lichess.org
Donate

How long is an opening?

How much ground must we gain before we change phases?

If all the space was behind the material, then I would say we have reached the middle game. The middle games would in theory start when there would be 16 empty squares behind the material. With that concept, I would say the phases is relative to the territorial space gained by the pawns and pieces. A game can end without a middle game or end game.

Lichess opening book stops at move 25 (ply 50). @x_trickster
https://lichess.org/opening
https://lichess.org/analysis#explorer

Lichess determined the opening ends on move 10 and the middle game starts on move 11.
https://lichess.org/study/kjnGc2Pt/DXpZEivG

Maybe the 3 phases can be determined by pawn structures. If the structure is complete then the opening phase is done or is it when the pawn phalanx phase starts? https://chessfox.com/pawn-structures-why-pawns-are-the-soul-of-chess/

How much ground must we gain before we change phases? If all the space was behind the material, then I would say we have reached the middle game. The middle games would in theory start when there would be 16 empty squares behind the material. With that concept, I would say the phases is relative to the territorial space gained by the pawns and pieces. A game can end without a middle game or end game. Lichess opening book stops at move 25 (ply 50). @x_trickster https://lichess.org/opening https://lichess.org/analysis#explorer Lichess determined the opening ends on move 10 and the middle game starts on move 11. https://lichess.org/study/kjnGc2Pt/DXpZEivG Maybe the 3 phases can be determined by pawn structures. If the structure is complete then the opening phase is done or is it when the pawn phalanx phase starts? https://chessfox.com/pawn-structures-why-pawns-are-the-soul-of-chess/

@x_trickster said in #8:

Conclusions?? I came here to find 'How long is an opening?'. Still no answer :)
@RitwinSathish said in #9:
Your blog post was very good, but you didn't explain about the conclusion of what happened to the opening depending on the elo.

Sorry for the inconvenience. I have added a conclusion section to the post.

@x_trickster said in #8: > Conclusions?? I came here to find 'How long is an opening?'. Still no answer :) @RitwinSathish said in #9: > Your blog post was very good, but you didn't explain about the conclusion of what happened to the opening depending on the elo. Sorry for the inconvenience. I have added a conclusion section to the post.

I was always taught that the opening ends when the rooks are connected.

I was always taught that the opening ends when the rooks are connected.

@oortcloud_o No inconvenience whatsoever. Love the articles, thank you.

@oortcloud_o No inconvenience whatsoever. Love the articles, thank you.

What games hold the records of shortest and longest phases ?
The opening ...
Shortest Fool's mate
Longest:

The middle game ...
Shortest:
Longest:

The end game ...
Shortest:
Longest:

Thanks for your answers.

What games hold the records of shortest and longest phases ? The opening ... Shortest Fool's mate Longest: The middle game ... Shortest: Longest: The end game ... Shortest: Longest: Thanks for your answers.

I think one can ask a question, and present various property of the data**, that are worth sharing as preliminary finding and suggestions, for further work. or even threads to answer the question. So we just got the lichess interpretation or rules for the server analysis to segment games into phases.. SF internals might have its own, now that it toy with depth searches by procuration (SF classical to NNue, until SF16 which is a mystery to me so far). Basically material attrition might be a guide. Or development.

about lichess servers analysis gist shared previously:

  • How often a small sliding window contains a mix of black and white pieces
    An interesting one this one, btw. Borrowing on physical entropy notions but with the constraint of balanced material. Material from both sides. something like that. Hmmmm...

Here, my impression, from an earlier version, was that the notion of movetime was used and we were not talking about chess games properties in them selves but about the player preparation version of when opening stops, per their move time move reflex time response.
A psychological notion of opening length. I wonder if that depends on rating (as surrogate measure of player experience, or the rating differential across the board).

I may not have read the latest version. I only looked fast at the figures. And some headers.
Edit: last pic (may have been there) is about some rating as factor. good.

** Data: if it could be specified to point that others could reproduce the result, and therefore also interpret future result without reproducing them first, as those specification would allow alternate points of view to retrace the reasoning that the figure would suggest, that is a general point, not specific to this blog, then the authors would have equals in the conversation, otherwise, the figures are suggestion. And there might be nothing wrong with that, if that was the intent. (here suggestion is my intent).

