What's wrong with deflating my inflated rating and working my way up?

I find this kind of thing more educational. In this game specifically, I had to work my rooks to get the win. I didn't have any of it calculated out, and now I can go over it and make a study of it.

I just played someone just under 1000. They threw out a Polerio (well, actually I did, but they gave me that choice). Usually, white plays Ba4 and then better players play Bd3. This player may have played an inferior line but it made me prove I was a better player. If I played 1400+ players they are more likely to go down a different path. I find this kind of thing more educational. In this game specifically, I had to work my rooks to get the win. I didn't have any of it calculated out, and now I can go over it and make a study of it.

AsDaGo

#22

@AlligatorChomps said in #20:

A game of chess where my rating jumps around a lot? Let's do the math.

A 1300 player wins their first two games and gets a rating of 1800+s. That means they will fall 500 points. This also means people are playing them with that huge of a difference.

I don't really understand what you're trying to say here, but if I lose or win to someone with a high RD, my rating won't change much but there's will. The lower the RD, the more accurate the rating is likely to be.

I am not going down to 300 or 600 or even 900. I think working on 1000-1500 is ideal for me and if I want to, I shouldn't be punished for it. If I can beat 1500s easily, then you can shift it to 1100-1600. After that 1200-1700, and so on. I honestly don't see the problem with that.

You can click "create a game" on the Lichess homepage and select a rating range if you only want to play with players of a certain rating. The problem is you ruining the experience for everyone else just because you think you understand the rating system better than Mark Glickman and the Lichess devs.

Just because I don't want to "get lucky" winning against a 1800 player shouldn't invalidate my joys in chess. I had meaningful games to study after I went down to 1000 and worked my way up.

I don't know why you wouldn't want to play against players of your own strength, but just use the rating range feature if you really want to, like I said.

But it will also give me the flip of the coin where I could play a 1700 rated player. If I fall down to 12-1300 there is no way I am going to stick around until I go back up to 1700 when I can just make a new account and get there sooner.

You can't just say "I want to be 1700" and get there immediately, if that's what you're saying.

See how stupid these high fluctuations (you call it doctor backwards)?

I do not. Of course there will be high fluctuations when you make a new account and Lichess has no idea what your strength is. RD stands for rating deviation, by the way. See https://lichess.org/faq#provisional. Lichess uses the Glicko-2 rating system. Look it up.

Or, they can listen to my ideas and try them. Why do nothing and have an army of followers spewing hatred in my direction?

Because your ideas come from a fundamental misunderstanding of the rating system, and you choose to ruin the experience of other players because of that.

I didn't create the system.

Believe me, I can tell.

I can only work with it to get the results I want.

That's called "cheating" (and/or "sandbagging"). You can't just choose the result of the game before you play it if you play legitimately.

I want to work my way up from 1000.

I want to EARN my points.

Then play legitimate games, and pretty soon your rating will reflect your actual strength.

I doubt you'll do that though, you seem very stubborn with your misunderstanding of the rating system. I reported you, by the way. Hopefully they'll IP ban you soon if you keep creating new accounts.

@AlligatorChomps said in #20: > A game of chess where my rating jumps around a lot? Let's do the math. > > A 1300 player wins their first two games and gets a rating of 1800+s. That means they will fall 500 points. This also means people are playing them with that huge of a difference. I don't really understand what you're trying to say here, but if I lose or win to someone with a high RD, my rating won't change much but there's will. The lower the RD, the more accurate the rating is likely to be. > I am not going down to 300 or 600 or even 900. I think working on 1000-1500 is ideal for me and if I want to, I shouldn't be punished for it. If I can beat 1500s easily, then you can shift it to 1100-1600. After that 1200-1700, and so on. I honestly don't see the problem with that. You can click "create a game" on the Lichess homepage and select a rating range if you only want to play with players of a certain rating. The problem is you ruining the experience for everyone else just because you think you understand the rating system better than Mark Glickman and the Lichess devs. > Just because I don't want to "get lucky" winning against a 1800 player shouldn't invalidate my joys in chess. I had meaningful games to study after I went down to 1000 and worked my way up. I don't know why you wouldn't want to play against players of your own strength, but just use the rating range feature if you really want to, like I said. > But it will also give me the flip of the coin where I could play a 1700 rated player. If I fall down to 12-1300 there is no way I am going to stick around until I go back up to 1700 when I can just make a new account and get there sooner. You can't just say "I want to be 1700" and get there immediately, if that's what you're saying. > See how stupid these high fluctuations (you call it doctor backwards)? I do not. Of course there will be high fluctuations when you make a new account and Lichess has no idea what your strength is. RD stands for rating deviation, by the way. See https://lichess.org/faq#provisional. Lichess uses the Glicko-2 rating system. Look it up. > Or, they can listen to my ideas and try them. Why do nothing and have an army of followers spewing hatred in my direction? Because your ideas come from a fundamental misunderstanding of the rating system, and you choose to ruin the experience of other players because of that. > I didn't create the system. Believe me, I can tell. > I can only work with it to get the results I want. That's called "cheating" (and/or "sandbagging"). You can't just choose the result of the game before you play it if you play legitimately. > I want to work my way up from 1000. > > I want to EARN my points. Then play legitimate games, and pretty soon your rating will reflect your actual strength. I doubt you'll do that though, you seem very stubborn with your misunderstanding of the rating system. I reported you, by the way. Hopefully they'll IP ban you soon if you keep creating new accounts.

