This article is within the scope of WikiProject Chess, a collaborative effort to improve the coverage of Chess on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.ChessWikipedia:WikiProject ChessTemplate:WikiProject Chesschess articles
The books that I have listed as "looks elementry/introductory" are books that I don't have, but I think that they are elementry or introductory based on sample pages or reviews. Bubba73
The paragraph under "queen endings" that begins "Endings with asymmetric piece possession ..." seems misplaced since it doesn't refer to queen endings, but I don't know where it should go. It was in this location before I expanded "endings without pawns" and moved knight endings and bishop endings above rook endings. Also, I'm not sure that the comment about asymetrical material arangements being less common. Does someone have an idea about where this paragraph should go, and if it should be changed?
Bubba73
I added the Misc Books section and moved some of the ones I marked as "looks elementry/introductory" there. I don't have those books, so if anyone has them, please feel free to make a comment about them and move them to one of the other sections, if appropriate. Bubba73
"All rook endings are drawn" - most sources say this quote was from Tarrasch, but Korchnoi says Tartakower instead. I don't know which is right, but I'm inclined to think that the majority (Tarrasch) is probably right. Also, though, most sources say "All rook endings are drawn" instead of "All rook AND PAWN endings are drawn", but I didn't make note of that. Again, I'm inclined to go with the majority and leave out "and pawn".
--Bubba7303:52, 21 May 2005 (UTC)[reply]
The article says "Endings with asymmetric piece possession are less common". Is this really true? I think I read somewhere that the most common endgame was a rook versus a minor piece, with pawns. Bubba73 20:37, August 24, 2005 (UTC)
I added a table of the most common endgames in actualgames, based on data in Muller & Lamprecht. However, they have "pawn endings" and then "K+P vs. K". I don't know whether the second is included in the first. Also, they have "R vs R" and then "R+P vs. R" and "R+2P vs R", and I don't know if "R vs R" includes the second and third category. Bubba73(talk), 20:17, 12 December 2005 (UTC)[reply]
I put in the table based on Muller & Lamprecht, then I added the second table, breaking down the "rook & minor piece" vs "rook & minor piece" category into its components, based on other data in M&L. A recent edit removed the first table as being redundant. It is to an extent, but M&L saw fit to remove the first table. I'm not going to argue about it, but I would like to hear some comments about whether what was the first table should be in there or not. Thanks. Bubba73(talk), 03:21, 21 December 2005 (UTC)[reply]
Please find my own work on endgame frequencies [1]. I don't think it's contradicting to the currently published, but it can't be directly compared (as eg. in Miller's data Rooks + any number of pawns is taken together as one set, while in my sample r+p would be different than r+p+p). Additionally to the endgame frequencies, I have published the frequency of player blunders in such positions (Basically meaning: which endgames are more complex to play). I will refrain from editing this article, as for the sake of transparency I'm working on chess-db.com so there is possible conflict of interest, and in any case I don't want my contributions on Wikipedia to be questioned for their intention. — Preceding unsigned comment added by ChessyD (talk • contribs) 18:14, 16 November 2012 (UTC)[reply]
That is very interesting data. Yes, there is a difference because your data is limited to five pieces. And you are right to be concerned about the possible conflict of interest. You should not added to the article, but I'm not sure what to do with it. I'd been telling people about the importance of R+P vs. R and K+P vs. K - this bears it out. Bubba73You talkin' to me?19:45, 16 November 2012 (UTC)[reply]
I added the number of moves for the NNNN vs. Q win (85).
For the 290 move win, I removed the references to the mate in a further 6 moves, since the 6 moves is really the shorted distance to mate or capture. Indeed, the distance to mate from the position after the rook capture on move 290 would be 8. However, that does not mean that the starting position for the 290 move win is a mate in 298. Some confusion has been caused in some circles by the inappropriate mixing of distance to win and distance to mate concepts. -Marc Bourzutschky, 12/18/2005.
