@tygxc, do you understand that you can't reach certainty with statistics?
Chess will never be solved, here's why
Sort:
Sep 8, 2022
0
#3723
Three avoidable errors.
- There is no excuse for thinking I "think you can reach certainty with statistics", since my previous post says "you can't reach certainty with statistics".
- It would require you to have suffered catastrophic amnesia to genuinely believe statistics was all I know
- The fact that we can't presently justify certainty whether 2. Ba6 loses is not based at all on statistics, any more than is the fact that we couldn't justify certainty whether "Fermat's last theorem" was a true proposition before this theorem was finally proved in the 1990s.
It's worth noting the distinction between propositions and meta-propositions here (the latter referring to the status of meta-knowledge, which is knowledge about what is known).
Optimissed wrote:
Nevertheless, your performance here has been so completely myopic and childish that I and others have entirely stopped taking you seriously. I stopped taking the 13 year old seriously, years ago. All he knows is personal attack. However, you have consistently evaded my very well-directed criticisms and it's clear you don't know how to answer them, since your only recourse has been to make personal attacks.
There's no excuse for the things you have written. I would suggest that perhaps your confidence has been bolstered but I would seriously reconsider the credentials of those agreeing with you. I think you're really a small minority which pretends to be a majority, because that's the only way to stay afloat. It would be better to sink and start again. If you actually answered my criticisms in stead of constant evasions, you would learn something. As it is, you will not learn.
I see your ego has reconstituted to its dysfunctional norm after running away for a week.
Optimissed wrote:
I've been very busy.
I believe that you don't understand the meaning of what "ego" was considered, in Freudian terms, to be. You're very, very confident but extremely ignorant of many things and I'd prefer not to be taking to you, because I find, in general, that you have nothing of interest to write. I think the problem is with YOUR ego, in that respect. If there was any real mental content, your posts would reflect it.
Yep, this is the norm I was referring to.
NervesofButter wrote:
LOL...btickler and optimissed have yet again invaded another forum thread and turned it into there own personal battlefield.
This post is well over a hundred pages late, if you are going to make it.
Optimissed wrote:
Please stop being so childish.
Other things are happening. The World is a different place today and yet you are sureally the same.
Yes, it's different. That's not some ticket to pretend you are.
I already posted my respects, before you posted your own thread on the subject. Don't bother trying to make today's events about you.
Optimissed wrote:
Without people to look down on, you would be completely lost. You'd have no reason to be here. You don't only disgust me so don't pretend it is just me. Somehow you need to grow up.
All of the above is applicable to you. The difference is that you cannot be any other way. I, on the the other hand, am just forcing you to show people who you are by mirroring your behavior back at you (minus the worst stuff that I won't stoop to), which you obligingly stumble into like clockwork.
Sep 8, 2022
0
#3729
Optimissed wrote:
NervesofButter wrote:
LOL...btickler and optimissed have yet again invaded another forum thread and turned it into there own personal battlefield.
Nerves of Butter, I'm sorry but this needs to be said. Don't comment if you can't recognise what the problem is.
This is supposed to be an open forum where anyone can express their opinion. NervesofButter has a different view of what "the problem" may be. Don't tell others that they shouldn't express a viewpoint that may be different from yours.
Optimissed wrote:
You didn't stay unblocked for long, did you.
You get the last comment, as usual. Forgive me if I don't reply but I haven't actually read your post even though it's short, because I imagine it will be as pretentious as all the rest of the stuff you write..
You can't discourage me with a win-win scenario 
. I will keep pointing out when you jump on other posters or are generally condescending to everyone around you. Respond, or don't, it doesn't matter.
NervesofButter wrote:
No problemo. You proved your maturity/mentality/paranoia with the message you just sent me. I used to think you were just a mildly deranged old man. But now? Yea dont worry i wont be in contact with you anymore.
Don't feel special, he reaches out to everyone he thinks he can enlist in his crusade. He really should look up that "Dark Lord" guy from the Niemann threads...that guy is right up his alley .
Optimissed wrote:
btickler wrote:
Optimissed wrote:
Without people to look down on, you would be completely lost. You'd have no reason to be here. You don't only disgust me so don't pretend it is just me. Somehow you need to grow up.