AlligatorChomps edited

#23

@AsDaGo said in #22:

I don't really understand what you're trying to say here, but if I lose or win to someone with a high RD, my rating won't change much but there's will. The lower the RD, the more accurate the rating is likely to be.

Ok, I will spell out the math then.

A 1500(?) player plays first game and wins, goes up 250 points.

The 1500(?) player is now seen as 1750. They play a second game and win. Maybe they don't get 250, but they get 200.

The 1500(?) player is now seen as 1950. They play a third game with a 1900 player and lose.

It doesn't matter if the RD being high with the 1500(?) player and the low RD with the 1900 doesn't change the 1900 player's rating much. The point is the 1900 has to potentially play the 1500(?) player when maybe they would rather play someone in their range.

This is the opposite argument people are telling me about lowering my rating, that 1000s don't want to play with me. Well, 1900s probably don't want to play with me either. The answer I see is to NOT inflate the ratings early on. Earn them. 5 wins at 20 points is 100. I see nothing wrong with this.

The problem is you ruining the experience for everyone else just because you think you understand the rating system better than Mark Glickman and the Lichess devs.

I think it is a wonderful setup for 2000+ players. The deviation between 1000-1500 is so much different than 2000-2500. Even when you become a GM at 2500, many don't stay at 2500 or move into that 2700-2750 super GM level.

Think of kids growing up. They start out small and can't pick up mom and dad. As they get older in their teens, it's possible. By high school no problem. Their strength then bottoms out when comparing non-weightlifters.

Look at Mishra Abimanyu now and his performance in the Spring classic. He became the youngest GM because he played a certain pool of players at a certain strength. Now he is playing a much tougher crowd. He is the lowest rated and at the bottom of the cross table.

This is similar to NMs not performing as well as IMs. The NM title only reflects their country's performance, while the IM title reflects a more accurate assessment. Both Mishra and the NM are "overrated" with respects to playing strength. So, I find a better assessment would be Mishra has the GM title in the first division of players. The second division would be those that win against 2600 GMs (or maybe even break it up into 2500, 2550, and 2600 by 50s instead which would definitely work better for women's chess).

I doubt you'll do that though, you seem very stubborn with your misunderstanding of the rating system. I reported you, by the way. Hopefully they'll IP ban you soon if you keep creating new accounts.

I hope they don't. I hope they listen to what I am stating here. I hope they try it out. I hope they don't resort to benign actions which don't help anyone.

If we are here to play chess, let's play chess. I am not telling you that you have to do what I am saying or you can't do what you want when playing chess, but you are definitely telling me I can't do what I want.