I think one can ask a question, and present various property of the data**, that are worth sharing as preliminary finding and suggestions, for further work. or even threads to answer the question. So we just got the lichess interpretation or rules for the server analysis to segment games into phases.. SF internals might have its own, now that it toy with depth searches by procuration (SF classical to NNue, until SF16 which is a mystery to me so far). Basically material attrition might be a guide. Or development. about lichess servers analysis gist shared previously: > * How often a small sliding window contains a mix of black and white pieces An interesting one this one, btw. Borrowing on physical entropy notions but with the constraint of balanced material. Material from both sides. something like that. Hmmmm... Here, my impression, from an earlier version, was that the notion of movetime was used and we were not talking about chess games properties in them selves but about the player preparation version of when opening stops, per their move time move reflex time response. A psychological notion of opening length. I wonder if that depends on rating (as surrogate measure of player experience, or the rating differential across the board). I may not have read the latest version. I only looked fast at the figures. And some headers. Edit: last pic (may have been there) is about some rating as factor. good. ** Data: if it could be specified to point that others could reproduce the result, and therefore also interpret future result without reproducing them first, as those specification would allow alternate points of view to retrace the reasoning that the figure would suggest, that is a general point, not specific to this blog, then the authors would have equals in the conversation, otherwise, the figures are suggestion. And there might be nothing wrong with that, if that was the intent. (here suggestion is my intent).

yep comments asking questions and offering suggestions @oortcloud_o are demonstrating the interest. It is a good thing.

yep comments asking questions and offering suggestions @oortcloud_o are demonstrating the interest. It is a good thing.

I don't wanna be mean or anything it's just I am 10 years old and the X and Y axis don't make sense so do you think that you can make it a bit simpler for me?
Thanks!

I don't wanna be mean or anything it's just I am 10 years old and the X and Y axis don't make sense so do you think that you can make it a bit simpler for me? Thanks!

@dboing said in #17:

I think one can ask a question, and present various property of the data**, that are worth sharing as preliminary finding and suggestions, for further work. or even threads to answer the question. So we just got the lichess interpretation or rules for the server analysis to segment games into phases.. SF internals might have its own, now that it toy with depth searches by procuration (SF classical to NNue, until SF16 which is a mystery to me so far). Basically material attrition might be a guide. Or development.

about lichess servers analysis gist shared previously:
An interesting one this one, btw. Borrowing on physical entropy notions but with the constraint of balanced material. Material from both sides. something like that. Hmmmm...

Here, my impression, from an earlier version, was that the notion of movetime was used and we were not talking about chess games properties in them selves but about the player preparation version of when opening stops, per their move time move reflex time response.
A psychological notion of opening length. I wonder if that depends on rating (as surrogate measure of player experience, or the rating differential across the board).

I may not have read the latest version. I only looked fast at the figures. And some headers.
Edit: last pic (may have been there) is about some rating as factor. good.

** Data: if it could be specified to point that others could reproduce the result, and therefore also interpret future result without reproducing them first, as those specification would allow alternate points of view to retrace the reasoning that the figure would suggest, that is a general point, not specific to this blog, then the authors would have equals in the conversation, otherwise, the figures are suggestion. And there might be nothing wrong with that, if that was the intent. (here suggestion is my intent).
Um... how long did it take you to write this???

@dboing said in #17: > I think one can ask a question, and present various property of the data**, that are worth sharing as preliminary finding and suggestions, for further work. or even threads to answer the question. So we just got the lichess interpretation or rules for the server analysis to segment games into phases.. SF internals might have its own, now that it toy with depth searches by procuration (SF classical to NNue, until SF16 which is a mystery to me so far). Basically material attrition might be a guide. Or development. > > about lichess servers analysis gist shared previously: > An interesting one this one, btw. Borrowing on physical entropy notions but with the constraint of balanced material. Material from both sides. something like that. Hmmmm... > > Here, my impression, from an earlier version, was that the notion of movetime was used and we were not talking about chess games properties in them selves but about the player preparation version of when opening stops, per their move time move reflex time response. > A psychological notion of opening length. I wonder if that depends on rating (as surrogate measure of player experience, or the rating differential across the board). > > I may not have read the latest version. I only looked fast at the figures. And some headers. > Edit: last pic (may have been there) is about some rating as factor. good. > > ** Data: if it could be specified to point that others could reproduce the result, and therefore also interpret future result without reproducing them first, as those specification would allow alternate points of view to retrace the reasoning that the figure would suggest, that is a general point, not specific to this blog, then the authors would have equals in the conversation, otherwise, the figures are suggestion. And there might be nothing wrong with that, if that was the intent. (here suggestion is my intent). Um... how long did it take you to write this???