Let's play chess

Questions on how we spend our money and our time - consumer goods and services, home and vehicle, leisure and recreational activities
Post Reply
User avatar
canadianbacon
Posts: 677
Joined: Sun Nov 10, 2019 9:04 pm

Re: Let's play chess

Post by canadianbacon »

Marseille07 wrote: Mon Nov 01, 2021 10:46 am 1500 being the baseline is also a valid observation; I noticed some players having 3100+, that's higher than Magnus Carlsen.
Magnus plays on Lichess as DrNykterstein; he has a bullet rating of 3191 and Blitz of 3160.
Bulls make money, bears make money, pigs get slaughtered.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Let's play chess

Post by Marseille07 »

canadianbacon wrote: Mon Nov 01, 2021 12:09 pm
Marseille07 wrote: Mon Nov 01, 2021 10:46 am 1500 being the baseline is also a valid observation; I noticed some players having 3100+, that's higher than Magnus Carlsen.
Magnus plays on Lichess as DrNykterstein; he has a bullet rating of 3191 and Blitz of 3160.
Oh I see, I've seen the handle but didn't know that's him.
User avatar
canadianbacon
Posts: 677
Joined: Sun Nov 10, 2019 9:04 pm

Re: Let's play chess

Post by canadianbacon »

willthrill81 wrote: Mon Nov 01, 2021 9:48 am I'm certainly not a great player. My bullet rating on Lichess is typically around 1900.
I can barely manage 1300 at bullet, so it's all relative :). I need to play/study more so I can power through most openings and any endgames if I'm going to play 2+1. It makes more sense for me to play time controls where I can gain understanding.
Bulls make money, bears make money, pigs get slaughtered.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Let's play chess

Post by Marseille07 »

willthrill81 wrote: Mon Nov 01, 2021 9:48 am I'm certainly not a great player. My bullet rating on Lichess is typically around 1900. It is true though that playing faster games exposes you to many positions and helps you to quickly recognize patterns, which is much of what playing chess is really all about.
Not sure...I've seen lots of gimmicky openings that don't work if a player knows how to defend them, but work because they're under time pressure. Personally I stay away from bullet.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Let's play chess

Post by willthrill81 »

Marseille07 wrote: Mon Nov 01, 2021 4:18 pm
willthrill81 wrote: Mon Nov 01, 2021 9:48 am I'm certainly not a great player. My bullet rating on Lichess is typically around 1900. It is true though that playing faster games exposes you to many positions and helps you to quickly recognize patterns, which is much of what playing chess is really all about.
Not sure...I've seen lots of gimmicky openings that don't work if a player knows how to defend them, but work because they're under time pressure. Personally I stay away from bullet.
That's just part of that style of play. It's not enough to play the best moves; you have to avoid losing before the clock runs out.
The Sensible Steward
halfnine
Posts: 2421
Joined: Tue Dec 21, 2010 12:48 pm

Re: Let's play chess

Post by halfnine »

Marseille07 wrote: Mon Nov 01, 2021 10:49 am
canadianbacon wrote: Mon Nov 01, 2021 5:08 am Your rating will jump around more at the start because there is less data so your actual ability level is not known (your rating is provisional). Your rating deviation after three games is 247.31. If you lose your next game you'll lose a lot of points. The more games you play, the less a particular result changes your rating. Think of it sort of like a binary search, it's "finding" your true talent level more quickly this way.
Yeah, what was interesting though is that the opponent didn't lose 250 points or anything like that. When I was playing Yahoo Chess back in the day, the rating change was more or less symmetrical (you gain 10, the opponent loses 10 points). I am not too familiar with the Glicko 2 system.
There is a calculator for rating change for the ELO system here. The most points that one can gain or lose during a game is essentially their K factor. Most people have a K factor of 20. They have changed the K factors over the year to try to address the shortcomings of the ELO system as it pertains to establishing the rating for new players and to juniors who progress rapidly. Those players have a K factor of 40. These factors are for Classical time control. I believe Blitz and Rapid everyone gets a K factor of 20. I think this is simply because people can play a lot of shorter time control games in a relatively quick amount of time so that their current playing ability quite quickly reaches their actual ratings.

So, like you mentioned, if two players with identical K factors play each other the same amount of points are an offer for each player. If one player gains 10 points the other will lose 10 points. One can crudely calculate how many points are on offer in their head. Half of the points are based on whether you win, draw, or lose. So for a K factor of 20 that would respectively be +10, 0, -10. The other 10 points are based on the rating difference between the two players. The rating difference is capped at 400. So for rating differences (-400,0,400) one would receive points of (-10,0,10). Essentially, one gains or loses 2.5 points for every 100 point rating difference between them are their opponent.

