Amazing results of LeelaKnightOdds-Dev

[Father]

It seems that fixing the bug change the table and now this is the table(I do not complain about reduction in the place of me that is logical but how it is possible that Catecan went down to 2291.9?)

I thought that even with dividing K by 2 considering the fact that he played so many games his rating should not go down by a lot because it already converged to his performance against the bot.
The only case when his rating should go down by a lot is the case that he improved significantly in the last games so the rating still did not converge to be close to the right rating.

Even I went down only by 30 elo or something like that and I could see 2246 or something close earlier after part of the update so I guess it was my rating before fixing the bug.

LeelaQueenOdds as white 1+0 (Reference) 2895.2
LeelaQueenOdds as black 1+0 (Reference) 2795.2
LeelaQueenOdds as white 1+1 (Reference) 2634.5
LeelaQueenOdds as white 3+0 (Reference) 2619.5
LeelaQueenOdds as black 1+1 (Reference) 2534.5
1 Orichess88 2524.5 73
LeelaQueenOdds as black 3+0 (Reference) 2519.5
2 Master_Unknown 2514.6 92
LeelaQueenOdds as white 3+2 (Reference) 2450.0
3 Ushkothebear 2443.7 184
4 Hissha 2398.4 94
5 Omnivor 2396.8 30
LeelaQueenOdds as white 5+3 (Reference) 2353.7
LeelaQueenOdds as black 3+2 (Reference) 2350.0
6 Respectful_Dave 2299.6 74
7 itore 2297.1 43
8 Catecan 2291.9 1288
LeelaQueenOdds as black 5+3 (Reference) 2253.7
9 Sakali 2251.1 146
10 Spreek 2229.8 45
11 sfy2011 2223.6 25
12 SilverMitt 2223.1 26
LeelaQueenOdds as white 10+5 (Reference) 2220.3
13 uriblass 2216.4 182
14 nn1133 2205.6 6346

Catecan went down because its recent performance is better than its older performance, all it has to do is maintain this performance and it will continue to rise.

Good afternoon Marcus91. I really need the help of mathematicians. If one day Mr. Larry Kaufman, or any of you, invites me to play chess computers, I will not bring a psychologist, a doctor, or a team of analysts. I will only require, humanly speaking, a mathematics teacher to help me. What has happened to my elo has been completely incomprehensible to me. I'm a real yo-yo that has me on the brink of collapse. I simply ask all of you to collaborate with me, I don't know what's happening, the more I play, the lower my rating goes, even though I'm feeling progress. Besides, I know that I will be able to win by the minute later, I just don't care at all at the moment. I'm simply doing the same thing that Kasparov did with Karpov before he defeated him and challenged him for the title of world champion. exhaust him, break him mentally. I am leading the algorithm to collapse. Marcus91. At this moment I have 1388 games. Please help me and tell me how much my elo is left... please: Today we have 100 new bullet games, we have reached 38 draws and 62 losses. Sincere thanks in advance.

[Uri Blass]

It seems that fixing the bug change the table and now this is the table(I do not complain about reduction in the place of me that is logical but how it is possible that Catecan went down to 2291.9?)

I thought that even with dividing K by 2 considering the fact that he played so many games his rating should not go down by a lot because it already converged to his performance against the bot.
The only case when his rating should go down by a lot is the case that he improved significantly in the last games so the rating still did not converge to be close to the right rating.

Even I went down only by 30 elo or something like that and I could see 2246 or something close earlier after part of the update so I guess it was my rating before fixing the bug.

LeelaQueenOdds as white 1+0 (Reference) 2895.2
LeelaQueenOdds as black 1+0 (Reference) 2795.2
LeelaQueenOdds as white 1+1 (Reference) 2634.5
LeelaQueenOdds as white 3+0 (Reference) 2619.5
LeelaQueenOdds as black 1+1 (Reference) 2534.5
1 Orichess88 2524.5 73
LeelaQueenOdds as black 3+0 (Reference) 2519.5
2 Master_Unknown 2514.6 92
LeelaQueenOdds as white 3+2 (Reference) 2450.0
3 Ushkothebear 2443.7 184
4 Hissha 2398.4 94
5 Omnivor 2396.8 30
LeelaQueenOdds as white 5+3 (Reference) 2353.7
LeelaQueenOdds as black 3+2 (Reference) 2350.0
6 Respectful_Dave 2299.6 74
7 itore 2297.1 43
8 Catecan 2291.9 1288
LeelaQueenOdds as black 5+3 (Reference) 2253.7
9 Sakali 2251.1 146
10 Spreek 2229.8 45
11 sfy2011 2223.6 25
12 SilverMitt 2223.1 26
LeelaQueenOdds as white 10+5 (Reference) 2220.3
13 uriblass 2216.4 182
14 nn1133 2205.6 6346

Catecan went down because its recent performance is better than its older performance, all it has to do is maintain this performance and it will continue to rise.

I see the leader has 73 games in the table but search found only 65 games

https://lichess.org/@/Orichess88/search ... order=desc

What is the reason for the difference?

[lkaufman]

It seems that fixing the bug change the table and now this is the table(I do not complain about reduction in the place of me that is logical but how it is possible that Catecan went down to 2291.9?)

