'm not that strong of a player of it consistantly enough to get a higher rating. I still play the game win or lose because I find it a fun game, but I try to win every game and am disappointed when I lose. It doesn't mean that I didn't have fun playing the game, it just means I'm disappointed in losing. If my opponent makes a good move and it's something to remember, it makes the loss feel better than if I made a bad move and gave the game away, but I can still have fun even if I lose. I never play a game to lose. I find winning a game more rewarding then losing a game. This is what the rating is supposed to show, not if you are having fun playing.

Vikings: There is a reason that pawns can't edit, it keeps them from being able to delete trouble post

I don't get it. I'd say that being able to delete "trouble posts" is a good thing.

the solution is getting the ladder system in place

There is a reason that pawns can't edit, it keeps them from being able to delete trouble post before getting caught(getting rid of troublemakers)

nobleheart: Correct, even other player's ratings that have been modified by playing these players are now inaccurate.

If possible the users should be deleted and the ratings recalculated.

If that is too difficult then at least turn off the accounts and remove them from the rating list.

Hrqls: their would be a big benefit to penalizing or banning cheaters.
we could depend on the ratings then to be realistic.
also a stern policy on cheating would be a good deterant for such behavior.

playBunny: if pawns were allowed to edit thier own posts.then we pawns could edit our own mistakes.do not moda have enough to do?
it has always seemed ilogical not not allow each to edit their own mistakes.
the ability to edit text would emininate typos thus inproving communication.

grenv: does anyone see this?agree?
cheating to artificially create a high rating..is easily transparent to most players.
is it not a hollow victory?

FriendJosh: This whole topic has arisen because two sets of cheats have just been unearthed and there's little room for doubt. In one case a whole series of games were resigned very shortly after the start of the game in a game (hypergammon) where resignation is very rare. See the backgammon board for details (starting about 40 messages back with ArtfulDodger: "Wouldn't it be odd").

In the other case the guy had two nicks playing each other. One from his first names and one with his surname. And his profiles spelled it out. Doh! See this board (about 60 messages back).

Hrqls: Oh, you can still ban them. But we cant delete the account - this would mess up everybody else's account. So their artifically high rating would still show even if they were banned. Thats where a indication that this is a "cheater" comes in - "cheaters" would keep their account and ratings, but would be visible as cheaters and not show up in the overall ratings. This would mean that we would keep the regular rating system without any modifications (allowing everyone to play as they want now), but still "remove" those who cheat the system.

As for identifying cheaters - that's the rub. How can you conclusively identify if someone is cheating 100%? In essence, the final authority has to be Fencer (or another admin maybe). When Fencer says you are cheating, he can remove you or punish you.

friendjosh

Rex Nihilo & anyone else :) : i dont know much about ratings .. i just played games in which you gain experience points and skill levels (rpgs) ... which is seen as a rating in there, but which just depends on time :)

i have some feeling for maths though .. and i like any ratings and to work for them :)

would it be hard to implement other types of ratings as well ? and maybe display as a side note to the ratings which are used now ? (the lists dont even have to sort by them)

weehee .. another feature request :)

FriendJosh: the whole problem is to identify cheaters, when they are identified with a 100% certainty .. ban them .. why keep them ?

playBunny: you are right i can see a group wanting to have a rating among themselves .. i didnt think of that .. i think i can add nothing to what we have said before and have no real sollution yet there are too many sides to ratings online in a lose 'organisation' like this .. ratings are tough to get right when its done online .. there are too many holes in which villains can hide .. and its too little controlled ... but still they are fun to work for :)

redsales: Ah, the voice of reason.

ScarletRose: i don't begrudge you the right to play unrated games, why should you return the favor to us who do want to play rated games. You say you want to have fun? Well, my definition of fun generally doesn't involve playing someone 1000 points higher or lower than me in a particular game, it usually is too easy or too hard, and no fun. How do I know this? From the ratings.

playBunny: I have also never turned down a game request by anyone with the "rating window" where I don't have to lay 14-1 without good reason and offering a counter proposal (usually a longer time limit). I have also, on occasion, accepted a game from someone 400 points below me.

I do play for rating. It's part of the competition on here. I will say ELO ratings are not a good fit for BG, but that's the way things are on here. I know different systems are out there, but I don't know enough about them to reccomend one.

On the feature request side, it would be nice if we could wager Brains one here. Not sure about the legality of that in the Czech Republic, and everywhere else for that matter, but it is worth a thought or two. I know I've mentioned it before. I'm not agitating, just thought it would be a good thing to bring up given the context.

Fencer: I'd like to request that Pawns be given the ability to edit their posts.