So a case in point if you draw against someone rated 200 points higher than you, you will get +0 points for the draw and +5 points for the difference in rating. This is why often when people play against higher rated players they will play lines that lead toward a draw. The higher rated player, not wanting to lose points on a draw, will then likely take a few chances or try to imbalance the position to try to prevent this.

Anyway, that is the cliff notes version. And that is the extent of my knowleged as I actually don't know the long version.
Northern Flicker
Posts: 15371
Joined: Fri Apr 10, 2015 12:29 am

Re: Let's play chess

Post by Northern Flicker »

Glicko 2 was originally developed by Mark Glickman as his doctoral dissertation in statistics. I am not very familiar with the details, but I believe it may be Bayesian in nature. One's current rating before a game and how long it has been since a rated game was played are inputs to the prior distribution for the outcome of the game, and the rating after the game is a property of the posterior distribution after the game outcome is known.

As a result, ratings are not a zero-sum game in Glicko 2 like they are in Arpad Elo's system. If player A has not played a rated game in a long time, then a rated game will have more influence on the rating for such a player than for a player B who has played many recent rated games. For player B, the rating at the start of the game provides a stronger prior opinion, so the influence of the one game on player B's rating is less than for player A, whose rating does not reflect recent performance.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Let's play chess

Post by Marseille07 »

halfnine wrote: Wed Nov 03, 2021 1:25 am
Marseille07 wrote: Mon Nov 01, 2021 10:49 am
canadianbacon wrote: Mon Nov 01, 2021 5:08 am Your rating will jump around more at the start because there is less data so your actual ability level is not known (your rating is provisional). Your rating deviation after three games is 247.31. If you lose your next game you'll lose a lot of points. The more games you play, the less a particular result changes your rating. Think of it sort of like a binary search, it's "finding" your true talent level more quickly this way.
Yeah, what was interesting though is that the opponent didn't lose 250 points or anything like that. When I was playing Yahoo Chess back in the day, the rating change was more or less symmetrical (you gain 10, the opponent loses 10 points). I am not too familiar with the Glicko 2 system.
There is a calculator for rating change for the ELO system here. The most points that one can gain or lose during a game is essentially their K factor. Most people have a K factor of 20. They have changed the K factors over the year to try to address the shortcomings of the ELO system as it pertains to establishing the rating for new players and to juniors who progress rapidly. Those players have a K factor of 40. These factors are for Classical time control. I believe Blitz and Rapid everyone gets a K factor of 20. I think this is simply because people can play a lot of shorter time control games in a relatively quick amount of time so that their current playing ability quite quickly reaches their actual ratings.

So, like you mentioned, if two players with identical K factors play each other the same amount of points are an offer for each player. If one player gains 10 points the other will lose 10 points. One can crudely calculate how many points are on offer in their head. Half of the points are based on whether you win, draw, or lose. So for a K factor of 20 that would respectively be +10, 0, -10. The other 10 points are based on the rating difference between the two players. The rating difference is capped at 400. So for rating differences (-400,0,400) one would receive points of (-10,0,10). Essentially, one gains or loses 2.5 points for every 100 point rating difference between them are their opponent.

So a case in point if you draw against someone rated 200 points higher than you, you will get +0 points for the draw and +5 points for the difference in rating. This is why often when people play against higher rated players they will play lines that lead toward a draw. The higher rated player, not wanting to lose points on a draw, will then likely take a few chances or try to imbalance the position to try to prevent this.

Anyway, that is the cliff notes version. And that is the extent of my knowleged as I actually don't know the long version.
Thank you, this is excellent :beer I remember Jeff Sonas analyzed ELO in depth and discovered that the ratings have inflated over time. This was a while ago and his site hasn't been updated for quite some time: http://www.chessmetrics.com
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Let's play chess

Post by Marseille07 »

People who play like this are really tiring. Why don't they play normal openings? I had to pull some Petrosian here and the guy left: https://lichess.org/qhMn9z7G/black
Northern Flicker
Posts: 15371
Joined: Fri Apr 10, 2015 12:29 am

Re: Let's play chess

Post by Northern Flicker »

16. Qh3 instead of 16. Qd2?! and white seems to doing ok there. Black's dark squares on the k-side are weak, and white has the only dark-squared bishop still on the board. This should give white an edge, though Black's extra center pawn is a source of counterplay, and f2 is a target.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Let's play chess

Post by Marseille07 »