All of the above is applicable to you. The difference is that you cannot be any other way. I, on the the other hand, am just forcing you to show people who you are by mirroring your behavior back at you (minus the worst stuff that I won't stoop to), which you obligingly stumble into like clockwork.
I suppose, if all else were to fail, you could try to be a comedian. You probably even believe you.
30 minutes flat for you to forget you were pretending you didn't read that post .
Optimissed wrote:
btickler wrote:
30 minutes flat for you to forget you were pretending you didn't read that post .
I didn't. I can quote something and not read it, believe it or not. You're the one who's deluded. Now, goodbye.
Sure, sounds perfectly plausible...
[and]
The truth is, I've been extremely busy with a venture that's working out well and also wanted to stay away for a couple of days. But I did want to come back just to say what I think of some of you. You know who you are. The ones who never make honest arguments and always pretend they already answered. The old, online trick. The ones who start personal attacks just so someone will answer them back. That's what they live for.
I'm aware that the majority of people here, I don't know. The second majority I like and get on will with. But there are some who never grew up and who treat this as a kids' playground, where they're the bullies. It's ok if someone disagrees with someone else but if there's someone who argues their opinions better, the gang of kids doesn't like that. Those kids just love pretending they are the big men.
They probably even believe it, because that's what happens when people never grow up. Maybe when they've spent their time in an institution which prevents growing up ... maybe the army, maybe a university. They think they're normal and right and good and proper. It's the ones who can function better than them, or at least, differently, who are the kids, with everything reversed. Apparently there's no cure. It's too late for one. From now on, I'll be playing more chess, instead of wasting my time, learning from fools. It's been instructive but probably time to move on.
Quoting for posterity .
Back on topic.
This is the most recent finished ICCF WC game.
It has been established this is > 99% sure to be a perfect game with no errors,
and has < 1 % chance to contain 2 errors that undo each other.
A draw was agreed because of a forced transition to a table base drawn 7 men rook ending.
https://www.iccf.com/game?id=1164280
Chess can be weakly solved using such a game as backbone.
Look at 3 alternatives for 119 Bd3. Do they draw too?
Look at 3 alternatives for 118 f6. Do they draw too?
Look at 3 alternatrives for 117 Rh1. Do they draw too?
...
Like that all the way down until it comes to some other ICCF WC drawn game.
tygxc wrote:
Back on topic.
This is the most recent finished ICCF WC game.
It has been established this is > 99% sure to be a perfect game with no errors,
and has < 1 % chance to contain 2 errors that undo each other.
A draw was agreed because of a forced transition to a table base drawn 7 men rook ending.
https://www.iccf.com/game?id=1164280
Chess can be weakly solved using such a game as backbone.
Look at 3 alternatives for 119 Bd3. Do they draw too?
Look at 3 alternatives for 118 f6. Do they draw too?
Look at 3 alternatrives for 117 Rh1. Do they draw too?
...
Like that all the way down until it comes to some other ICCF WC drawn game.
"It has been established this is > 99% sure to be a perfect game with no errors,
and has < 1 % chance to contain 2 errors that undo each other."
This was established by you using poisson distribution that is not applicable to determine probability of chess errors.
@4708
"This was established by you using poisson distribution that is not applicable to determine probability of chess errors."
++ It is applicable.
Please come up with another plausible distribution of errors per game.
I say: 136 ICCF WC games = 127 draws with 0 error + 9 decisive games with 1 error.
What do you say with whatever distribution you deem more applicable?
tygxc wrote:
@4708
"This was established by you using poisson distribution that is not applicable to determine probability of chess errors."
++ It is applicable.
Please come up with another plausible distribution of errors per game.
I say: 136 ICCF WC games = 127 draws with 0 error + 9 decisive games with 1 error.
What do you say with whatever distribution you deem more applicable?
Events that unpredictably affect the probability of the next event can't really be reliably distributed like that. The probabilities are unpredictable. We would have to solve chess to be sure.
@4710
"Events that unpredictably affect the probability of the next event can't really be reliably distributed like that."
++ Telephone switching gear and hospital emergency rooms are designed using Poisson.
They should neither be overwhelmed, nor too expensive and idle. During e.g. 9/11 that might have been broken, with phone calls or hospital emergencies stemming from the same event.