I thought that even with dividing K by 2 considering the fact that he played so many games his rating should not go down by a lot because it already converged to his performance against the bot.
The only case when his rating should go down by a lot is the case that he improved significantly in the last games so the rating still did not converge to be close to the right rating.

Even I went down only by 30 elo or something like that and I could see 2246 or something close earlier after part of the update so I guess it was my rating before fixing the bug.

LeelaQueenOdds as white 1+0 (Reference) 2895.2
LeelaQueenOdds as black 1+0 (Reference) 2795.2
LeelaQueenOdds as white 1+1 (Reference) 2634.5
LeelaQueenOdds as white 3+0 (Reference) 2619.5
LeelaQueenOdds as black 1+1 (Reference) 2534.5
1 Orichess88 2524.5 73
LeelaQueenOdds as black 3+0 (Reference) 2519.5
2 Master_Unknown 2514.6 92
LeelaQueenOdds as white 3+2 (Reference) 2450.0
3 Ushkothebear 2443.7 184
4 Hissha 2398.4 94
5 Omnivor 2396.8 30
LeelaQueenOdds as white 5+3 (Reference) 2353.7
LeelaQueenOdds as black 3+2 (Reference) 2350.0
6 Respectful_Dave 2299.6 74
7 itore 2297.1 43
8 Catecan 2291.9 1288
LeelaQueenOdds as black 5+3 (Reference) 2253.7
9 Sakali 2251.1 146
10 Spreek 2229.8 45
11 sfy2011 2223.6 25
12 SilverMitt 2223.1 26
LeelaQueenOdds as white 10+5 (Reference) 2220.3
13 uriblass 2216.4 182
14 nn1133 2205.6 6346

Catecan went down because its recent performance is better than its older performance, all it has to do is maintain this performance and it will continue to rise.

Good afternoon Marcus91. I really need the help of mathematicians. If one day Mr. Larry Kaufman, or any of you, invites me to play chess computers, I will not bring a psychologist, a doctor, or a team of analysts. I will only require, humanly speaking, a mathematics teacher to help me. What has happened to my elo has been completely incomprehensible to me. I'm a real yo-yo that has me on the brink of collapse. I simply ask all of you to collaborate with me, I don't know what's happening, the more I play, the lower my rating goes, even though I'm feeling progress. Besides, I know that I will be able to win by the minute later, I just don't care at all at the moment. I'm simply doing the same thing that Kasparov did with Karpov before he defeated him and challenged him for the title of world champion. exhaust him, break him mentally. I am leading the algorithm to collapse. Marcus91. At this moment I have 1388 games. Please help me and tell me how much my elo is left... please: Today we have 100 new bullet games, we have reached 38 draws and 62 losses. Sincere thanks in advance.

If you are able to improve enough to get one draw out of every two games (so up from 38% draws to 50% draws), and keep that level for hundreds of games, your rating should approach 2515 under current rules, which would be a tie for second place right now. If draws were fully counted, it would approach (2795-191 = 2604), so counting only half the draws (technically half of each draw) reduces the expected rating by nearly 90 elo. But it also hurts the elo of all the other players who play at bullet time controls where they are usually only getting draws at best, so it's not just lowering your elo. But it does reward players for choosing time limits where they can actually win games, which is after all supposed to be the goal of a chess game. Since Leela doesn't mind making draws when down a queen, it seems reasonable to me. Note that all the players above you on the list do play at time limits where they win a lot of games.

[Uri Blass]

It seems that fixing the bug change the table and now this is the table(I do not complain about reduction in the place of me that is logical but how it is possible that Catecan went down to 2291.9?)

I thought that even with dividing K by 2 considering the fact that he played so many games his rating should not go down by a lot because it already converged to his performance against the bot.
The only case when his rating should go down by a lot is the case that he improved significantly in the last games so the rating still did not converge to be close to the right rating.

Even I went down only by 30 elo or something like that and I could see 2246 or something close earlier after part of the update so I guess it was my rating before fixing the bug.

LeelaQueenOdds as white 1+0 (Reference) 2895.2
LeelaQueenOdds as black 1+0 (Reference) 2795.2
LeelaQueenOdds as white 1+1 (Reference) 2634.5
LeelaQueenOdds as white 3+0 (Reference) 2619.5
LeelaQueenOdds as black 1+1 (Reference) 2534.5
1 Orichess88 2524.5 73
LeelaQueenOdds as black 3+0 (Reference) 2519.5
2 Master_Unknown 2514.6 92
LeelaQueenOdds as white 3+2 (Reference) 2450.0
3 Ushkothebear 2443.7 184
4 Hissha 2398.4 94
5 Omnivor 2396.8 30
LeelaQueenOdds as white 5+3 (Reference) 2353.7
LeelaQueenOdds as black 3+2 (Reference) 2350.0
6 Respectful_Dave 2299.6 74
7 itore 2297.1 43
8 Catecan 2291.9 1288
LeelaQueenOdds as black 5+3 (Reference) 2253.7
9 Sakali 2251.1 146
10 Spreek 2229.8 45
11 sfy2011 2223.6 25
12 SilverMitt 2223.1 26
LeelaQueenOdds as white 10+5 (Reference) 2220.3
13 uriblass 2216.4 182
14 nn1133 2205.6 6346

Catecan went down because its recent performance is better than its older performance, all it has to do is maintain this performance and it will continue to rise.