Northern Flicker wrote: Sun Nov 07, 2021 12:36 am 16. Qh3 instead of 16. Qd2?! and white seems to doing ok there. Black's dark squares on the k-side are weak, and white has the only dark-squared bishop still on the board. This should give white an edge, though Black's extra center pawn is a source of counterplay, and f2 is a target.
Whoa, you must be really good :beer I just found out today that we can directly link games like this, which is very nice imo.
User avatar
Topic Author
LadyGeek
Site Admin
Posts: 95704
Joined: Sat Dec 20, 2008 4:34 pm
Location: Philadelphia
Contact:

Re: Let's play chess

Post by LadyGeek »

Yes, you can. Interested readers can review, modify, or set game preferences by clicking on the various moves and menu items in the right sidebar.

The target icon (2nd from left at the bottom of the sidebar) will start a new game.

Caution: Only link anonymous games. Otherwise, your lichess username will be connected with your forum username and your anonymity will be lost.
Wiki To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
investingdad
Posts: 2140
Joined: Fri Mar 15, 2013 10:41 pm

Re: Let's play chess

Post by investingdad »

I can’t get past the 1980 rating hump on lichess playing 3 minute games. That’s with nearly 10,000 games played.

Ah well.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Let's play chess

Post by Marseille07 »

Any training material online / offline? I have an old version of CT-Art, but I don't use Windows these days and I don't bother enough to fire up Wine to run it even if it is runnable on linux.

I found LucasChess which has linux distributions...but their training material wasn't easy to use, or maybe I don't know how to use it.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Let's play chess

Post by Marseille07 »

Turned out we can download PGN even for anonymous games: https://lichess.org/LArW9J0q/white

At the bottom, it says FEN & PGN; that's where you can download it. I just had a good rundown checkmate after the opponent blundered :beer
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Let's play chess

Post by willthrill81 »

investingdad wrote: Sun Nov 07, 2021 6:59 am I can’t get past the 1980 rating hump on lichess playing 3 minute games. That’s with nearly 10,000 games played.

Ah well.
That's good. My 2/1 rating go up to nearly 2000 and then plummeted to the low 1800s, and that's with thousands of games played.
The Sensible Steward
halfnine
Posts: 2421
Joined: Tue Dec 21, 2010 12:48 pm

Re: Let's play chess

Post by halfnine »

investingdad wrote: Sun Nov 07, 2021 6:59 am I can’t get past the 1980 rating hump on lichess playing 3 minute games. That’s with nearly 10,000 games played.

Ah well.
From what I saw watching my kid play the 2000 barrier on Lichess is in general a tough barrier to cross. When people are at or around 2000 they appear to selectively play rated games only if they feel they are currently performing well. If not they will either play non-rated games, play at different time controls, or even play on other chess sites often for weeks or months.
cheesepep
Posts: 1295
Joined: Wed Feb 17, 2010 9:58 pm

Re: Let's play chess

Post by cheesepep »

For anyone who wants to play Chinese chess or wants to learn, I highly recommend xiangqi.com. You can customize the board to use western style pieces in the form of discs. The rules are kind of similar also.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Let's play chess

Post by Marseille07 »

Any predictions for the World Chess Championship? I am rooting for Nepomniachtchi to unseat Carlsen.
User avatar
canadianbacon
Posts: 677
Joined: Sun Nov 10, 2019 9:04 pm

Re: Let's play chess

Post by canadianbacon »

Marseille07 wrote: Sat Nov 13, 2021 10:34 pm Any predictions for the World Chess Championship? I am rooting for Nepomniachtchi to unseat Carlsen.
It seems like Alireza has the best chance to be the next champion, and it will take him a few years to be ready.
Bulls make money, bears make money, pigs get slaughtered.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Let's play chess

Post by Marseille07 »

canadianbacon wrote: Sun Nov 14, 2021 2:54 pm
Marseille07 wrote: Sat Nov 13, 2021 10:34 pm Any predictions for the World Chess Championship? I am rooting for Nepomniachtchi to unseat Carlsen.
It seems like Alireza has the best chance to be the next champion, and it will take him a few years to be ready.
I've heard of his name but I wasn't aware how rapidly he's progressing.

Players come and go though. Not long ago people talked about Ding, Shankland and Vachier-Lagrave, but I don't think they're doing too well lately.