I don't think it's a privilege to be able to correct and clarify your postings. FriendJosh, by posting a correction, has done us a service by making his message more meaningful. Would it not be good to make it tidier and easier for Pawns to do that?


Spirou: :-))) Sometimes I'm wise. Some would say I'm just a wiseass. Sometimes I'm just an ass. Hee hee hee.

ps... i meant "by a value" instead of "but a value" :-)

Fencer,

Alot of the discussion around here is how to correct ratings for cheaters, but what if you just identify who those cheaters are? That is, since the ratings in all the games (even the win/loss ratios) can be "cheated", why dont you identify the users who are "cheaters" but a value (boolean?). This value would display their name in green or something so that others could see they were known cheaters. This value could also be used to remove them from ratings lists or tournaments.

If you want to be fair, you could also have this value reset after X time or stay permanent (if the account was banned maybe).

friendjosh

ScarletRose: Your highest Hyper BKR was 1956.

playBunny: You may be right, we all be wrong. It is a question of generation. Please don't charge me to be nostalgic from my youth :-). BUT! I am absolutely not jealous. But! Rating is good as you explain though also bad due to the fact many (not all) are prone to cheat in order to protect this meaningless stuff.
Therefore I maintain: rating is a poison as is money.
The nickname of chess (and his variant I suppose) is "The noble game" because (I was teached to know) his (or was?) the pleasure to play is self-sufficient.
I dream, if all of us could think and play as you say, we were all intelligent and wise.
But as I can see actually, rating prone an horrible greediness. And greediness conducts to poor behaviour.

As you say we are all wrong, because we (me included of course!) are not wise. Could we strive toward a wise behaviour? "I have a dream..." said someone, follow my eyes ;-).

Signed an old drone :-)))

ScarletRose: You were too quick to fly away there, oh scarlet , and failed the comprehension test. I said that you were not jealous.

Strangely enough, I consider conversations on discussion boards to be part of the real world.

Lolol. Any more slurs that you'd like to cast? ;-p

AbigailII: Lolol. So he did. I got the 14 bit but didn't like the whole idea and forgot the bit about the year.

Hrqls: My sympathies. :-) I hate laptop keyboards too.

It depends what you want ratings to be. It's surely very unlikely but it's possible that a small group of players would want to play for ratings only between themselves. If they stick to playing within their own group then the ratings will be accurate and useful - for them. That seems a reasonable usage of the rating system as it adds value to their experience of this site.

Those ratings will appear within the game pool, of course, yet be unrelated to any outside their own group. That, from the point of view of the entire ratings/rankings list, is a bad thing (though many individual players may not care two hoots). I, being interested in my standing within the pool, would care. You too, it would seem. I'd prefer that ratings were as accurate as possible** and would wish that sub-groups couldn't affect the mainstream ratings pool. But then I'd also like the players who play only a few games, get a high rating and then stay at the top of the board by not playing, to also be "adjusted" for (or compelled to play, lol). It comes back to the idea in my earlier post about "choppy seas". There are many influences in the rating system that make it imperfect. Which areas, if any, should be addressed can only be decided as a matter of site policy.

** Accurate ratings would be better served in backgammon by using the correct backgammon formula ... ah, but that's a different feature request, lol.

George: Totally agree. ;-)
(ps. Make a move! ;-p)

chessmec: yeppers.. at one time I had like 2400 bkr for Hyper Gammon.. of course.. then I played the game a few times.. haha..

Not jealous as Bunny states.. just happy to play and make friends.. welp.. gotta run.. got things to do in the real world..

playBunny: A high rating demonstrates mastery

It depends on the amount of played games. When e.g. a player plays only a few games and gets more than 2400 at first ... it demonstrates NOTHING.

Spirou, ScarletRose: I don't know why but it never fails to impress me when someone's sole contribution is to say "I don't understand your viewpoint therefore it's rubbish". Now that's what I call childish. For me, and probably many high-rated players, the rating aspect is part of the fun. I enjoy playing, I enjoy chatting and I enjoy the attempt to get my rating high. A high rating demonstrates mastery and guess what, that's a nice thing to do if you are lucky enough to have the talent and have worked at it. For students of the game, striving to improve, the rating history provides a visualisation of their progress. You guys want to call it lame? Then I'll call you jealous - and we'll all be wrong.

ScarletRose: You have a good point, but I want to state that I _DO NOT_ cheat to achieve my backgammon rating.

That said, Spirou's comment applies to backgammon, as well. As "Big Jim" at the Flint BG Club says, "Ultimately, it all comes down to money."

(edit) -- And I want to invite you all to the Flint BC Club. We meet at the Days Inn on Bristol Rd. right accross from the GM Plant. (http://www.flintbg.com/). You can find me in the $2 chuoette.

