lichess.org
Donate

rating distribution is broken

#21
@Wisherwood said in #20:
> You don't even play any games on this site, you have less than 100.
If you bothered to look more carefully, you might have noticed that the number is result of (1) me playing here only for few months and (2) playing almost exclusively longer time controls.

> If you had more experience playing then you would know that the server restart only affects you if...
If you bothered to read the comment (one that you replied to earlier) more carefully, you might have noticed that I almost never play around the time server is usually restarted (9:00 in the morning). Maybe even never, without "almost".

Your lack of attention is something I cannot help you with either.
#22
@Wisherwood said in #20:
> You don't even play any games on this site, you have less than 100.
Actually, this is really funny (the information is from our profiles):

User Wisherwoood: member since 2016-11-30 (i.e. more than 7 years), time spent playing: 5 days 17 hours 17 minutes
User mkubecek: member since 2023-09-30 (i.e. less than 7 months), time spent playing: 2 days 22 hours 11 minutes

In other words, in less than one twelfth of the time, I played more than half of how much you did. Come again, which one of us is not playing much here? :-)
#23
@mkubecek I was speaking of experience in the abstract sense rather than concrete. So from the data you've provided, and what I said earlier, about how one can easily take 10 minutes here or there during a winning 60 minute classical position with 30 minutes on your clock for example where your opponent still hasn't resigned when they could've. It's technically possible, and no I'm not going to go through your games and see how much of your time usage was wasted, when you could have basically just played your move instantly because it was 100% forced else loss for example yet taking several minutes, searching for these kinds of things. Will I search for them? No.. I don't really care to go through for both of us and resolve how our aggregation of experiences on lichess would amount to who the stronger chess player is. I arguably have a shot against you in classical, just due to personal reasons involving relationships between myself and masters. And it seems you wouldn't even begin to cope against me in ultra or hyper for example, though you also have hidden experiences not known by me for example you might have an alt with thousands of ultra games or maybe you've played 1/2+0 or 1/4+0 anonymously on incognito lichess.org for example and developed plans to overcome a top 1000 player and you're ready to execute, then you can definitely win some games. But I mean if you want to play a match this is a format that could work:

3 classical games
5 rapid games
7 blitz games
11 bullet games
11 ultra games

Each row constitutes 1155 total points, and you sum each row your score is and whoever is larger at the end wins.
For example
in classical, suppose my score was LWW = 2/3
I would get 1155 * (2/3) = 770 points and you 385.
We do that for each rated time control for standard chess on lichess.org, to make sure we account for every rated category in standard lichess.org

The match's winner is determined by whose sum of scores across all the rows is larger. It's possible though seems fairly unlikely to draw the match. Each time control is taken to be the minimum amount of time necessary to be deemed that category. For example, that means 1/2+0 is for bullet. It's hard to say whether 1/4+whatever is shorter than 3+0, so the tournament organizer would need to write a script to iterate over the lichess database and total up the distributions for amount of time taken in each possible time control for each category of time controls, like 3+0 versus 1/4+whatever it takes to become blitz, which of these two have faster games on average? I guess we do it like that, but honestly we could also pick one time control at a time, like, I pick 1/2+0 for bullet then you say 45+45 classical, then I say 8+0 rapid, then you say 5+4 blitz.

Get it?

I think that's better where we just hand it off one at a time picking a time control for each possible time control category, then slowly we eventually play all the games. The way the games will be distributed is also important. So we would need to agree on that too, and a time and a date.

But anyway, this all assumes we can get a prize pool and someone or organization to sponsor it anyway, and it seems weird to want to give money to furnish an online match between two lichess ppl, one who is obsessed with bullet, and the other, classical. But anyway I'm sure you're not interested in playing me in ultra/hyper but I think it's a reasonable definition for covering 'chess knowledge' to be able to play off instincts as well as deeper more strategical calculation, and the passing off of deciding the time controls helps to keep the whole thing more balanced towards equality. Though whoever goes first in that also has to have 1 more black game than the other player, I think that should be possible to do by just alternating and they start with black if they chose the first time control.
#24
@Wisherwood I'm afraid you completely missed the point. The point was that you questioned my experience and (game playing) activity which was based only on the game count. I pointed out that this is misleading as you can hardly count a 1+1 game as equal to a 45+45 game. And that when you look at the time spent playing, the picture suddenly looks quite different.