Yes, thanks, I believe you're right. It is easy (for me at least) to get confused between shortest path to mate and shortest path to conversion. Bubba73(talk), 03:17, 21 December 2005 (UTC)[reply]
I think that section shouldn't be there, because everything is already covered in draw (chess). That being said, I'm not removing it because I'd like to know if there's any chance of fixing that into the scope of endgame draws, principally stalemate and the fifty-move rule. fetofsHello!18:34, 4 June 2006 (UTC)[reply]
That's what I said in the comment to my edit a few hours ago. I also doubt that it needs to be here, for the same reason as you mentioned. But I didn't see fit to remove it just yet. Perhaps what you are talking about is giving examples of endgames that are stalemates and draws under the 50-move rule? There is probably already an example at stalemate. Bubba73(talk), 23:07, 4 June 2006 (UTC)[reply]
Statemate? That's a new one for me ;). But, going back into the discussion, if we aren't going to remove it just yet, we have to do something with it just yet. I'll try to rescope it. fetofsHello!12:34, 5 June 2006 (UTC)[reply]
Thinking about it some more, I think it should be removed. It is a way games can end, but not necessarily an "ending", as we usually use the term. A draw by agreement certainly doesn't have to come in an ending and 3-fold doesn't either. 50-move probably does, but not necessarily; same with stalemate. Bubba73(talk), 14:36, 5 June 2006 (UTC)[reply]
Yes, but a draw is not specific to the endgame, and there is already an artticle on draws. Also, the section says that the most common reasons for a draw are stalemate, 50-moves, and insufficient material; and that others are agreement and 3-fold. Draws by agreement are the most common in the endgame, probably followed by insufficient material. 50-move is quite rare. Bubba73(talk), 20:35, 5 June 2006 (UTC)[reply]
From above "A draw is a common resolution, and may often occur in the endgame". Resignations, some time forfeits, and sometimes checkmates occur in the endgame too. But all of these (including draws) cn occur in any phase of the game, so surely this is mentioned elsewhere, such as the main chess article. I'm not seeing much significance of the section of draws in this article. Bubba73(talk), 03:34, 6 June 2006 (UTC)[reply]
I favor that, but I'm not going to do it unless we have a consensus. It doesn't hurt to leave it in, I just don't think it is needed in this article, since it is in the others I mentioned. Bubba73(talk), 19:54, 6 June 2006 (UTC)[reply]
I decided to take the section out, because it doesn't directly relate to endgames and it is covered in at least three other articles. If someone objects, feel free to discuss it here. Bubba73(talk), 03:31, 10 June 2006 (UTC)[reply]
Has anybody noticed that the article suggests that for other uses of the word Endgame, we look at the article for... Endgame? Is this a leftover from when there were multiple articles on the subject, or what?--Raguleader06:08, 10 July 2006 (UTC)[reply]
Your observation is correct. It probably is a left over from something, I'm not sure though. Perhaps this should be moved to Endgame (chess) and this page could have links to it and other uses, but I don't know of a pages of other uses. Bubba73(talk), 03:03, 11 July 2006 (UTC)[reply]
I replaced the assumption that white is the attacking side and replaced "black" and "white" with "defender" and "attacker", resp. Does anyone have any thoughts? Bubba73(talk), 03:15, 16 October 2006 (UTC)[reply]
I fixed another instance of black/white instead of defender/attacker in the section labeled "Bishop and pawn versus bishop on the same color". I also reworded an apparent confusion in wording between the color of the bishop itself, and the color of the squares it travels on. I hope someone can double-check this change. CosineKitty (talk) 19:41, 13 November 2008 (UTC)[reply]
Thank you, Bubba73. Incidentally, I keep looking at the diagram labeled "Centurini, 1847". The caption says White (to move) can win, but it looks like a draw. It also seems to match the rules in this section for a draw. Can someone clarify this? If it is a win for White, how does White win? Otherwise the caption should be fixed. Thanks! CosineKitty (talk) 21:14, 14 November 2008 (UTC)[reply]
The reference does give that as a win, but the second criteria for a draw does seem to apply. Perhaps it means two diagonals that cross the pawn's path, and the black bishop can't maneuver on the a7-b8 diagonal. I'll try to figure it out. Bubba73(talk), 05:32, 15 November 2008 (UTC)[reply]
The defending bishop must have two diagonals, each of which have at least two squares to which it can move, not counting the square it is on. I've changed it to reflect this. Bubba73(talk), 03:51, 16 November 2008 (UTC)[reply]
The article was getting long, and the lengthly bibliography was moved to the new article Chess endgame literature, which also has two paragraphs about the history. Several of the currently common and accessible books were left here. Bubba73(talk), 20:58, 23 July 2007 (UTC)[reply]
It is sort of covered under the "Basic checkmates" section and the checkmate article. It states that two bishops or a bishop and knight are required. That implies that one minor piece is not sufficient, but that isn't stated. Do you think it needs to be in there? Bubba73(talk), 19:38, 9 September 2007 (UTC)[reply]
Yes - not so much in checkmate where a non-checkmate is somewhat off the subject, but definitely in endgame where this is one of the most fundamental facts, and I don't think we can expect people to look in checkmate for this. For instance, Endgame#Knight and pawn versus knight says the knight can be sacrificed for the pawn, but takes it for granted that we all know why we should give up 3 points for 1. Endgame#Basic checkmates says that 2 bishops or a bishop and a knight will work, but it doesn't say they are required. So I think this basic fact should be made explicit. Art LaPella05:26, 10 September 2007 (UTC)[reply]
I made it more explicit in the Basic Checkmates section. However, the section did state that two knights was not sufficient and that clearly implies that one knight isn't sufficient. It also states that two bishops on the same color can't checkmate, so it is clear that one bishop can't do it. At least that is obvious to anyone familiar to chess, but a lot of readers aren't that familiar with it, so now it is in there explicitly. Thank you for pointing this out. Bubba73(talk), 05:33, 10 September 2007 (UTC)[reply]
I think edits to chess endgame are excellent (better then on Chess openings for sure). I've reverted. Be precise about what you percieve to be the problem looks like you are getting personal with this editor. Stay calming. Regards, SunCreator(talk)01:26, 2 August 2011 (UTC)[reply]
I think User:Jaxdelaguerre really understands the subject and the edits are well intended but also it's way to flowery with redundancy that can be cut and requires copy-editing afterwards. But, no big deal. It's nice to have fresh input. There is always WP:BRD, stay cool! Regards, SunCreator(talk)01:03, 2 August 2011 (UTC)[reply]
To get the ball rolling, I object to the first paragraph:
The chess endgame (or end game or ending) is a non-rigorous term referring to the stage of the game when the strategic patterns laid out in the opening and executed in the middlegame have not led to sudden checkmate and the cumulative effect of irreversible piece capture (unlike chess's Japanese cognate shogi in which pieces return) has resulted in a reduced number of pieces remaining on the board in a position progressively more subject to concrete calculation by highly skilled players and especially by chess-playing computers.