Hikaru Nakamura was also very promising but he turned into a YouTuber. I guess not everyone wants to be the world champion.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Let's play chess

Post by Marseille07 »

Lichess is kind of weird. Even though it says 20000 games are in progress, when I play pick up games I notice I'm playing the same opponent over and over (from peculiar opening lines they play). The odds of this should be small but I notice this from time to time.
halfnine
Posts: 2421
Joined: Tue Dec 21, 2010 12:48 pm

Re: Let's play chess

Post by halfnine »

Marseille07 wrote: Sun Nov 14, 2021 9:42 pm Lichess is kind of weird. Even though it says 20000 games are in progress, when I play pick up games I notice I'm playing the same opponent over and over (from peculiar opening lines they play). The odds of this should be small but I notice this from time to time.
Not necessarily.

First of all there a lot of people that are playing in tournaments. Lichess has Arena tournaments that run throughout the day as you can see from the Play drop down menu. But these are only tournaments run by Lichess and there are plenty of other online tournaments held by individuals, groups or chess clubs. For instance, recently my children played as part of a team in a tournament which had 30 teams and over 400 players. And the team they are associated with joins these types of tournaments almost every day.

And, then because of the various time controls, an individual's settings, casual/rated, and variants (although less so) there are probably a lot fewer people than you think that are playing exactly what you want to play within your rating range. Obviously quick pairing helps as it tends to at least greatly diminish the time controls the majority of people are choosing. Now, if you go the Lobby you will see the game requests that are available to you as they pop up (I assume these include quick pairings). I prefer to look at it with "graph" mode which can be selected by clicking the graph symbol on the top left of the table. If you play at the same time everyday and your rating is relatively stable you will likely see the same user names popping up from time to time.

When my kids were less than 1700 they would actually hang out in the Lobby and select opponents from there instead of using quick pairing or setting out their own custom game. This is simply because otherwise they got a lot of "?" in terms of ratings. And those games would largely be a waste of time. In the Lobby they can see the player and their details before committing to a game. Nowadays, they send out a custom request simply because they can select a rating range of their opponent. Typically they will choose -50 to +150 as that gives them the best games.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Let's play chess

Post by Marseille07 »

halfnine wrote: Sun Nov 14, 2021 11:01 pm
Marseille07 wrote: Sun Nov 14, 2021 9:42 pm Lichess is kind of weird. Even though it says 20000 games are in progress, when I play pick up games I notice I'm playing the same opponent over and over (from peculiar opening lines they play). The odds of this should be small but I notice this from time to time.
Not necessarily.

First of all there a lot of people that are playing in tournaments. Lichess has Arena tournaments that run throughout the day as you can see from the Play drop down menu. But these are only tournaments run by Lichess and there are plenty of other online tournaments held by individuals, groups or chess clubs. For instance, recently my children played as part of a team in a tournament which had 30 teams and over 400 players. And the team they are associated with joins these types of tournaments almost every day.

And, then because of the various time controls, an individual's settings, casual/rated, and variants (although less so) there are probably a lot fewer people than you think that are playing exactly what you want to play within your rating range. Obviously quick pairing helps as it tends to at least greatly diminish the time controls the majority of people are choosing. Now, if you go the Lobby you will see the game requests that are available to you as they pop up (I assume these include quick pairings). I prefer to look at it with "graph" mode which can be selected by clicking the graph symbol on the top left of the table. If you play at the same time everyday and your rating is relatively stable you will likely see the same user names popping up from time to time.

When my kids were less than 1700 they would actually hang out in the Lobby and select opponents from there instead of using quick pairing or setting out their own custom game. This is simply because otherwise they got a lot of "?" in terms of ratings. And those games would largely be a waste of time. In the Lobby they can see the player and their details before committing to a game. Nowadays, they send out a custom request simply because they can select a rating range of their opponent. Typically they will choose -50 to +150 as that gives them the best games.
Holy cow, I wasn't aware all of those tournament games are going on throughout the day, and that people outside of lichess can host them at lichess.

I know I should be playing more rated games but I'm still picking up anonymous games as I'm making way too many blunders :D
halfnine
Posts: 2421
Joined: Tue Dec 21, 2010 12:48 pm

Re: Let's play chess

Post by halfnine »