Good afternoon Marcus91. I really need the help of mathematicians. If one day Mr. Larry Kaufman, or any of you, invites me to play chess computers, I will not bring a psychologist, a doctor, or a team of analysts. I will only require, humanly speaking, a mathematics teacher to help me. What has happened to my elo has been completely incomprehensible to me. I'm a real yo-yo that has me on the brink of collapse. I simply ask all of you to collaborate with me, I don't know what's happening, the more I play, the lower my rating goes, even though I'm feeling progress. Besides, I know that I will be able to win by the minute later, I just don't care at all at the moment. I'm simply doing the same thing that Kasparov did with Karpov before he defeated him and challenged him for the title of world champion. exhaust him, break him mentally. I am leading the algorithm to collapse. Marcus91. At this moment I have 1388 games. Please help me and tell me how much my elo is left... please: Today we have 100 new bullet games, we have reached 38 draws and 62 losses. Sincere thanks in advance.

I see that I missed your post but your rating already improved.

Did not post the new rating list earlier because I do not understand how the leader has 73 games and not 65 games(my number is fixed) but you improved to rating of 2336.6(I also can get it if I continue to beat Leela at 15+10 with white but I will need a lot of games for it and I will probably try the next level that is white at 10+5 time control later but I do not hurry and I guess I will devote more time to try to improve in normal chess against humans).

1 Orichess88 2524.5 73
LeelaQueenOdds as black 3+0 (Reference) 2519.5
2 Master_Unknown 2514.6 92
LeelaQueenOdds as white 3+2 (Reference) 2450.0
3 Ushkothebear 2443.7 184
4 Hissha 2398.4 94
5 Omnivor 2396.8 30
LeelaQueenOdds as white 5+3 (Reference) 2353.7
LeelaQueenOdds as black 3+2 (Reference) 2350.0
6 Catecan 2336.6 1396
7 Respectful_Dave 2299.6 74
8 itore 2297.1 43
9 Bonitta 2286.6 29
LeelaQueenOdds as black 5+3 (Reference) 2253.7
10 Sakali 2251.1 146
11 Spreek 2229.8 45
12 uriblass 2226.3 185
13 sfy2011 2223.6 25
14 SilverMitt 2223.1 26
LeelaQueenOdds as white 10+5 (Reference) 2220.3
15 nn1133 2205.6 6346
16 AxelBoldt 2199.9 306

[Father]

It seems that fixing the bug change the table and now this is the table(I do not complain about reduction in the place of me that is logical but how it is possible that Catecan went down to 2291.9?)

I thought that even with dividing K by 2 considering the fact that he played so many games his rating should not go down by a lot because it already converged to his performance against the bot.
The only case when his rating should go down by a lot is the case that he improved significantly in the last games so the rating still did not converge to be close to the right rating.

Even I went down only by 30 elo or something like that and I could see 2246 or something close earlier after part of the update so I guess it was my rating before fixing the bug.

LeelaQueenOdds as white 1+0 (Reference) 2895.2
LeelaQueenOdds as black 1+0 (Reference) 2795.2
LeelaQueenOdds as white 1+1 (Reference) 2634.5
LeelaQueenOdds as white 3+0 (Reference) 2619.5
LeelaQueenOdds as black 1+1 (Reference) 2534.5
1 Orichess88 2524.5 73
LeelaQueenOdds as black 3+0 (Reference) 2519.5
2 Master_Unknown 2514.6 92
LeelaQueenOdds as white 3+2 (Reference) 2450.0
3 Ushkothebear 2443.7 184
4 Hissha 2398.4 94
5 Omnivor 2396.8 30
LeelaQueenOdds as white 5+3 (Reference) 2353.7
LeelaQueenOdds as black 3+2 (Reference) 2350.0
6 Respectful_Dave 2299.6 74
7 itore 2297.1 43
8 Catecan 2291.9 1288
LeelaQueenOdds as black 5+3 (Reference) 2253.7
9 Sakali 2251.1 146
10 Spreek 2229.8 45
11 sfy2011 2223.6 25
12 SilverMitt 2223.1 26
LeelaQueenOdds as white 10+5 (Reference) 2220.3
13 uriblass 2216.4 182
14 nn1133 2205.6 6346

Catecan went down because its recent performance is better than its older performance, all it has to do is maintain this performance and it will continue to rise.

Good afternoon Marcus91. I really need the help of mathematicians. If one day Mr. Larry Kaufman, or any of you, invites me to play chess computers, I will not bring a psychologist, a doctor, or a team of analysts. I will only require, humanly speaking, a mathematics teacher to help me. What has happened to my elo has been completely incomprehensible to me. I'm a real yo-yo that has me on the brink of collapse. I simply ask all of you to collaborate with me, I don't know what's happening, the more I play, the lower my rating goes, even though I'm feeling progress. Besides, I know that I will be able to win by the minute later, I just don't care at all at the moment. I'm simply doing the same thing that Kasparov did with Karpov before he defeated him and challenged him for the title of world champion. exhaust him, break him mentally. I am leading the algorithm to collapse. Marcus91. At this moment I have 1388 games. Please help me and tell me how much my elo is left... please: Today we have 100 new bullet games, we have reached 38 draws and 62 losses. Sincere thanks in advance.