The paragraph is one extremely long, convoluted sentence. The sentence is about 90 words long! Just tell what the endgame is, instead of go into all of this. Skip the "non-rigorous" bit, skip the "strategic patterns", skip the "sudden checlmate", skip the "cumulative effect of irreversible piece capture", don't bring up shogi, "reduced pieces" is too vague, "position progressively more subject to concrete calculation" is true but it doesn't belong here, and skip "skilled players" and "computers". In short revert the whole paragraph. Bubba73You talkin' to me?01:56, 2 August 2011 (UTC)[reply]
I really don't have much time for this, but I will ONE TIME defend my edit. After that if I am troll-reverted again I simply will withdraw from editing the Chess pages.
The chess endgame (or end game or ending) is a non-rigorous term
An article intro is effectively a definition of the term under discussion. The notion of "endgame" is among chessplayers proverbially not rigorous. Do you know any chessplayers? Have you ever discussed the endgame with a master? And if you don't like precise definition, why edit Wikipedia?
strategic patterns laid out in the opening and executed in the middlegame
This is to connect the reader to what they understand of the rest of the chess game and guide them to the portion of the game under discussion.
have not led to sudden checkmate
This is precisely what causes an endgame to occur, no midgame checkmate. Do you actually play chess?
cumulative effect of irreversible piece capture
Because a) in some games (including variants of western chess) pieces come back and b) the concrete, definitive calculability of endgames depends precisely on this phenomenon of reduced piece count
"reduced pieces"
I wrote "reduced piece count" not "reduced pieces". If you don't read my text, don't argue in favor of reverting it.
"position progressively more subject to concrete calculation" true but it doesn't belong here
Precisely where it belongs. It is part and parcel of the definition of the endgame as chessplayers understand it. The game BECOMES CONCRETELY CALCULABLE (by experts and computers, if not by beginners reading the article who may not be aware that this is the case in the ending) instead of being STRATEGICALLY CONDUCTED BY GENERAL PRINCIPLES. Again, do you actually play chess? And if so, have you ever studied the endgame?
skip "skilled players" and "computers"
Important information, methinks. The distinction is illustrated by the current discussion, perhaps. In any event, it's hardly grounds to revert. More like reason to edit if you disagree with prose.
I have up to now had a wonderful time editing Wikipedia. I have never before been reverted in years of editing music pages. As a longtime chess player (50 years) I should have perhaps anticipated the prickly ambiance over here in Wikipedia chessland, but I don't like it and won't give of my precious time improving the section if this is what it is going to be like every time I tighten things up and remove the fuzzy, mystical B-player babble from some corner of the material in favor of an exposition of modern scientific chess. My unattainable goal, to which end I would like to participate, is to help raise all the Wikipedia Chess pages to the forensic quality exhibited by First-move advantage in chess (with which I had nothing to do but which is the best chess page, truly world class, that I have yet encountered in the whole collection!) . JacquesDelaguerre (talk) 03:19, 2 August 2011 (UTC)[reply]
It isn't possible for anyone to really think that
The chessendgame (or end game or ending) is a non-rigorous term referring to the stage of the game when the strategic patterns laid out in the opening and executed in the middlegame have not led to sudden checkmate and the cumulative effect of irreversible piece capture (unlike chess's Japanese cognate shogi in which pieces return) has resulted in a reduced number of pieces remaining on the board in a position progressively more subject to concrete calculation by highly skilled players and especially by chess-playing computers.
is an improvement over the previous text
In chess and chess-like games, the endgame (or end game or ending) refers to the stage of the game when there are few pieces left on the board.