ScarletRose: I applaud to your message.
This struggle is like a frog battle in little marshland. No offense to anyone but an invitation to see far away.

Why bother with a rating system anyway.. it isn't like you are getting world wide acknowledgement.. and besides.. there are peeps out their cheating everyday that mostly go unnoticed.. Just my opinion..

Come on peeps.. I come play online to have fun.. NOT to measure how much better I am.. geez.. so lame..

A just create a Janus game with 1 hour for each. Is somehow interested in? I will wait until 17 hour around.

El Cid: I know, I know, I am from a past generation when winning (sometime) against a weaker offer the same satisfaction than suffering a lost against a stronger player.

Oh folks I feel old ... LOL

Fencer: I like that idea, but I think a year is too long for the counter to reset. Maybe every 3-4 months.

Also, I would like to say that tournament games should not be added to the counter, but only non-tournament games should.

Spirou: It gives the possibility to enter in top50 tournaments (though in the top50 Battleboats Plus, i am playing, there are people that are on the 2xxth in rating position)

In order to summarize : rating is a poison!

In card games the "motor" is money, in chess in my young time the "motor" was the pleasure to play. Now it is this meaningless rating. Sad, sad, saaaad ... If at least rating points was money, I could probably understand this struggle. LOL

Aren't we all a bit childish? Kindly yours :-)))

playBunny: N games per year? I'm not sure where that idea came from.

Fencer, in his posting earlier today. With N == 14.

Fencer: using that formula, as Walter pointed out, many legitimate players will be hurt as well. The cheaters can circumvent it by creating more accounts. Ultimately, you will deal with them on a case-by-case basis, there is no blanket solution to this problem.

playBunny: argh! i hate laptops .. pressed backspace by accident

i agree that players should be free in their choice of opponents .. but when they want to play for the rating some requirements could be applied .. for example needing a certain variety in opponents .. (lets say when a player played more than 50% of his total games with the same opponent then the rating will be slightly adjusted .. i like abigails idea of normalizing) .. of course this doesnt prevent someone to create 2 extra nicks to cheat with

the 'success' can be calculated better from the win/loss ratio

i am ok with the current system though ;)

AbigailII: N games per year? I'm not sure where that idea came from. Are you referring back to anything I should know about? Either way I agree; it's not useful to impose any such condition.

As for restricting the influence of a pair's large number of games? Horrors, that's the last thing I'd want (if these are legitimate games). Any of the various playing patterns: two people playing a lot against each other, a group playing a lot between themselves, high-level players only playing other high-level players, lower ratings only playing lower ratings, etc - these all do their bit to make the flat lake of perfect ratings/rankings into a somewhat choppy sea. Removing the cheats who cause waves is a good thing but to control any of the other winds and currents .. is it necessary?

But if it is, I reckon the only fair way to ensure the accuracy of all BKRs is to force every player to play every other player in the same number of games within a given time frame. To ensure that last condition, slow players would have their moves made for them by an artificial intelligence whose playing level exactly matches the players style and BKR.

I prepare a new system of Brains which is more important at the moment.

Andersp: I was waiting for your reaction, as usually. I don't have time to make such changes anyway.

I think it's great that the message box can now have its size specified. Unfortunately its the same size in all contexts and on the game boards I find that more than 4 lines is too big (given that I mostly don't use it), while on the discussion boards 4 lines is too small. Going back and forth to the Change message area size is a cumbersome workaround.

What would be really handy is an extra javascript capability - a button to immediately (but temporarily) resize the message box. Or for it to expand to a designated size when it gets focus, or when there's a subject in the subject box (which would encourage people to actually provide a subject) .. or some other such big-only-when-required mechanism.

playBunny: The same applies but to a diluted degree if a player restricts themselves to a small group of opponents who may pr may not play more widely themselves. That's the issue that I think Fencer is looking at and it's a fair one from the game pool/rankings perspective.

That's a fair one, but IMO, that also isn't solved by only counting the first N games played against each other in a year. You're not getting a more accurate rating by giving more weight to games played in January than the games played in December.

If you want to restrict the influence of a large number of games a pair of people played against each other on the BKR, perhaps something can be done to "normalize" the results. Say you want to count all the games as "as most 14 games", and a pair of players, say A and B, have played 48 games, with 30 wins for A, and 18 for B. Then normalize that as 8.75 for A and 5.25 for B. (30 / 48 * 14 = 8.75 and 18 / 48 * 14 = 5.25). Of course, this is simpler said that done, as the in the current system not only the number of wins/losses matters, but also the order in which they were made, and what the ratings were at the moment games finished.

NOT a floosie: I know you were addressing some naughty rooks down below but if it's pawns who you find straying from the path, they might not understand it when you plant a *109* for their attention.
;-)