I see that I missed your post but your rating already improved.

Did not post the new rating list earlier because I do not understand how the leader has 73 games and not 65 games(my number is fixed) but you improved to rating of 2336.6(I also can get it if I continue to beat Leela at 15+10 with white but I will need a lot of games for it and I will probably try the next level that is white at 10+5 time control later but I do not hurry and I guess I will devote more time to try to improve in normal chess against humans).

1 Orichess88 2524.5 73
LeelaQueenOdds as black 3+0 (Reference) 2519.5
2 Master_Unknown 2514.6 92
LeelaQueenOdds as white 3+2 (Reference) 2450.0
3 Ushkothebear 2443.7 184
4 Hissha 2398.4 94
5 Omnivor 2396.8 30
LeelaQueenOdds as white 5+3 (Reference) 2353.7
LeelaQueenOdds as black 3+2 (Reference) 2350.0
6 Catecan 2336.6 1396
7 Respectful_Dave 2299.6 74
8 itore 2297.1 43
9 Bonitta 2286.6 29
LeelaQueenOdds as black 5+3 (Reference) 2253.7
10 Sakali 2251.1 146
11 Spreek 2229.8 45
12 uriblass 2226.3 185
13 sfy2011 2223.6 25
14 SilverMitt 2223.1 26
LeelaQueenOdds as white 10+5 (Reference) 2220.3
15 nn1133 2205.6 6346
16 AxelBoldt 2199.9 306

Mr. Larry Kaufman. With my soul: "Thank you." I've been a little apprehensive. My intention is not to hurt anyone on the Leela team or offend the other participants in the chat. I have simply always been alone in all this and I do not know the dynamics of the federations and clubs. I wish Uri, you and everyone the best. What's more: Lea I offer all my effort against leelaqueenodds. I entrust you to my Programmer.

[Uri Blass]

I believe that Leela can play better espacielly against father without spoiling the results against other players and without knowing nothing except the fact that the opponent does not try to win based on the opponent's moves and the times on the clock.

Here are the last draws and some analysis by me:

https://lichess.org/A0LsV2p6

The last capture was 31.gxh4
There was no reason to allow repetition at move 70 when white had only 17 seconds on the clock and black could continue to waste move without a draw at least until move 79(white need to make move 81 with no pawn move or capture to claim a draw by the 50 move rule so the plan should be to play 80...e5 or 80...Ng3 and then to spend more 50 moves in order to win on time(I believe it can work in bullet 1+0 even against GM's and the way they win some game is not by closing the position.

Second game:

https://lichess.org/cO2DIgtV

30...c5 was the last pawn move. white need to make move 80 with no pawn move to claim a draw.
79...cxd4 could prevent it and I believe black could have good chances to win on time.
(white had only 14.4 seconds on the clock)

third game:
https://lichess.org/s0HwRz13

35...a4 was too early.
I think that black could spend many moves before playing this move without a risk of losing regardless of time control amd also play many moves after finally playing a4 in order to win on time.

[Uri Blass]

It seems that fixing the bug change the table and now this is the table(I do not complain about reduction in the place of me that is logical but how it is possible that Catecan went down to 2291.9?)

I thought that even with dividing K by 2 considering the fact that he played so many games his rating should not go down by a lot because it already converged to his performance against the bot.
The only case when his rating should go down by a lot is the case that he improved significantly in the last games so the rating still did not converge to be close to the right rating.

Even I went down only by 30 elo or something like that and I could see 2246 or something close earlier after part of the update so I guess it was my rating before fixing the bug.

LeelaQueenOdds as white 1+0 (Reference) 2895.2
LeelaQueenOdds as black 1+0 (Reference) 2795.2
LeelaQueenOdds as white 1+1 (Reference) 2634.5
LeelaQueenOdds as white 3+0 (Reference) 2619.5
LeelaQueenOdds as black 1+1 (Reference) 2534.5
1 Orichess88 2524.5 73
LeelaQueenOdds as black 3+0 (Reference) 2519.5
2 Master_Unknown 2514.6 92
LeelaQueenOdds as white 3+2 (Reference) 2450.0
3 Ushkothebear 2443.7 184
4 Hissha 2398.4 94
5 Omnivor 2396.8 30
LeelaQueenOdds as white 5+3 (Reference) 2353.7
LeelaQueenOdds as black 3+2 (Reference) 2350.0
6 Respectful_Dave 2299.6 74
7 itore 2297.1 43
8 Catecan 2291.9 1288
LeelaQueenOdds as black 5+3 (Reference) 2253.7
9 Sakali 2251.1 146
10 Spreek 2229.8 45
11 sfy2011 2223.6 25
12 SilverMitt 2223.1 26
LeelaQueenOdds as white 10+5 (Reference) 2220.3
13 uriblass 2216.4 182
14 nn1133 2205.6 6346

Catecan went down because its recent performance is better than its older performance, all it has to do is maintain this performance and it will continue to rise.