Really. Honestly. Truly. Seriously. That is absolutely atrocious; maybe the worst writing I've ever encountered on wikipedia. Suncreator, I refuse to believe that you think that's an improvement. I'm asking you to be specific in explaining how that edit is "excellent" and "better in almost all ways". I'm reverting again. Quale (talk) 04:41, 2 August 2011 (UTC)[reply]
Your missing the good points of this edit and focus on the bad parts. It needs copyediting and cuts for sure, but it does raise both the impreciseness of the term and the irreversible of exchanges(not in article) two things missing from the original. Also the original is dubious because it claims few pieces and at times endgames contain more then a few pieces, indeed there are examples in the article with five piece on the board and I suggest there can be more. Regards, SunCreator(talk)11:22, 2 August 2011 (UTC)[reply]
The issue of the impreciseness of the term has an entire section in the article - "The start of the endgame". And that is summarized perfectly well in the lead with the sentence that starts "The line between middlegame and endgame is often not clear..." And I don't see what "irreversible exchanges" has to do with it at all in anything close to standard chess. Bubba73You talkin' to me?14:39, 2 August 2011 (UTC)[reply]
I agree on impreciseness as I missed that was already further down the lead. "irreversible exchanges" or whatever term it found does want to be covered in the article. There is no reason to assume the reader is aware that a chess endgame is irreversible, indeed with the promotion of pawns to pieces it may easily thought not to be irreversible. Regards, SunCreator(talk)16:27, 2 August 2011 (UTC)[reply]
In short, that paragraph doesn't even come close to telling the reader what an endgame is. I know I'm beating a dead horse, but it mentions "sudden checkmate in the middlegame". What is that? What about a gradual checkmate in the endgame? What if a person resigns in the middlegame or his flag falls? Bubba73You talkin' to me?02:26, 3 August 2011 (UTC)[reply]
And I ran it through Microsoft Word's Flesch–Kincaid readability test. The Flesch-Kincaid "reading ease" is 0.0. The Flesch-Kincaid grade level is 37.8. A grade level of 12 corresponds to a 12th grader, etc. It is unreadable. For comparison, the current lead section has a F-K reading ease of 59.1 (60 is easily understandable by 13- to 15-year-olds) and a grade level of 9.5 (between 9th and 10th grade level). Bubba73You talkin' to me?03:24, 3 August 2011 (UTC)[reply]
I find it really difficult to believe that you actually think that edit to chess endgame was "excellent", and "better in almost all ways", but I invite you to explain exactly what you prefer about it at Talk:chess endgame#recent edits. The primary problem isn't the chess (although there are some issues there), it's the writing. Quale (talk) 04:52, 2 August 2011 (UTC)[reply]
Agree on both those lasts points, which is why I reverted! Just because you don't like the writing it no reason to revert, if it's copy edited you'll end up with a better article. The edits made highlighted many issues in the article. You are failing to assume good faith and your actions are WP:BITEy to this editor who it new to the chess articles. Regards, SunCreator(talk)11:22, 2 August 2011 (UTC)[reply]
This is the English Wikipedia - writing is supposed to be in English, not Klingon. I think that all of the useful stuff that was in that 90-word sentence is in the article in a readable form. My daughter has been playing chess for nearly 8 years. I'm going to see if she can understand that sentence. Bubba73You talkin' to me?14:33, 2 August 2011 (UTC)[reply]
(To SunCreator) I think I did fix the problem. You don't improve an article by trying to wordsmith an edit like that one. The article was instantly improved by WP:UNDO, and this is not disproportionate for an edit of that quality. I don't think I failed to WP:AGF, although I did wonder at first if the edit was some kind of a joke. Now I think it's more likely to be an issue of WP:COMPETENCE.
You've said that the edits highlighted "many issues" with the article, but the only specific example you've mentioned so far is the "cumulative effect of irreversible piece capture". I agree with Bubba73 that this is just a consequence of the Laws of Chess, but if you feel strongly about it, we can try to find chess books that highlight this in conjunction with the endgame. If they don't, I'm not certain it's worth mentioning here, but we can certainly still discuss it. What are the other issues the edit highlighted? Quale (talk) 00:12, 3 August 2011 (UTC)[reply]
The importance of endgame theory and its finiteness
There is a short paragraph in the lead that I think isn't very good:
Many of the greatest players throughout history have considered the endgame to be of paramount importance because endgame theory is finite. Whereas chess opening theory changes frequently, giving way to middlegame positions that fall in and out of popularity, endgame theory always remains constant.
Anyone want to take a crack at that one? Not only do I not like the wording, I don't think it's very accurate. "Finiteness" and "remaining constant" don't have much to do with the importance of the endgame. The lead should summarize the rest of the article and I don't see where that is taken up in the body. Maybe it should just be removed. Quale (talk) 03:43, 3 August 2011 (UTC)[reply]
Stronger could mean the side with material or positional advantage or the stronger(higher rated) player. As such it's use in the article makes for ambiguous reading or misreading. Regards, Sun Creator(talk)05:54, 14 February 2013 (UTC)[reply]
Checking the last diagram - two bishops vs knight - under the "Effect of tablebases" headline with the shredderchess online database, it says "win in 57" (which is to win, not to convert). I presume that the article's "wins in 45 moves" means "captures the knight in 45 moves" (I did find a capture in 48).
