RHP ratings inflated?

Originally posted by Steve Exeter
RTh, I didn't come here to whinge about time outs at RHP and I'm not playing any games because whilst I don't agree with the decision to ban me from playing here, I have to respect it. However, I see no harm in my commenting in the forums from time to time.

Actually, I was just curious to know if the differing ways in which the two chess sites (th ...[text shortened]... o see at least one player rated over 2200 and many around the 2000 mark but it's not so. 😕

Doesn't ChessKnights have only a very small pool of players?

Originally posted by dottewell
Doesn't ChessKnights have only a very small pool of players?

With a very shallow gene pool end.

Originally posted by dottewell
Doesn't ChessKnights have only a very small pool of players?

Does that matter? If the pool is dominated by engine users then the ratings would increase exponentially, surely? When I started playing here 2 years ago players rated over 2000 were confined to the top half of the 1st page of the player tables, not the first 4 pages. 😕

Originally posted by Steve Exeter
Does that matter? If the pool is dominated by engine users then the ratings would increase exponentially, surely? When I started playing here 2 years ago players rated over 2000 were confined to the top half of the 1st page of the player tables, not the first 4 pages. 😕

I didn't realise we were talking about engine use.

I thought your point was that RHP ratings might be inflated and ChessKnights were more "accurate".

Any rating system is relative to the pool, but the bigger the pool, the more likely your rating (or at least your ranking on the player tables) will give a good idea of your absolute strength.

I'd rather think of a ratings spread, than inflation. It has everything to do with the passage of time and the quantitty of players in the ratings pool, and nothing at all to do with engine use. The ratings spread is equally apparent on the lower end of the ratings.

All that engine use can do is change the order in which the players are ranked, it has no impact on the process itself.

Imagine a large bucket and small bucket of sand, poured out into seperate heaps. They will have the same shape, but not only will the larger pile be taller, it will also be wider. It spreads further.

Since everyone starts out at 1200, the spread also increases with time. Again, you can use the sand analogy, tap the surface the piles are on, and the sand will settle more and spread further. And because the pool is constrantly increasing, more sand is being piled on, and there is yet more spread.

In order for a highly rated player to increase his rating, he has to have similarly rated players to play against. You could not for example, have a 2400 player if there were no other 2000+ players in the ratings pool, even if he beat everyone, everytime, his rating would stagnate until he has opponents with similar ratings.

It also gets harder the higher the ratings get. Brakes are applied. In order to gain 8 rating points, a player rated below 2100 must beat a player rated 200 point lower than them. A 2100 player, must beat a player rated 125 points lower than them. A 2400 player has to beat another 2400 player.

Ratings should be seen as a relative score among the players in the same pool. You can't directly compare it to other ratings pools or systems. There will be some similarities, but there are also many differences. One player may have vastly different ratings across pools, depending on the type of chess played, and the size and age of the pool.

At RHP, timeouts are an issue. Ratings get severely warped by timeouts, but even more so with mass resignations. I once gained over 100 points in a few days solely because 2 strong players resigned all their games and left the site. RHP ratings are live, instant and fluid. FIDE ratings, by comparison, are far more static, and only revised once every three months and even then all anomalies are removed.

There could be improvements, though. Especially a ratings floor (at least for calculation purposes) for those who allow all their games to timeout or indulge in mass-resignation binges. Its also important for the stability of the rating system that points gained are equivalent to points lost.

Originally posted by Gatecrasher
At RHP, timeouts are an issue. Ratings get severely warped by timeouts, but even more so with mass resignations. I once gained over 100 points in a few days solely because 2 strong players resigned all their games against me. RHP ratings are live, instant and fluid. FIDE ratings, by comparison, are far more static, and only revised every three months and even then all anomalies are removed.

That was my original supposition, before the subject was sidetracked by the issue of engine use at Chess Knights. I just feel that my rating there without the swings from time outs and mass resignations is more stable than my one at RHP which never would have reached 2000+, however briefly, without the help of the Skull!

Originally posted by Steve Exeter
That was my original supposition, before the subject was sidetracked by the issue of engine use at Chess Knights. I just feel that my rating there without the swings from time outs and mass resignations is more stable than my one at RHP which never would have reached 2000+, however briefly, without the help of the Skull!

Your rating should have never reached 2000+ without the chess engine.

The irony of your statement is delicious. Now run along now loser.


RTh

Originally posted by Ringtailhunter
Your rating should have never reached 2000+ without the chess engine.

The irony of your statement is delicious. Now run along now loser.