Your assumption that people spend big part of long(er) time control games by getting bored and doing something else (on their time) may reflect your personal approach but I can assure you it's far from universal. Perhaps it's the case for people who prefer blitz and do not actually enjoy long(er) games but people who focus on classical mostly have little need to waste the time. Personally, I usually only take short breaks to visit a toilet or drink something and even that happens mostly on my opponent's moves. When you try to give it your best, even 45+45 is not as long as it seems - or 90+30 for 40 moves, FWIW.

I definitely didn't want to compare who is a stronger player as that would make no sense when each of us focuses on very different type of games. And, more important, it would be even less relevant for this topic than comparing playing experience on lichess. Not to mention that the type of match you suggest would be completely pointless, giving only 40% of the score to time controls where I played at least one game since 1988 (maybe longer) and 40% to something that has nothing to do with chess skills. If you want to claim you are stronger, just say it and don't invent absurd schemes like this to support such claim.
#25
@mkubecek Little to do with chess skills? Ultra bullet is chess in 15 seconds per side. Classical is chess in 60 minutes per side. At the core they are both playing around the chess infrastructure however just imposing time restraints. Whether or not you've understood it yet, chess itself is what the engine says, classical is a game we play on the infrastructure of chess in a certain fixed amount of time, and ultra bullet is a game we play on the infrastructure of chess in a certain fixed amount of time. Whether or not it's classical, if you have a 3:1 time advantage, you have a time advantage. If I'm down 3:1 on time in a winning endgame, in bullet, there's a very good chance I convert it to a win, so that situation is a pleasant one for me though I took the risk of playing a bit slower due to forcing combinations on the board. They both have similar trade offs and again are just approximations of chess with a differing time constant, so to say that it has nothing to do with chess is preposterous. If anything, it is another lens in which to analyze both the competitive mindset of a human as well as another lens in which to understand chess in a different light. For example, in ultra for extremely low times, there are 100% calculated perfect wins that force you for example to make a premove as well as click and drag another piece to hover over a square to react to just one move and when your premove flicks off you're immediately making the next forcing move and so on. It adds many layers on top of the chess, yes, layers inherent to being competitive in shorter time spans, however I hope you see my point that just because classical is important to you does not mean that it is any "better" of a lens in which to analyze chess as a competitive game than those created from different time constants. Perhaps it could be marketed better than how lichess does it, but such a match is not pointless but rather an accumulation of a single match sample of all possible competitive rated chess games offered at lichess.org. To me that's what chess skill amounts to, is how well you are able to adapt to any situation involving chess - time control both short and long, correspondence, zerk chess (half your time to start than opponent), tournaments on lichess.org which incentivize you to play quickly as if you play more games you have more chances to get streaks and gain points - so losses are only penalized in that you've wasted that amount of time on the game, as your position in the tournament is exactly the same as it was at the start of the game minus the fact that time has passed, and so on. The better chess player is the one who can win arbitrary games setup from the chess infrastructure - For example, Hikaru sits down at an endgame which has never been played in a real game before, he calculates a bit and speaks the winning move for both sides. This situation wasn't chess, however, due to his experiences, he was able to solve the situation accordingly and act in a winning fashion, whereas someone who knew nothing of chess would not win in such a situation. Even though that situation isn't chess, it says something about how we understand chess. Similar , t=60/2^k minutes, at what point is (chess, t) no longer a "real game"? The answer, in my opinion, only depends on the infrastructure the chess being played on is good enough to register moves as soon as you've made them, and shows moves as soon as they have been made.