Furthermore, compare this to the phrasing "(The fifty-move rule is not taken into account in these studies.)"; one can of course take that into account with distance-to-conversion databases, so: could someone clean up and find which examples are actually won with the 50-move rule in place and which examples are "hypothetical wins" in the absence of the rule? — Preceding unsigned comment added by 129.240.49.40 (talk) 09:00, 1 December 2014 (UTC)[reply]
I've changed wins in 45 to wins in 57. I don't see why DTC databases databases would necessarily give correct results either, but the Syzygy DTZ50 databases should. Please feel free to investigate and correct. Martin Rattigan (talk) 15:41, 19 May 2016 (UTC)[reply]
As you say, in the given situation the knight can be captured in the kbbkn position mentioned within 50 moves, so the position will classified as won by any type of EGTB. 129.240.49.40's query regarding this position was about the number of moves required to mate. The number given was probably taken from a DTC EGTB, which gives the minimum number of moves required to mate or convert to a winning position in a different endgame (in this case by taking the knight). In fact, because the capture of the knight can be achieved faster than mate in the absence of the capture, a DTC EGTB will return the number of moves required to capture the knight. The correct mating distance is 57 as 129.240.49.40 states.
The second part of 129.240.49.40's query regards the classification of all positions on the site as won, drawn or lost within the 50 move rule. For almost all positions this would depend on the number of plies since the last pawn move or capture of a piece. The use of a DTC database would not solve this, though the number of moves required to mate can sometimes be more accurate than the DTM figure if the rule is taken into account (see diagram right), but the opposite is normally true. The classification of positions is the same whether the EGTB is DTM or DTC. I believe the Syzygy DTZ50 databases would correctly classify positions as won, drawn or lost with the 50 move rule taken into account for particular values (or ranges) of the ply count, but I think there also the number of moves required to mate would not necessarily be correct. What is needed for correct move counts under the rule are DTM50 EGTBs (not currently available).
In the meantime FIDE have introduced a mandatory (as opposed to claimable) 75 move rule for competition play as of 1 July 2014, which may render some of Wikipedia's analysis (together with that in any publications prior to the FIDE publication) invalid in relation to competition play. Any move with a ply count greater than 150 is illegal in such a game (and any move from a position where there is no possible mate within the remaining 150 ply count limit).
Incidentally, I believe there are constructed DTC EGTBs that incorrectly assume a winning conversion must be a promotion or capture by the winning side. In this case they would also give mate in 2 in the diagram. — Preceding unsigned comment added by Martin Rattigan (talk • contribs) 13:00, 20 May 2016 (UTC)[reply]
I inserted a citation required flag for the sentence, These are positions in which one side has only a king and the other side has one or two pieces and can checkmate the opposing king, with the pieces working together with their king. There is no general agreement on what constitutes a basic endgame and it is not Wikipedia's job to establish a standard meaning. Fine Basic Chess Endings 1941 includes an endgame where the stronger side has three pieces and Müller and Lamprecht Fundamental Chess Endings 2001 includes an endgame where neither side has only a king. Other sources can probably be found that correspond to none of these; I seem to remember Teach yourself chess including kqkr, but I may be wrong - It's a long time since I saw the book. Any classification in Wikipedia should adopt a neutral point of view, giving several differing sources.
Martin Rattigan (talk) 22:21, 28 July 2015 (UTC)[reply]
a
b
c
d
e
f
g
h
8
8
7
7
6
6
5
5
4
4
3
3
2
2
1
1
a
b
c
d
e
f
g
h
White to play and mate in 1
Added citation required flag to both sentences in If the pawn is blocked by a knight on or behind the Troitzky line, the knights have a long theoretical win. There are some other positions when the pawn is past the Troitzky line in which the knights can force checkmate, but the procedure is long and difficult. The knights do not necessarily have a win in the first case and in neither case is the win necessarily long or difficult. As here.
I replaced the discussion of two knights endgame to bring it into line with the main article..
I also reverted Bubba73's amendment because it's misleading
Bubba73 changed "... if the weaker side also has other material ..." to "... if the weaker side also has a pawn (or in rare cases, a knight or more than one pawn) ...".
(i) The White winning positions in knnkp for wtm/btm constitute 31.31%/13.56% of total wtm/btm positions. The White winning positions in knnkpp for wtm/btm constitute 29.52%/11.55% of total wtm/btm positions. There seems to be little reason to preface the latter, but not the former, with the phrase "in rare cases". Numerically there will, of course, be considerably more of the winning two knights versus two pawns cases.
(ii) There would seem to be no reason to single out, or mention, a knight. Winning chances with two knights against a knight are of a similar order to two knights against a bishop. In both cases they constitute less than 0.1% of the total positions with either side to move. The longest win in knnkn is 7 moves. Sporadic short depth winning positions are trivial to construct for two knights against very many combinations of opposing pieces (e.g. diagram left). Troitzky constructed many long mates with a variety opposing material.