@AsDaGo said in #22: > I don't really understand what you're trying to say here, but if I lose or win to someone with a high RD, my rating won't change much but there's will. The lower the RD, the more accurate the rating is likely to be. Ok, I will spell out the math then. A 1500(?) player plays first game and wins, goes up 250 points. The 1500(?) player is now seen as 1750. They play a second game and win. Maybe they don't get 250, but they get 200. The 1500(?) player is now seen as 1950. They play a third game with a 1900 player and lose. It doesn't matter if the RD being high with the 1500(?) player and the low RD with the 1900 doesn't change the 1900 player's rating much. The point is the 1900 has to potentially play the 1500(?) player when maybe they would rather play someone in their range. This is the opposite argument people are telling me about lowering my rating, that 1000s don't want to play with me. Well, 1900s probably don't want to play with me either. The answer I see is to NOT inflate the ratings early on. Earn them. 5 wins at 20 points is 100. I see nothing wrong with this. > The problem is you ruining the experience for everyone else just because you think you understand the rating system better than Mark Glickman and the Lichess devs. I think it is a wonderful setup for 2000+ players. The deviation between 1000-1500 is so much different than 2000-2500. Even when you become a GM at 2500, many don't stay at 2500 or move into that 2700-2750 super GM level. Think of kids growing up. They start out small and can't pick up mom and dad. As they get older in their teens, it's possible. By high school no problem. Their strength then bottoms out when comparing non-weightlifters. Look at Mishra Abimanyu now and his performance in the Spring classic. He became the youngest GM because he played a certain pool of players at a certain strength. Now he is playing a much tougher crowd. He is the lowest rated and at the bottom of the cross table. This is similar to NMs not performing as well as IMs. The NM title only reflects their country's performance, while the IM title reflects a more accurate assessment. Both Mishra and the NM are "overrated" with respects to playing strength. So, I find a better assessment would be Mishra has the GM title in the first division of players. The second division would be those that win against 2600 GMs (or maybe even break it up into 2500, 2550, and 2600 by 50s instead which would definitely work better for women's chess). > I doubt you'll do that though, you seem very stubborn with your misunderstanding of the rating system. I reported you, by the way. Hopefully they'll IP ban you soon if you keep creating new accounts. I hope they don't. I hope they listen to what I am stating here. I hope they try it out. I hope they don't resort to benign actions which don't help anyone. If we are here to play chess, let's play chess. I am not telling you that you have to do what I am saying or you can't do what you want when playing chess, but you are definitely telling me I can't do what I want.

Toadofsky

edited

#24

This is one of the most un-enlightened discussions about ratings I've ever read; there are standards, for example:
http://glicko.net/glicko.html

Those proposing improvements should first fully read and second fully understand how Glicko, Glicko-2, and Glicko-Boost work, then we can talk.
https://xkcd.com/675

I'll start with a basic definition: a player's rating ESTIMATE is defined by an interval:
r ± 2*RD
where RD serves as an estimate of standard error. There is no such thing as a "1500" player.

This is one of the most un-enlightened discussions about ratings I've ever read; there are standards, for example: http://glicko.net/glicko.html Those proposing improvements should first fully read and second fully understand how Glicko, Glicko-2, and Glicko-Boost work, then we can talk. https://xkcd.com/675 I'll start with a basic definition: a player's rating ESTIMATE is defined by an interval: r ± 2*RD where RD serves as an estimate of standard error. There is no such thing as a "1500" player.

AlligatorChomps edited

#25

@Toadofsky said in #24:

I'll start with a basic definition: a player's rating ESTIMATE is defined by an interval:
r ± 2*RD

Somehow 2*RD = 250 then.

RD would have been more than enough = 60.

I have always had my accounts start with 1500.

I would much rather start at 1000 and win 6 games to get those 120 points.

Still not convinced by the Glicko sheeple.

@Toadofsky said in #24: > I'll start with a basic definition: a player's rating ESTIMATE is defined by an interval: > r ± 2*RD Somehow 2*RD = 250 then. RD would have been more than enough = 60. I have always had my accounts start with 1500. I would much rather start at 1000 and win 6 games to get those 120 points. Still not convinced by the Glicko sheeple.

Cedur216

#26

Go to chesscom if you whine so much about a system that brings you to your correct rating range as quickly as possible.

Deadban

#27

@AlligatorChomps said in #25:

RD would have been more than enough = 60.

I reckon it would be pointless to counter your absurd arguments since you clearly have zero idea about how the rating system works and even less will to learn about it.

Instead there we go, you claim to know better than mathematician Prof. Glickman, Boston University, inventor or the glicko system. So now what I want from you is to know your credentials, who are you? What are you studies? What's your background?

I would much rather start at 1000 and win 6 games to get those 120 points.

You can do that in chess.com, paradise of sandbaggers like you. Have fun among your fellow.

@AlligatorChomps said in #25: > RD would have been more than enough = 60. I reckon it would be pointless to counter your absurd arguments since you clearly have zero idea about how the rating system works and even less will to learn about it. Instead there we go, you claim to know better than mathematician Prof. Glickman, Boston University, inventor or the glicko system. So now what I want from you is to know your credentials, who are you? What are you studies? What's your background? > I would much rather start at 1000 and win 6 games to get those 120 points. You can do that in chess.com, paradise of sandbaggers like you. Have fun among your fellow.