Marseille07 wrote: Sat Nov 06, 2021 10:35 pm People who play like this are really tiring. Why don't they play normal openings? I had to pull some Petrosian here and the guy left: https://lichess.org/qhMn9z7G/black
I can't speak specifically to what your opponent played but I can speculate as to why you may not see normal openings. And there are probably two main reasons. The first is simply the shorter the time control the more gambits and odd openings you will see simply because they work. While it is obvious that the moves are inaccurate it is impossible to figure out why in a short period of time. The second reason kind of ties into the tournaments that I mentioned above. Lichess largely has Arena style tournaments. In Arena tournaments you don't play a set number of rounds. In Arena tournaments you play for a duration (often about 90 minutes) and the winner is simply the one who can win the most games in that time frame. The incentive then is to forget about one's rating and either simply win quickly or lose quickly. Winning quickly is typically done with dubious gambits, ultra-aggressive play, or pet openings where you know the thematics and your opponent doesn't. And then if the opening doesn't pan out try a few aggressive ideas and if that doesn't work simply resign. And, although you are not playing these opponents in a tournament setting, the mentality often remains.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Let's play chess

Post by Marseille07 »

halfnine wrote: Sun Nov 14, 2021 11:31 pm
Marseille07 wrote: Sat Nov 06, 2021 10:35 pm People who play like this are really tiring. Why don't they play normal openings? I had to pull some Petrosian here and the guy left: https://lichess.org/qhMn9z7G/black
I can't speak specifically to what your opponent played but I can speculate as to why you may not see normal openings. And there are probably two main reasons. The first is simply the shorter the time control the more gambits and odd openings you will see simply because they work. While it is obvious that the moves are inaccurate it is impossible to figure out why in a short period of time. The second reason kind of ties into the tournaments that I mentioned above. Lichess largely has Arena style tournaments. In Arena tournaments you don't play a set number of rounds. In Arena tournaments you play for a duration (often about 90 minutes) and the winner is simply the one who can win the most games in that time frame. The incentive then is to forget about one's rating and either simply win quickly or lose quickly. Winning quickly is typically done with dubious gambits, ultra-aggressive play, or pet openings where you know the thematics and your opponent doesn't. And then if the opening doesn't pan out try a few aggressive ideas and if that doesn't work simply resign. And, although you are not playing these opponents in a tournament setting, the mentality often remains.
A very insightful post :beer

What you're describing makes sense, since chess is a game where chess engines beat humans; this might be taking away some aspect of studying the game diligently, and Arena tournaments and the like might cater those who just want to have fun winning regardless of quality games.

I sometimes tune into those 1+0 games or even 1/2+0 games, I guess those pros are good but not so sure the quality of the games or why they play with such a short time control. It's almost as if they're playing a video game.
Northern Flicker
Posts: 15371
Joined: Fri Apr 10, 2015 12:29 am

Re: Let's play chess

Post by Northern Flicker »

At the GM level, 2 and 3 minute games are surprisingly high quality. Sure, there are inaccuracies relative to GMs playing at slow time controls, but they are more accurate at that speed than most amateurs are at slow time controls.
halfnine
Posts: 2421
Joined: Tue Dec 21, 2010 12:48 pm

Re: Let's play chess

Post by halfnine »

Marseille07 wrote: Sun Nov 14, 2021 11:44 pm
halfnine wrote: Sun Nov 14, 2021 11:31 pm
Marseille07 wrote: Sat Nov 06, 2021 10:35 pm People who play like this are really tiring. Why don't they play normal openings? I had to pull some Petrosian here and the guy left: https://lichess.org/qhMn9z7G/black
I can't speak specifically to what your opponent played but I can speculate as to why you may not see normal openings. And there are probably two main reasons. The first is simply the shorter the time control the more gambits and odd openings you will see simply because they work. While it is obvious that the moves are inaccurate it is impossible to figure out why in a short period of time. The second reason kind of ties into the tournaments that I mentioned above. Lichess largely has Arena style tournaments. In Arena tournaments you don't play a set number of rounds. In Arena tournaments you play for a duration (often about 90 minutes) and the winner is simply the one who can win the most games in that time frame. The incentive then is to forget about one's rating and either simply win quickly or lose quickly. Winning quickly is typically done with dubious gambits, ultra-aggressive play, or pet openings where you know the thematics and your opponent doesn't. And then if the opening doesn't pan out try a few aggressive ideas and if that doesn't work simply resign. And, although you are not playing these opponents in a tournament setting, the mentality often remains.
A very insightful post :beer

What you're describing makes sense, since chess is a game where chess engines beat humans; this might be taking away some aspect of studying the game diligently, and Arena tournaments and the like might cater those who just want to have fun winning regardless of quality games.