Good afternoon Marcus91. I really need the help of mathematicians. If one day Mr. Larry Kaufman, or any of you, invites me to play chess computers, I will not bring a psychologist, a doctor, or a team of analysts. I will only require, humanly speaking, a mathematics teacher to help me. What has happened to my elo has been completely incomprehensible to me. I'm a real yo-yo that has me on the brink of collapse. I simply ask all of you to collaborate with me, I don't know what's happening, the more I play, the lower my rating goes, even though I'm feeling progress. Besides, I know that I will be able to win by the minute later, I just don't care at all at the moment. I'm simply doing the same thing that Kasparov did with Karpov before he defeated him and challenged him for the title of world champion. exhaust him, break him mentally. I am leading the algorithm to collapse. Marcus91. At this moment I have 1388 games. Please help me and tell me how much my elo is left... please: Today we have 100 new bullet games, we have reached 38 draws and 62 losses. Sincere thanks in advance.

If you are able to improve enough to get one draw out of every two games (so up from 38% draws to 50% draws), and keep that level for hundreds of games, your rating should approach 2515 under current rules, which would be a tie for second place right now. If draws were fully counted, it would approach (2795-191 = 2604), so counting only half the draws (technically half of each draw) reduces the expected rating by nearly 90 elo. But it also hurts the elo of all the other players who play at bullet time controls where they are usually only getting draws at best, so it's not just lowering your elo. But it does reward players for choosing time limits where they can actually win games, which is after all supposed to be the goal of a chess game. Since Leela doesn't mind making draws when down a queen, it seems reasonable to me. Note that all the players above you on the list do play at time limits where they win a lot of games.

I use the following table and get slightly different results(not that it is very important)

https://handbook.fide.com/chapter/B022024

I translate 75% to 193 elo difference and not 191 elo.
I translate 83.333% to 277 elo difference and not 280 elo.

[Father]

It seems that fixing the bug change the table and now this is the table(I do not complain about reduction in the place of me that is logical but how it is possible that Catecan went down to 2291.9?)

I thought that even with dividing K by 2 considering the fact that he played so many games his rating should not go down by a lot because it already converged to his performance against the bot.
The only case when his rating should go down by a lot is the case that he improved significantly in the last games so the rating still did not converge to be close to the right rating.

Even I went down only by 30 elo or something like that and I could see 2246 or something close earlier after part of the update so I guess it was my rating before fixing the bug.

LeelaQueenOdds as white 1+0 (Reference) 2895.2
LeelaQueenOdds as black 1+0 (Reference) 2795.2
LeelaQueenOdds as white 1+1 (Reference) 2634.5
LeelaQueenOdds as white 3+0 (Reference) 2619.5
LeelaQueenOdds as black 1+1 (Reference) 2534.5
1 Orichess88 2524.5 73
LeelaQueenOdds as black 3+0 (Reference) 2519.5
2 Master_Unknown 2514.6 92
LeelaQueenOdds as white 3+2 (Reference) 2450.0
3 Ushkothebear 2443.7 184
4 Hissha 2398.4 94
5 Omnivor 2396.8 30
LeelaQueenOdds as white 5+3 (Reference) 2353.7
LeelaQueenOdds as black 3+2 (Reference) 2350.0
6 Respectful_Dave 2299.6 74
7 itore 2297.1 43
8 Catecan 2291.9 1288
LeelaQueenOdds as black 5+3 (Reference) 2253.7
9 Sakali 2251.1 146
10 Spreek 2229.8 45
11 sfy2011 2223.6 25
12 SilverMitt 2223.1 26
LeelaQueenOdds as white 10+5 (Reference) 2220.3
13 uriblass 2216.4 182
14 nn1133 2205.6 6346

Catecan went down because its recent performance is better than its older performance, all it has to do is maintain this performance and it will continue to rise.

Good afternoon Marcus91. I really need the help of mathematicians. If one day Mr. Larry Kaufman, or any of you, invites me to play chess computers, I will not bring a psychologist, a doctor, or a team of analysts. I will only require, humanly speaking, a mathematics teacher to help me. What has happened to my elo has been completely incomprehensible to me. I'm a real yo-yo that has me on the brink of collapse. I simply ask all of you to collaborate with me, I don't know what's happening, the more I play, the lower my rating goes, even though I'm feeling progress. Besides, I know that I will be able to win by the minute later, I just don't care at all at the moment. I'm simply doing the same thing that Kasparov did with Karpov before he defeated him and challenged him for the title of world champion. exhaust him, break him mentally. I am leading the algorithm to collapse. Marcus91. At this moment I have 1388 games. Please help me and tell me how much my elo is left... please: Today we have 100 new bullet games, we have reached 38 draws and 62 losses. Sincere thanks in advance.

If you are able to improve enough to get one draw out of every two games (so up from 38% draws to 50% draws), and keep that level for hundreds of games, your rating should approach 2515 under current rules, which would be a tie for second place right now. If draws were fully counted, it would approach (2795-191 = 2604), so counting only half the draws (technically half of each draw) reduces the expected rating by nearly 90 elo. But it also hurts the elo of all the other players who play at bullet time controls where they are usually only getting draws at best, so it's not just lowering your elo. But it does reward players for choosing time limits where they can actually win games, which is after all supposed to be the goal of a chess game. Since Leela doesn't mind making draws when down a queen, it seems reasonable to me. Note that all the players above you on the list do play at time limits where they win a lot of games.