RTh

Which chess engine was that? I'm all ears to know which so I can purchase it and check all of my games! 😛

You still haven't explained why the ratings at Chess Knights seem naturally capped under 2000 if engine use is so rife?

Feel free to check any of my games there also.

Originally posted by Steve Exeter
Which chess engine was that? I'm all ears to know which so I can purchase it and check all of my games! 😛

You still haven't explained why the ratings at Chess Knights seem naturally capped under 2000 if engine use is so rife?

Feel free to check any of my games there also.

If the majority of games are played by chess engines the ratings will be flat because of the number of draws. Also without the volume and winning those games rating will move slower. It took (correct me if I'm wrong) about 4-5 years for rhp to get players who stayed above 2200. The high ratings here are pretty recent.

I cannot tell you what engine you have. I got pretty good matchups with winboard and also some of the freeware UCI engines. I don't think you used an engine for every move in every game but you certainly did when it counted.

Originally posted by Ringtailhunter
If the majority of games are played by chess engines the ratings will be flat because of the number of draws.

Mike, have you made any study of the players, ratings and results of the matches at Chess Knights or are you just talking off the top of your head? 😕

I have played 130 games at Chess Knights, my current rating is 1694, and my highest rating has been 1823. I have won 76 [59%], lost 49 [38%] but only drawn 5 [3%] so either I'm not using an engine (or I am but the majority of people I have beaten aren't) and those that I've lost to have a much stronger engine than I have.

Either way your 'theory' that the majority of players at Chess Knights use engines can't be right because there isn't the high number of draws you talk about. In fact the rating spread of the top 100 ranges from 1993 down to 1568.

I seriously believe that if I offered the .pgns of my Chess Knight games to the Game Mod team here that the evidence of the moves alone would not be enough to ban me from RHP. People can draw their own conclusions from that, I know what mine are. However, if any of the Mods are willing to take a look and do some real analysis then please contact me via PM and I’ll forward them the games.

Originally posted by Steve Exeter
Which chess engine was that? I'm all ears to know which so I can purchase it and check all of my games! 😛

You still haven't explained why the ratings at Chess Knights seem naturally capped under 2000 if engine use is so rife?

Feel free to check any of my games there also.

if you didn't use an engine, wouldn't it be pretty easy to prove you'd have the potential to reach your rhp-rating by showing comparative ratings from other (reliable) sources?

maybe you have an official rating, and can figure out how to prove your identity? if you were a subscriber the admins would already know your real identity from the billing. or maybe some stronger player here lives near you, and could play a couple of games otb to verify your level? maybe you could show you can perform tactically according to your rhp-rating at chess tactics server? none of these are conclusive evidence, as they're not exactly the same as CC, and even an official CC-rating could probably be 'engined'? but they would likely prove your ability beyond reasonable doubt.

a young player here keeps saying he's a strong player but his rhp-rating is 1500-ish. normally I don't believe such claims, of course. but then, one day I asked if he could show at CTS how he performed. -now, there was an account by his name there, but it had been idle for a couple of months. I also had my doubts about wether he really was the owner of that account. within an hour after my request he had activated the account, and outperformed people I knew to be 2100+ fide. and although tactical ability isn't the same as playing CC, that pretty much cleared all my doubts.

Originally posted by wormwood
if you didn't use an engine, wouldn't it be pretty easy to prove you'd have the potential to reach your rhp-rating by showing comparative ratings from other (reliable) sources?

maybe you have an official rating, and can figure out how to prove your identity? if you were a subscriber the admins would already know your real identity from the billing. or m tactical ability isn't the same as playing CC, that pretty much cleared all my doubts.

What's the point in that, it didn't save Ironman31! In any case, I play chess online because I enjoy a game, it's not my whole life, I don't have time in real life to sit down and play chess at competition level OTB. There are people here who can attest to my real ability and did when I was first banned.

I just never expected in my wildest dreams to be booted off here before some of the players who haven't lost a game and have ratings approaching 2300!

We ought to get this thread back on topic, or at least off the topic of my ban, or it will be closed or deleted.

🙄

Originally posted by Steve Exeter
Did I speak to soon?! He's no longer on the removed players list and if you check out his tournament victories it says "Ironman31 Wins" 😲

Has Ironman's section (b) ban been officially overturned?

IronMan31 section 3(b) 01 Sep '05 16:32

4th from the bottom

Originally posted by Chakan
IronMan31 section 3(b) 01 Sep '05 16:32

4th from the bottom

Yeah, I saw that, I thought they were in chronological order, but they're not. Still don't see why his tournament wins haven't been scratched through like the others.