I sometimes tune into those 1+0 games or even 1/2+0 games, I guess those pros are good but not so sure the quality of the games or why they play with such a short time control. It's almost as if they're playing a video game.
Another thing to be mindful of is not to take a Blitz rating of a random opponent to seriously as it might as much as 200 plus/minus points off. For instance, there is a kid who we frequently encounter around here. If you play him in blitz in an Arena tournament he will perform at around an 1800 level but if you play him outside the Arena format he will perform at around a 2000 level. His actual rating will fluctuate between those values depending on how frequent his Arena tournament playing is at the time. So, for instance, you could play him outside of an Arena tournament and see the 1800 ranking only to find yourself playing against someone who is really performing at a 2000 level. Alternatively, you could encounter him in an Arena tournament with a rating near 2000 only to find him performing around an 1800 level during the tournament.

HIs strategy is actually fairly effective for his goals as he finishes quite high in the Arena tournaments he plays soley based on the number of wins he accumulates. HIs blitz rating is also quite negatively impacted by the fact that in Arena tournaments he often (80%) berserks. This is when you play with half the time of your opponent (and no increment) but in return get an extra point if you win.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Let's play chess

Post by Marseille07 »

halfnine wrote: Mon Nov 15, 2021 12:51 am Another thing to be mindful of is not to take a Blitz rating of a random opponent to seriously as it might as much as 200 plus/minus points off. For instance, there is a kid who we frequently encounter around here. If you play him in blitz in an Arena tournament he will perform at around an 1800 level but if you play him outside the Arena format he will perform at around a 2000 level. His actual rating will fluctuate between those values depending on how frequent his Arena tournament playing is at the time. So, for instance, you could play him outside of an Arena tournament and see the 1800 ranking only to find yourself playing against someone who is really performing at a 2000 level. Alternatively, you could encounter him in an Arena tournament with a rating near 2000 only to find him performing around an 1800 level during the tournament.

HIs strategy is actually fairly effective for his goals as he finishes quite high in the Arena tournaments he plays soley based on the number of wins he accumulates. HIs blitz rating is also quite negatively impacted by the fact that in Arena tournaments he often (80%) berserks. This is when you play with half the time of your opponent (and no increment) but in return get an extra point if you win.
That's interesting. The rating at lichess itself is also dubious as well; I wish they used more standardish ELO system like Yahoo Chess used to. The interface is much nicer though, we couldn't play those split-second moves on their Flash-based interface. The technology has improved for sure.
halfnine
Posts: 2421
Joined: Tue Dec 21, 2010 12:48 pm

Re: Let's play chess

Post by halfnine »

Marseille07 wrote: Mon Nov 15, 2021 12:59 am
halfnine wrote: Mon Nov 15, 2021 12:51 am Another thing to be mindful of is not to take a Blitz rating of a random opponent to seriously as it might as much as 200 plus/minus points off. For instance, there is a kid who we frequently encounter around here. If you play him in blitz in an Arena tournament he will perform at around an 1800 level but if you play him outside the Arena format he will perform at around a 2000 level. His actual rating will fluctuate between those values depending on how frequent his Arena tournament playing is at the time. So, for instance, you could play him outside of an Arena tournament and see the 1800 ranking only to find yourself playing against someone who is really performing at a 2000 level. Alternatively, you could encounter him in an Arena tournament with a rating near 2000 only to find him performing around an 1800 level during the tournament.

HIs strategy is actually fairly effective for his goals as he finishes quite high in the Arena tournaments he plays soley based on the number of wins he accumulates. HIs blitz rating is also quite negatively impacted by the fact that in Arena tournaments he often (80%) berserks. This is when you play with half the time of your opponent (and no increment) but in return get an extra point if you win.
That's interesting. The rating at lichess itself is also dubious as well; I wish they used more standardish ELO system like Yahoo Chess used to. The interface is much nicer though, we couldn't play those split-second moves on their Flash-based interface. The technology has improved for sure.
I take a bit of different approach. Currently, I find the Lichess Rapid rating the most up to date reflection of someone’s current performance. And then this number can quite easily be correlated to a FIDE rating either in one’s head if one does it frequently enough or based on the chart that I referenced up thread. At the current time, most other ratings are currently outdated as OTB was largely non-existent for the last 18 months. And blitz ratings are subject to many flaws that have also been indicated up thread. Of course, as time progresses ELO ratings may eventually catch up, but for now Lichess Rapid (at least to me) seems to be the best indicator of how one’s opponent stacks up.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Let's play chess