I have added the sentence "The winning chances with two knights are insignificant except against a few pawns". (Database statistics are not available beyond two pawns, but I would guess that winning chances could probably not be described as insignificant till at least beyond five pawns.)
Martin Rattigan (talk) 02:38, 1 October 2015 (UTC)[reply]
Added citation required flag to, A draw results if the defending king can reach the square in front of the pawn or the square in front of that (or capture the pawn). Black here can reach e8 but still loses.
A citation was given. I think this is probably a misquote or taken out of context, but I don't have the cited source.
The sentence is still incorrect. I have added a note and diagram to that effect.
Given that the text now states a rule and immediately says that it doesn't necessarily hold whatever file the pawn is on, I feel it would be best to delete both the original text and my addition and leave any such discussion to the main article (which appears to be more accurate).
Can somebody who has M&L's Secrets of Pawn Endings verify that the original text is as stated in the book please? Martin Rattigan (talk) 02:38, 1 October 2015 (UTC)[reply]
I also moved the last paragraph in this section to the end of the previous section since that was where it was obviously intended to go. — Preceding unsigned comment added by Martin Rattigan (talk • contribs) 04:57, 1 October 2015 (UTC)[reply]
Added citation required to sentences, An outside passed pawn can outweigh a protected passed central pawn, unlike king and pawn endgames.[citation needed] A knight blockading a protected passed pawn attacks the protector, while the knight blockading an outside passed pawn is somewhat out of action. Not obvious enough, or obviously a useful enough generalization to be included without any citation (even though it may be true in many situations). — Preceding unsigned comment added by Martin Rattigan (talk • contribs) 01:26, 29 July 2015 (UTC)[reply]
"Graham Burgess...does not cover the bishop and knight checkmate 'since it is too difficult to be regarded as a basic mate.'"
I was quite surprised to see so much unwarranted text in this section devoted to answering the arbitrary question:
What is a "Basic checkmate"?
Let's hope another author doesn't come along and write, "These are the Simple Basic Checkmates compared to the more Complicated Basic Checkmates." This whole section would then need to be reworked again. I'm kidding of course.
In addition, the section named "Basic checkmates" is somewhat of a misnomer since "checkmate" denotes a position while "checkmating" denotes a process. It isn't just the checkmates, but the process or technique needed to arrive at checkmate from a starting position with just a few pieces. Maybe this section should be retitled: "Limited Material Checkmating" and leave out all the nonsense about what constitutes a "Basic checkmate" since apparently there is no consensus for that term.
Oh, and I was almost going to suggest the following definition for "Basic Checkmates":
Any simplified endgame not covered by Dvoretsky's Endgame Manual 4th edition.
But then I found that he somewhat reluctantly included KBN vs K.
OK, it can be changed to "almost all sources...". It needs to make it clear that a "basic checkmate" (a term used in many books) means that it involves minimal combinations of pieces that can force checkmate. Bubba73You talkin' to me?04:29, 4 May 2016 (UTC)[reply]
I disagree. In your answer you have defined "basic checkmate" contrary to some of the cited sources. It can only be said that trying to pigeon-hole endgames concisely into specific general classifications is not feasible due to their very nature.
If you want to make a change as you suggest, it should be "All sources agree on three types of positions....", and the corresponding changes to the rest of the paragraph. The purpose of that first sentence is to state the "consensus" types. Inserting "almost" destroys the reason for that sentence.
What I really think should happen is this entire section should be rewritten without trying to indicate that some authors have arbitrarily defined "basic checkmates" differently, but having looked at some of the history of this section I doubt all others would agree.
I've amended the "final" version slightly as follows:
Added "of opposite colours" to kbbk. I know that most references include also the case with bishops of the same colour but these are not mates, so I would have said they could be ignored in the context of this paragraph. I'm not averse to these being reinstated, but then kbk and knk should also appear, because these are usually also mentioned in the same references.
Added "(or others)" to the intro to the second list. This is clearly true.
Added knnk to the secondary list. Not the most fascinating endgame, but it does have 1-ply mates and is covered in at least two of the references given at the start (i.e. the ones I have; BCE and M&L) and it also gets a section in Fine (I don't have Fine & Benko). I would expect it to be covered in any reference that includes knnnk or knnkp anyway.
Is that statement really true? Perhaps with the advent of 7 piece EGTBs it is true for all endgames with 7 or fewer pieces (where "pieces" includes kings and pawns). It certainly wasn't a true statement before the year 2000.
"Whereas chess opening theory changes frequently, giving way to middlegame positions that fall in and out of popularity, endgame theory always remains constant." — Preceding unsigned comment added by TommyWP (talk • contribs) 19:37, 4 May 2016 (UTC)[reply]
I agree. It's bollocks. The endgame references I've seen all take delight in showing the mistakes of previous authors and people like Troitsky come along. But feel free to alter the text if you think the effort is warranted. Martin Rattigan (talk) 12:30, 19 May 2016 (UTC)[reply]
This is an old observation that no one has acted on, but should we remove or rewrite the sentence about endgame theory remaining constant? I think I understand what is intended, but it isn't true in a literal sense and I think it is just confusing. Quale (talk) 05:03, 20 June 2018 (UTC)[reply]
I have removed the phrase "and generally it is" from the sentence, " This was thought to be a draw and generally it is, ...".