Brian-E

edited

#28

@AlligatorChomps Aside from all the detailed discussion about ratings in your thread above, please keep in mind the following. When you challenge a player or accept their challenge for a game of chess, your opponent will be expecting you to play chess properly and give them a real game, as opposed to having you deliberately throw it. You are wasting that opponent's time when they could be playing someone else who would give them a proper game. If I thought I had just time for a nice quick game of chess, played you and had you obviously deliberately lose, I'd be seriously annoyed. It is not acceptable.

Toadofsky

#29

@AlligatorChomps said in #25:

Still not convinced by the Glicko sheeple.

Then go invent your own math, or make your own chess web site. Surely there are similar-minded people in the world, if what you are saying is true.

@AlligatorChomps said in #25: > Still not convinced by the Glicko sheeple. Then go invent your own math, or make your own chess web site. Surely there are similar-minded people in the world, if what you are saying is true.

AlligatorChomps

#30

@Brian-E said in #28:

@AlligatorChomps Aside from all the detailed discussion about ratings in your thread above, please keep in mind the following. When you challenge a player or accept their challenge for a game of chess, your opponent will be expecting you to play chess properly and give them a real game, as opposed to having you deliberately throw it. You are wasting that opponent's time when they could be playing someone else who would give them a proper game. If I thought I had just time for a nice quick game of chess, played you and had you obviously deliberately lose, I'd be seriously annoyed. It is not acceptable.

Thank you for taking a different slant than those "If you don't like it here, get out of our country (and go to chessDOTcom)" replies. Now, what I am "experiencing" here is nowhere near what you may have experienced as mentioned in your profile.

However, I see a drop of similarity. The initial reaction is that you aren't degrading me for "my math".

You state:

When you challenge a player or accept their challenge for a game of chess, your opponent will be expecting you to play chess properly and give them a real game, as opposed to having you deliberately throw it.

This is a big reason why I prefer longer chess over blitz. We have that choice. We can choose the time controls. I won't go into the rating discussion more here, because you already referenced it. If you have a rating comment or question feel free to give that. We also have the choice who we can play. This one might be a tricky issue.

This is not a human rights issue. It's not saying, "I don't want to play with gays." or "I don't want to play Russians."

This should be an individual decision. I want to be "transgender rating". I want to change the rating I was born with on this site. I want to play starting at 1000 (analogous to chopping something off if you want to carry the analogy further). Others may want the opposite (I have no idea how women before transitioning get one attached down there) and start with 2000.

That should be the choice of each and all of us. If you think you are 2000, great, start at 2000. If you want to start at 1500, start there. If I want to start at 1000 I should have this choice too. We all should be allowed to start where we want.

Let's work together and fight this luddite movement.

@Brian-E said in #28: > @AlligatorChomps Aside from all the detailed discussion about ratings in your thread above, please keep in mind the following. When you challenge a player or accept their challenge for a game of chess, your opponent will be expecting you to play chess properly and give them a real game, as opposed to having you deliberately throw it. You are wasting that opponent's time when they could be playing someone else who would give them a proper game. If I thought I had just time for a nice quick game of chess, played you and had you obviously deliberately lose, I'd be seriously annoyed. It is not acceptable. Thank you for taking a different slant than those "If you don't like it here, get out of our country (and go to chessDOTcom)" replies. Now, what I am "experiencing" here is nowhere near what you may have experienced as mentioned in your profile. However, I see a drop of similarity. The initial reaction is that you aren't degrading me for "my math". You state: > When you challenge a player or accept their challenge for a game of chess, your opponent will be expecting you to play chess properly and give them a real game, as opposed to having you deliberately throw it. This is a big reason why I prefer longer chess over blitz. We have that choice. We can choose the time controls. I won't go into the rating discussion more here, because you already referenced it. If you have a rating comment or question feel free to give that. We also have the choice who we can play. This one might be a tricky issue. This is not a human rights issue. It's not saying, "I don't want to play with gays." or "I don't want to play Russians." This should be an individual decision. I want to be "transgender rating". I want to change the rating I was born with on this site. I want to play starting at 1000 (analogous to chopping something off if you want to carry the analogy further). Others may want the opposite (I have no idea how women before transitioning get one attached down there) and start with 2000. That should be the choice of each and all of us. If you think you are 2000, great, start at 2000. If you want to start at 1500, start there. If I want to start at 1000 I should have this choice too. We all should be allowed to start where we want. Let's work together and fight this luddite movement.

This topic has been archived and can no longer be replied to.