Post by Marseille07 »

halfnine wrote: Mon Nov 15, 2021 1:27 am I take a bit of different approach. Currently, I find the Lichess Rapid rating the most up to date reflection of someone’s current performance. And then this number can quite easily be correlated to a FIDE rating either in one’s head if one does it frequently enough or based on the chart that I referenced up thread. At the current time, most other ratings are currently outdated as OTB was largely non-existent for the last 18 months. And blitz ratings are subject to many flaws that have also been indicated up thread. Of course, as time progresses ELO ratings may eventually catch up, but for now Lichess Rapid (at least to me) seems to be the best indicator of how one’s opponent stacks up.
That's a good point. Lichess provides different ratings for different time controls, and I agree Rapid or longer would provide more trustworthy ratings.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Let's play chess

Post by Marseille07 »

Strongest chess engines over time: https://www.youtube.com/watch?v=wljgxS7tZVE

I remember the days when Fritz, Shredder, HIARCS and Junior were competing for the top spot. Stockfish seems like the strongest today, and what's incredible is that it is open sourced.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Let's play chess

Post by Marseille07 »

Did lichess just add the dark mode today? Not that I had looked for this option but it's on the top page today.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Let's play chess

Post by Marseille07 »

World Chess Championship starting on Wednesday and Game 1 on Friday. I think Nepo has a shot, as he's really strong in classical TC.
halfnine
Posts: 2421
Joined: Tue Dec 21, 2010 12:48 pm

Re: Let's play chess

Post by halfnine »

Marseille07 wrote: Mon Nov 22, 2021 1:58 pm World Chess Championship starting on Wednesday and Game 1 on Friday. I think Nepo has a shot, as he's really strong in classical TC.
This will be the first World Chess Championship I'll be following in real time. It will be interesting to see how it transpires. Kasparov only thinks Nepo has a chance if Magnus isn't playing well. He also thinks the Magnus is currently not playing very well.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Let's play chess

Post by Marseille07 »

halfnine wrote: Tue Nov 23, 2021 10:24 am
Marseille07 wrote: Mon Nov 22, 2021 1:58 pm World Chess Championship starting on Wednesday and Game 1 on Friday. I think Nepo has a shot, as he's really strong in classical TC.
This will be the first World Chess Championship I'll be following in real time. It will be interesting to see how it transpires. Kasparov only thinks Nepo has a chance if Magnus isn't playing well. He also thinks the Magnus is currently not playing very well.
Yeah I'm rooting for Nepo. Nepo is 4W 8D 1L vs Carlsen in classical TC, he has a decent shot imo so long as the series won't go to tiebreakers where Magnus is just unbeatable.
halfnine
Posts: 2421
Joined: Tue Dec 21, 2010 12:48 pm

Re: Let's play chess

Post by halfnine »

Marseille07 wrote: Tue Nov 23, 2021 10:31 am
halfnine wrote: Tue Nov 23, 2021 10:24 am
Marseille07 wrote: Mon Nov 22, 2021 1:58 pm World Chess Championship starting on Wednesday and Game 1 on Friday. I think Nepo has a shot, as he's really strong in classical TC.
This will be the first World Chess Championship I'll be following in real time. It will be interesting to see how it transpires. Kasparov only thinks Nepo has a chance if Magnus isn't playing well. He also thinks the Magnus is currently not playing very well.
Yeah I'm rooting for Nepo. Nepo is 4W 8D 1L vs Carlsen in classical TC, he has a decent shot imo so long as the series won't go to tiebreakers where Magnus is just unbeatable.
Vishy and Kramnik have offered their thoughts as well, here
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Let's play chess

Post by Marseille07 »

halfnine wrote: Thu Nov 25, 2021 3:32 am
Marseille07 wrote: Tue Nov 23, 2021 10:31 am
halfnine wrote: Tue Nov 23, 2021 10:24 am
Marseille07 wrote: Mon Nov 22, 2021 1:58 pm World Chess Championship starting on Wednesday and Game 1 on Friday. I think Nepo has a shot, as he's really strong in classical TC.
This will be the first World Chess Championship I'll be following in real time. It will be interesting to see how it transpires. Kasparov only thinks Nepo has a chance if Magnus isn't playing well. He also thinks the Magnus is currently not playing very well.
Yeah I'm rooting for Nepo. Nepo is 4W 8D 1L vs Carlsen in classical TC, he has a decent shot imo so long as the series won't go to tiebreakers where Magnus is just unbeatable.
Vishy and Kramnik have offered their thoughts as well, here
Carlsen remains the favorite but I think people underestimate Nepo.

Nepo playing white in Game 1, if he starts off with a bang, he can win this thing.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Let's play chess

Post by Marseille07 »

Game 1 underway...looks like a draw. Not a good start for Nepo.

I didn't know he primarily plays e4.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Let's play chess

Post by Marseille07 »

I wonder why lichess provides an option to call it a draw when your opponent leaves the game (which happens quite often for anonymous games).