Two grandmaster pronouncements are given on this:
Nunn - The general result is undoubtedly a draw ...
Müller & Lamprecht - although nearly 90 percent of all of these positions are wins for the queen, it is generally a draw if the king is not separated from the knights and they are on reasonable squares
These are very different statements and Wikipedia should not be taking sides in the matter. Müller and Lamprecht's statement is more guarded and cannot be taken to mean that the ending is generally drawn.
In fact though M&L is no doubt correct (given a suitable interpretation of reasonable squares), it's a little like the Scottish crofter who, when asked the way to Aberdeen, replied, "Och, if I were going tae Aberdeen, I wadnae start fra here". Your opponent, should you reach the wrong side of this ending would probably be unsympathetic if you asked him would he mind awfully if you moved your knights nearer the king and shifted them all to reasonable squares.
I would personally say that Nunn at least overstates the case.
Nalimov gives the results of the position left as won for Black. Neither White's a8 knight nor his king are close enough to stop the pawn and the black king prevents the f1 knight from doing so. This means that White must find a drawn KNNKQ position for his pieces in the three moves it takes for the black pawn to promote. But this can't be done. There are no such positions that White can reach with any three moves of his pieces.
This example doesn't appear to be isolated among KNNKP positions where White cannot stop the pawn.
Of course if the White pieces cannot stop the pawn this suggests they may be in unreasonable positions within the meaning of M&L's statement. But that brings us back to the Scottish crofter.
I also replaced, "On the other hand, Batsford Chess Endings states that 89.7 percent of the starting positions are wins for the queen (Speelman, Tisdall & Wade 1993:7)", with, "On the other hand 73.44% of positions are won by the queen.<ref name=wilhelm>((cite web|url=http://www.geocities.com/rba_schach2000/overview_english.htm |title=Chess program Wilhelm |deadurl=unfit |archiveurl=https://web.archive.org/web/20081208015339/http://www.geocities.com/rba_schach2000/overview_english.htm |archivedate=December 8, 2008 )) + ((cite web|url=http://auto-chess.blogspot.ch/2012/11/download-nalimov-end-game-tablebases-2.html|title=Nalimov Engame Tablebases|website=AutoChess))</ref>".
The BCE statistics are White to play only, with a target of either mating or winning a piece. The replacement is either side to move with a target of mate. (Nevertheless the statistics as Nunn & M&L point out can be misleading.)
A recent rephrasing of the sentence describing exceptions introduced a couple inaccuracies, so I reverted.
"bishops on opposite color with other pieces – the stronger side should avoid exchanging the other pieces" => "the stronger side should avoid exchanging pieces if bishops are on opposite colors" -- this is not correct since exchanging the opposite colored bishops would generally be good, the rewrite lost the "with other pieces" bit that is important
"when all of the pawns are on the same side of the board, often the stronger side must exchange pawns to try to create a passed pawn" => "when more pawns are on the same side of the board, the stronger side should exchange pawns to try to create a passed pawn" -- not sure what "more pawns are on the same side of the board" is supposed to mean, but this isn't quite correct either. Quale (talk)
I explained the reason I reverted in the General considerations section above. I agree with you that the sentence can be improved, but since the edit introduced at least two inaccuracies it needed to be undone. Unfortunately it can be hard to improve the English and logical flow of a sentence on a technical subject. Often the words have been chosen with some care to be accurate, and the important details can easily be lost in a rewrite since they can be hard to perceive unless you have some expertise in the subject matter. We can work here on the talk page to try to find a better phrasing that remains correct. Quale (talk) 04:58, 20 June 2018 (UTC)[reply]
Lucas chess has a formula for game phase (I believe lichess uses a very similar one?) Can I include a simplified version of the formula in the article? Alshfik (talk) 13:39, 17 August 2021 (UTC)[reply]
Sorry I didn't reply earlier. I'm not sure that formula is really relevant for this page, it's certainly not anything a human would ever use when playing or analyzing a game. Also, how is the reader to interpret the parameter m (total material)? The formula could be relevant on a page for the Lucas chess software, if such a page existed. Quale (talk) 05:17, 19 August 2021 (UTC)[reply]
I agree with User:Wereon here. The underpromotion of a knight, something which virtually never happens, cannot be part of a "basic mate". A basic mate is one that we are all taught as children when we start learning chess, ie Q v K, R v K, 2R v K, 2B v K, and B+N v K. Ok the latter two don't happen very often in practice but are at least plausible. The totally theoretical three knights is not, and should not be included in this article. Pawnkingthree (talk) 00:00, 6 September 2021 (UTC)[reply]
I agree. If no one can point to a game where such a thing has actually happened, this mate is little more than an interesting thought exercise and not anything to do with the game of chess as it is actually played. SaltySaltyTears (talk) 00:09, 6 September 2021 (UTC)[reply]
The definitions of "virtually never happens" and "at least plausible" being used here are unstated and, I assume, rather loose. The article itself doesn't try to decide what "should" be basic, but apparently goes by the criteria of looking at sources. I admit I had not noticed K + 3N vs. K in my old, now lost, copy of Fine, but if it's there, well, Fine is certainly the acme of English-language sources.