I use the following table and get slightly different results(not that it is very important)

https://handbook.fide.com/chapter/B022024

I translate 75% to 193 elo difference and not 191 elo.
I translate 83.333% to 277 elo difference and not 280 elo.

[pgn][Event "Casual bullet game"]
[Site "https://lichess.org/w4WIftQu"]
[Date "2024.12.02"]
[White "Catecan"]
[Black "LeelaQueenOdds"]
[Result "1-0"]
[UTCDate "2024.12.02"]
[UTCTime "06:27:18"]
[WhiteElo "2030"]
[BlackElo "2000"]
[BlackTitle "BOT"]
[Variant "From Position"]
[TimeControl "60+0"]
[ECO "?"]
[Opening "?"]
[Termination "Normal"]
[FEN "rnb1kbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1"]
[SetUp "1"]
[Annotator "lichess.org"]

1. d4 Nf6 2. e3 g5 3. f4 gxf4 4. exf4 e6 5. Nf3 b6 6. c3 Bb7 7. Bd3 Rg8 8. Qe2 h5 9. Rg1 Nc6 10. Ba6 Bxa6 11. Qxa6 Nd5 12. Qe2 O-O-O 13. g3 f6 14. Rg2 h4 15. Nxh4 e5 16. fxe5 fxe5 17. dxe5 Re8 18. Nf3 Rg4 19. Bf4 Nxf4 20. gxf4 Rxf4 21. Nbd2 Ra4 22. a3 Rf4 23. O-O-O Kb7 24. Re1 Re7 25. Qe3 Rff7 26. Ree2 Bg7 27. Kb1 Bxe5 28. Nxe5 Rxe5 29. Qxe5 Nxe5 30. Rxe5 Rh7 31. Ree2 a5 32. Nf1 a4 33. Rg3 Rf7 34. Reg2 Rxf1+ 35. Ka2 Rh1 36. h3 Kc6 37. Rf2 b5 38. Rff3 d6 39. Rf4 Kb6 40. h4 c5 41. Rg5 Kc6 42. h5 Rh2 43. Rf6 Rh1 44. h6 Rh2 45. Rg7 Kd5 46. h7 b4 47. axb4 cxb4 48. cxb4 a3 49. Kxa3 Rh3+ 50. b3 Rh1 51. Rf8 Ra1+ 52. Kb2 Rh1 53. h8=Q Rxh8 54. Rxh8 Ke5 55. Rh6 d5 56. Rgg6 d4 57. Kc1 Ke4 58. Kd2 d3 59. Rg5 Kf4 60. Rhh5 Ke4 61. Re5+ Kf3 62. Rd5 Kf4 63. Rxd3 Kg4 64. Rh1 Kf4 65. Re1 Kg4 66. Ke2 Kf5 67. Kf2 Kf4 68. Rf3+ Kg5 69. Rg1+ Kh4 70. Kg2 Kg5 71. Rgf1 Kg6 72. Rg3+ Kh7 73. Rh1# { White wins by checkmate. } 1-0
1+0 • Amistosa • FROM POSITION
hace 3 minutos
Catecan (2030?)
BOT LeelaQueenOdds (2000?)
Jaque mate • Las blancas ganan
[/pgn]

Good evening everyone. I want to dedicate this game to the memory of my Father and my Mother. There is a saying or proverb that teaches us: "Consistency achieves what happiness cannot achieve." A drop of water, over time, shapes a rock. To my Programmer from this world I say: Thank you, this achievement is for you who made the algorithm of my soul, my body, and my mind. In the face of difficulties, the strength of the human being is "hope." We should not be afraid, just "hope." We are not alone. To Uri and Mr. Larry Kaufman, I also dedicate this award for perseverance.

[lkaufman]

It seems that fixing the bug change the table and now this is the table(I do not complain about reduction in the place of me that is logical but how it is possible that Catecan went down to 2291.9?)

I thought that even with dividing K by 2 considering the fact that he played so many games his rating should not go down by a lot because it already converged to his performance against the bot.
The only case when his rating should go down by a lot is the case that he improved significantly in the last games so the rating still did not converge to be close to the right rating.

Even I went down only by 30 elo or something like that and I could see 2246 or something close earlier after part of the update so I guess it was my rating before fixing the bug.

LeelaQueenOdds as white 1+0 (Reference) 2895.2
LeelaQueenOdds as black 1+0 (Reference) 2795.2
LeelaQueenOdds as white 1+1 (Reference) 2634.5
LeelaQueenOdds as white 3+0 (Reference) 2619.5
LeelaQueenOdds as black 1+1 (Reference) 2534.5
1 Orichess88 2524.5 73
LeelaQueenOdds as black 3+0 (Reference) 2519.5
2 Master_Unknown 2514.6 92
LeelaQueenOdds as white 3+2 (Reference) 2450.0
3 Ushkothebear 2443.7 184
4 Hissha 2398.4 94
5 Omnivor 2396.8 30
LeelaQueenOdds as white 5+3 (Reference) 2353.7
LeelaQueenOdds as black 3+2 (Reference) 2350.0
6 Respectful_Dave 2299.6 74
7 itore 2297.1 43
8 Catecan 2291.9 1288
LeelaQueenOdds as black 5+3 (Reference) 2253.7
9 Sakali 2251.1 146
10 Spreek 2229.8 45
11 sfy2011 2223.6 25
12 SilverMitt 2223.1 26
LeelaQueenOdds as white 10+5 (Reference) 2220.3
13 uriblass 2216.4 182
14 nn1133 2205.6 6346

Catecan went down because its recent performance is better than its older performance, all it has to do is maintain this performance and it will continue to rise.