Regardless, as calculated for the ICCF WC draws, there is > 99% sure either 0 or 1 error per game, with < 1% probablitity of a drawn game with 2 errors.
If there is > 99% sure only 0 or 1 error, then there is no next event,
so whether the non existent next event depends on the first event or not does not matter.
"We would have to solve chess to be sure."
++ Now we can say the game @4707 is > 99% sure to be a perfect game with no errors.
With another distribution it may be 99.7% or 99.1% or even 98.4%, whatever.
Thus such ICCF WC drawn games are suitable as a backbone to weakly solve chess with the procedure explained @4707.
Those who claim to know something about statistics and probability should realise that its purpose is to extract good knowledge from bad data. It purpose is not to say we cannot tell. Gauss developed much of it to extract precise information from very blurred astronomical data. He did not say: bad data, I cannot tell.
tygxc wrote:
@4710
"Events that unpredictably affect the probability of the next event can't really be reliably distributed like that."
++ Telephone switching gear and hospital emergency rooms are designed using Poisson.
They should neither be overwhelmed, nor too expensive and idle. During e.g. 9/11 that might have been broken, with phone calls or hospital emergencies stemming from the same event.
Regardless, as calculated for the ICCF WC draws, there is > 99% sure either 0 or 1 error per game, with < 1% probablitity of a drawn game with 2 errors.
If there is > 99% sure only 0 or 1 error, then there is no next event,
so whether the non existent next event depends on the first event or not does not matter.
"We would have to solve chess to be sure."
++ Now we can say the game @4707 is > 99% sure to be a perfect game with no errors.
With another distribution it may be 99.7% or 99.1% or even 98.4%, whatever.
Thus such ICCF WC drawn games are suitable as a backbone to weakly solve chess with the procedure explained @4707.
Those who claim to know something about statistics and probability should realise that its purpose is to extract good knowledge from bad data. It purpose is not to say we cannot tell. Gauss developed much of it to extract precise information from very blurred astronomical data. He did not say: bad data, I cannot tell.
Chess errors don't follow a poisson distribution. First of all we have no knowledge of when an error actually occurs. It's one possibility that chess is a forced win and the first error has a probability of 99,9% or any other high number, its unknown. The next error could have a probability of 5%. In complex positions where a series of only moves will maintain the current evaluation probability of errors will suddenly increase. If the first error leads to a position where plenty lines force a draw, probability of the next error could go close to 0%. Obviously the probability of these errors don't fit the poisson distribution.
The example you gave of emergency services is a problematic one for poisson distribution and it would not handle well a possibility of mass emergencies where people would call the emergency line all at once. The more there is available data the better it works. With chess we have very little accurate data of when errors occur.
@4712
"Chess errors don't follow a poisson distribution." ++ How do you know that? A Poisson distribution makes sense as in similar problems. Even @Elroch suggested it. Anyway, please then use another distribution you think is more appropriate and then tell how many games with 0, 1, 2... errors there are in the 136 ICCF WC games with 127 draws, 6 white wins and 3 black wins.
"we have no knowledge of when an error actually occurs" ++ We do not need to know that.
The error is usually the last move in an ICCF WC finals decisive game.
"It's one possibility that chess is a forced win"
++ Yes, a priori chess can be a draw, a white win, or a black win.
It is impossible to fit a Poisson distribution where the probability of an odd number of errors is 127/136, so that rules out a white or black win.
It is possible to fit a Poisson distribution where the probability of an odd number of errors is 9/136 with average 0.07 error per game and thus the 127 draws are > 99% sure to contain 0 error and the 9 decisive games are > 99% sure to contain exactly 1 error.
"Obviously the probability of these errors don't fit the poisson distribution."
++ Obviously a Poisson distribution is good here as in many similar problems.
Please show an alternative distribution of errors to explain the 30th ICCF WC finals' results.
I say: chess is a draw, 127 games with 0 error, 9 games with 1 error
You say: chess is a draw / a white win / a black win with:
... games with 0 error, ... games with 1 error, ... games with 2 errors, ... errors with 3 errors.
Bored or bothered, whatsoever?
This thread turned into a string of pegs for hangers-on, no?