Does anyone really go, "Hmm, my opponent just left the game...I think it's a draw"?
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Let's play chess

Post by willthrill81 »

Marseille07 wrote: Fri Nov 26, 2021 2:40 pm I wonder why lichess provides an option to call it a draw when your opponent leaves the game (which happens quite often for anonymous games).

Does anyone really go, "Hmm, my opponent just left the game...I think it's a draw"?
I've never understood that option either.
The Sensible Steward
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Let's play chess

Post by willthrill81 »

Marseille07 wrote: Fri Nov 26, 2021 10:21 am Game 1 underway...looks like a draw. Not a good start for Nepo.
I agree. Nepo apparently missed 21.Bf4, and after playing the move at 22, the game was apparently dead drawn. However, Magnus was perhaps a bit reckless with 8.Na5, though who am I to argue with one of the greatest players of all time?
The Sensible Steward
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Let's play chess

Post by Marseille07 »

willthrill81 wrote: Fri Nov 26, 2021 4:31 pm
Marseille07 wrote: Fri Nov 26, 2021 10:21 am Game 1 underway...looks like a draw. Not a good start for Nepo.
I agree. Nepo apparently missed 21.Bf4, and after playing the move at 22, the game was apparently dead drawn. However, Magnus was perhaps a bit reckless with 8.Na5, though who am I to argue with one of the greatest players of all time?
Levy was talking about 8. Na5 as well. A very unusual move but Nepo couldn't punish Carlsen.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Let's play chess

Post by Marseille07 »

Game 2 coming up; Nepo has to draw this one as Carlsen plays white.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Let's play chess

Post by Marseille07 »

Pretty good Game 2 recap by Levy: https://www.youtube.com/watch?v=CcXcdoG0W2c

Interesting that everyone talks about 23 ... Ba8 but stockfish thinks that was fine and 24 .. c3 was not.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Let's play chess

Post by willthrill81 »

Marseille07 wrote: Sat Nov 27, 2021 11:44 am Pretty good Game 2 recap by Levy: https://www.youtube.com/watch?v=CcXcdoG0W2c

Interesting that everyone talks about 23 ... Ba8 but stockfish thinks that was fine and 24 .. c3 was not.
It was a very interesting game, to be sure. The match doesn't look good for Nepo yet to me.
The Sensible Steward
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Let's play chess

Post by Marseille07 »

willthrill81 wrote: Sat Nov 27, 2021 8:47 pm
Marseille07 wrote: Sat Nov 27, 2021 11:44 am Pretty good Game 2 recap by Levy: https://www.youtube.com/watch?v=CcXcdoG0W2c

Interesting that everyone talks about 23 ... Ba8 but stockfish thinks that was fine and 24 .. c3 was not.
It was a very interesting game, to be sure. The match doesn't look good for Nepo yet to me.
Nepo's tactical approach is interesting so far. I don't play e4 but apparently that's all he plays. E4 tends to invite open & tactical games than d4.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Let's play chess

Post by willthrill81 »

Marseille07 wrote: Sat Nov 27, 2021 10:35 pm
willthrill81 wrote: Sat Nov 27, 2021 8:47 pm
Marseille07 wrote: Sat Nov 27, 2021 11:44 am Pretty good Game 2 recap by Levy: https://www.youtube.com/watch?v=CcXcdoG0W2c

Interesting that everyone talks about 23 ... Ba8 but stockfish thinks that was fine and 24 .. c3 was not.
It was a very interesting game, to be sure. The match doesn't look good for Nepo yet to me.
Nepo's tactical approach is interesting so far. I don't play e4 but apparently that's all he plays. E4 tends to invite open & tactical games than d4.
1. e4 is all I play.
The Sensible Steward
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Let's play chess

Post by Marseille07 »

willthrill81 wrote: Sat Nov 27, 2021 11:48 pm 1. e4 is all I play.
I'm not a huge fan of e4 because my tactics isn't that great, and e4 often walks into the Sicilian defense.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Let's play chess

Post by willthrill81 »

Marseille07 wrote: Sun Nov 28, 2021 12:59 am
willthrill81 wrote: Sat Nov 27, 2021 11:48 pm 1. e4 is all I play.
I'm not a huge fan of e4 because my tactics isn't that great, and e4 often walks into the Sicilian defense.
Different strokes. I prefer the more open positions that e4 tends to lead to, and yes, one must be prepared for the Sicilian. Actually, I believe the French Defense to be at least as good for equalizing the position for black.
The Sensible Steward
Post Reply