Good afternoon Marcus91. I really need the help of mathematicians. If one day Mr. Larry Kaufman, or any of you, invites me to play chess computers, I will not bring a psychologist, a doctor, or a team of analysts. I will only require, humanly speaking, a mathematics teacher to help me. What has happened to my elo has been completely incomprehensible to me. I'm a real yo-yo that has me on the brink of collapse. I simply ask all of you to collaborate with me, I don't know what's happening, the more I play, the lower my rating goes, even though I'm feeling progress. Besides, I know that I will be able to win by the minute later, I just don't care at all at the moment. I'm simply doing the same thing that Kasparov did with Karpov before he defeated him and challenged him for the title of world champion. exhaust him, break him mentally. I am leading the algorithm to collapse. Marcus91. At this moment I have 1388 games. Please help me and tell me how much my elo is left... please: Today we have 100 new bullet games, we have reached 38 draws and 62 losses. Sincere thanks in advance.

If you are able to improve enough to get one draw out of every two games (so up from 38% draws to 50% draws), and keep that level for hundreds of games, your rating should approach 2515 under current rules, which would be a tie for second place right now. If draws were fully counted, it would approach (2795-191 = 2604), so counting only half the draws (technically half of each draw) reduces the expected rating by nearly 90 elo. But it also hurts the elo of all the other players who play at bullet time controls where they are usually only getting draws at best, so it's not just lowering your elo. But it does reward players for choosing time limits where they can actually win games, which is after all supposed to be the goal of a chess game. Since Leela doesn't mind making draws when down a queen, it seems reasonable to me. Note that all the players above you on the list do play at time limits where they win a lot of games.

I use the following table and get slightly different results(not that it is very important)

https://handbook.fide.com/chapter/B022024

I translate 75% to 193 elo difference and not 191 elo.
I translate 83.333% to 277 elo difference and not 280 elo.

I believe the small differences are because FIDE use the normal distribution whereas USchess and most of the servers and websites use the logistic distribution. Both versions were described by prof. elo in his book written around 1970 (I independently invented the logistic distribution version of the Elo system around the same time for rating local club tournaments!). As you say, there is not much difference, except when talking about very large rating differences (800 elo means 100 to 1 odds in the logistic model but way more than that in the normal dist. model). The Logistic distribution is much easier to compute, and also is probably more realistic for the data from what I've read, so FIDE should really switch over. Basically, the normal distribution model says that 800 elo upsets should almost never occur, but they do.

[Uri Blass]

It seems that fixing the bug change the table and now this is the table(I do not complain about reduction in the place of me that is logical but how it is possible that Catecan went down to 2291.9?)

I thought that even with dividing K by 2 considering the fact that he played so many games his rating should not go down by a lot because it already converged to his performance against the bot.
The only case when his rating should go down by a lot is the case that he improved significantly in the last games so the rating still did not converge to be close to the right rating.

Even I went down only by 30 elo or something like that and I could see 2246 or something close earlier after part of the update so I guess it was my rating before fixing the bug.

LeelaQueenOdds as white 1+0 (Reference) 2895.2
LeelaQueenOdds as black 1+0 (Reference) 2795.2
LeelaQueenOdds as white 1+1 (Reference) 2634.5
LeelaQueenOdds as white 3+0 (Reference) 2619.5
LeelaQueenOdds as black 1+1 (Reference) 2534.5
1 Orichess88 2524.5 73
LeelaQueenOdds as black 3+0 (Reference) 2519.5
2 Master_Unknown 2514.6 92
LeelaQueenOdds as white 3+2 (Reference) 2450.0
3 Ushkothebear 2443.7 184
4 Hissha 2398.4 94
5 Omnivor 2396.8 30
LeelaQueenOdds as white 5+3 (Reference) 2353.7
LeelaQueenOdds as black 3+2 (Reference) 2350.0
6 Respectful_Dave 2299.6 74
7 itore 2297.1 43
8 Catecan 2291.9 1288
LeelaQueenOdds as black 5+3 (Reference) 2253.7
9 Sakali 2251.1 146
10 Spreek 2229.8 45
11 sfy2011 2223.6 25
12 SilverMitt 2223.1 26
LeelaQueenOdds as white 10+5 (Reference) 2220.3
13 uriblass 2216.4 182
14 nn1133 2205.6 6346

Catecan went down because its recent performance is better than its older performance, all it has to do is maintain this performance and it will continue to rise.

Good afternoon Marcus91. I really need the help of mathematicians. If one day Mr. Larry Kaufman, or any of you, invites me to play chess computers, I will not bring a psychologist, a doctor, or a team of analysts. I will only require, humanly speaking, a mathematics teacher to help me. What has happened to my elo has been completely incomprehensible to me. I'm a real yo-yo that has me on the brink of collapse. I simply ask all of you to collaborate with me, I don't know what's happening, the more I play, the lower my rating goes, even though I'm feeling progress. Besides, I know that I will be able to win by the minute later, I just don't care at all at the moment. I'm simply doing the same thing that Kasparov did with Karpov before he defeated him and challenged him for the title of world champion. exhaust him, break him mentally. I am leading the algorithm to collapse. Marcus91. At this moment I have 1388 games. Please help me and tell me how much my elo is left... please: Today we have 100 new bullet games, we have reached 38 draws and 62 losses. Sincere thanks in advance.

If you are able to improve enough to get one draw out of every two games (so up from 38% draws to 50% draws), and keep that level for hundreds of games, your rating should approach 2515 under current rules, which would be a tie for second place right now. If draws were fully counted, it would approach (2795-191 = 2604), so counting only half the draws (technically half of each draw) reduces the expected rating by nearly 90 elo. But it also hurts the elo of all the other players who play at bullet time controls where they are usually only getting draws at best, so it's not just lowering your elo. But it does reward players for choosing time limits where they can actually win games, which is after all supposed to be the goal of a chess game. Since Leela doesn't mind making draws when down a queen, it seems reasonable to me. Note that all the players above you on the list do play at time limits where they win a lot of games.

I use the following table and get slightly different results(not that it is very important)

https://handbook.fide.com/chapter/B022024

I translate 75% to 193 elo difference and not 191 elo.
I translate 83.333% to 277 elo difference and not 280 elo.

[pgn][Event "Casual bullet game"]
[Site "https://lichess.org/w4WIftQu"]
[Date "2024.12.02"]
[White "Catecan"]
[Black "LeelaQueenOdds"]
[Result "1-0"]
[UTCDate "2024.12.02"]
[UTCTime "06:27:18"]
[WhiteElo "2030"]
[BlackElo "2000"]
[BlackTitle "BOT"]
[Variant "From Position"]
[TimeControl "60+0"]
[ECO "?"]
[Opening "?"]
[Termination "Normal"]
[FEN "rnb1kbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1"]
[SetUp "1"]
[Annotator "lichess.org"]

1. d4 Nf6 2. e3 g5 3. f4 gxf4 4. exf4 e6 5. Nf3 b6 6. c3 Bb7 7. Bd3 Rg8 8. Qe2 h5 9. Rg1 Nc6 10. Ba6 Bxa6 11. Qxa6 Nd5 12. Qe2 O-O-O 13. g3 f6 14. Rg2 h4 15. Nxh4 e5 16. fxe5 fxe5 17. dxe5 Re8 18. Nf3 Rg4 19. Bf4 Nxf4 20. gxf4 Rxf4 21. Nbd2 Ra4 22. a3 Rf4 23. O-O-O Kb7 24. Re1 Re7 25. Qe3 Rff7 26. Ree2 Bg7 27. Kb1 Bxe5 28. Nxe5 Rxe5 29. Qxe5 Nxe5 30. Rxe5 Rh7 31. Ree2 a5 32. Nf1 a4 33. Rg3 Rf7 34. Reg2 Rxf1+ 35. Ka2 Rh1 36. h3 Kc6 37. Rf2 b5 38. Rff3 d6 39. Rf4 Kb6 40. h4 c5 41. Rg5 Kc6 42. h5 Rh2 43. Rf6 Rh1 44. h6 Rh2 45. Rg7 Kd5 46. h7 b4 47. axb4 cxb4 48. cxb4 a3 49. Kxa3 Rh3+ 50. b3 Rh1 51. Rf8 Ra1+ 52. Kb2 Rh1 53. h8=Q Rxh8 54. Rxh8 Ke5 55. Rh6 d5 56. Rgg6 d4 57. Kc1 Ke4 58. Kd2 d3 59. Rg5 Kf4 60. Rhh5 Ke4 61. Re5+ Kf3 62. Rd5 Kf4 63. Rxd3 Kg4 64. Rh1 Kf4 65. Re1 Kg4 66. Ke2 Kf5 67. Kf2 Kf4 68. Rf3+ Kg5 69. Rg1+ Kh4 70. Kg2 Kg5 71. Rgf1 Kg6 72. Rg3+ Kh7 73. Rh1# { White wins by checkmate. } 1-0
1+0 • Amistosa • FROM POSITION
hace 3 minutos
Catecan (2030?)
BOT LeelaQueenOdds (2000?)
Jaque mate • Las blancas ganan
[/pgn]

Good evening everyone. I want to dedicate this game to the memory of my Father and my Mother. There is a saying or proverb that teaches us: "Consistency achieves what happiness cannot achieve." A drop of water, over time, shapes a rock. To my Programmer from this world I say: Thank you, this achievement is for you who made the algorithm of my soul, my body, and my mind. In the face of difficulties, the strength of the human being is "hope." We should not be afraid, just "hope." We are not alone. To Uri and Mr. Larry Kaufman, I also dedicate this award for perseverance.

a good game of you but Leela did not play well.
27...Bxe5 is a bad move.