I got an opinion on the Chess project page before restoring it, this time with a reference to Fine. The 1941 edition of Fine is the primary source. The 2003 edition has only a very brief mention of it under "elementary mates".
I've had it in informal games - intentionally getting to just three knights.
Bruce is right - the article doesn't decide what to include, it lists what several writers have included, which includes Q+R, etc, in some cases.
Fundamental Chess Endings by Muller and Lamprect lists 3N vs B and 3N vs N (under pawnless endings), but not 3N. These require up to 91 and 86 moves, resp. They must, in cases, get to 3N vs. K. Bubba73You talkin' to me?01:23, 6 September 2021 (UTC)[reply]
Looking at the section some more, I think it needs some attention:
Citing 11 endgame books without quoting or paraphrasing any of them is silly. Whatever citations are for in Wikipedia, this isn't it.
Using sources to define what is "basic" and what isn't doesn't work very well. Of course, if you've been editing Wikipedia long enough, you want to use sources for everything, but that has problems here. We have more than 11 sources, most of them reliable, and they disagree with each other about what is "basic".
I guess we should come up with our own definition(s) of what is "basic" and work with them. For instance, one definition would be "endgames with king and some pieces vs. king". The textbook I started with, many years ago, taught KQ/K, KR/K, KBB/K, and KBN/K as "basic". In talking about these, we would also mention that:
KB/K, KN/K, and KNN/K are drawn.
KNNN/K is a win, but can't occur without a pawn promotion.
Note that this definition of "basic" leaves out KP/K and KNN/KP. The former is in some sense "basic" in that you really have to know it. The latter is more of a curiosity, but of course, it's famous. We should mention both, and should explain what their significance is.
I like your ideas. The authors disagree about which ones are fundamental/elementary/basic. Some don't include KBN/K. Some include that but don't include KBB/K. We could say which books have which endings, but I don't like that. Why don't we use editorial judgement and (1) list KQ, KR, KBB, KBN as basic, and (2) have another section saying (a) KB/K, KN/K, and KNN/K can't force checkmate, (b) have ones like 3N, Q vs. R and NN/P. Bubba73You talkin' to me?02:16, 6 September 2021 (UTC)[reply]
I don't think it hurts and it gives information about how many people are going to the page. It is interesting that it was ~ 200 a year ago then got to nearly 1,000, then tailed off. Bubba73You talkin' to me?06:12, 14 September 2021 (UTC)[reply]
The shape of the curve gives you an idea of the effect of The Queen's Gambit. But that's an awful lot of screen real estate to use for a small item of information that isn't really relevant to the topic of this article. Bruce leverett (talk) 15:35, 14 September 2021 (UTC)[reply]
"The winning chances with two knights are insignificant except against a few pawns. " (I can't access the database.) But this sentence sounds like the two knights have significant chances against several pawns. Generally 2N vs P is a win if the pawn can be blocked on or before the Troitsky line. Bubba73You talkin' to me?06:27, 14 September 2021 (UTC)[reply]
I agree that this is problematic. By the way, the lead section of Two knights endgame has the same problem, though that article has a later section that makes things more clear. KNNvKP is famous, thanks at least partly to Troitzky; all other KNNvK<any> endings, if there are any interesting ones, are obscure, and we shouldn't mention them unless we have to. Bruce leverett (talk) 15:32, 14 September 2021 (UTC)[reply]
The lead is messy, weirdly talks about "other games" when the title is chess endgame (and other games are mentioned nowhere in the article), uses subjective phrases, and has a strange structure. I am not certain why my attempts to improve it were reverted. @Jasper Deng
@Jack234567: You were introducing a clear grammar error in the first sentence: "Chess endgame, better known as just endgame or ending, is the final stage of a chess game which occurs after the middlegame". This is missing an article. This is the main reason why you were reverted. Also, endgames definitely do exist in other chess variants, such as xiangqi and shogi. Also, changing "most" to "many" in "Most endgames are not solved, and textbooks teach useful strategies and tactics for them." changes the meaning, since most is not the same as many. The changes to the second paragraph are not improvements either, since the increasing importance of pawns is not the only change that comes with the beginning of an endgame, so the existing language that emphasizes general strategic differences is better.--Jasper Deng(talk)10:05, 24 October 2023 (UTC)[reply]
But why does the first sentence talk about other games when the article is only about chess endgame and not chess variants endgame? I am not sure which article is missing either.
I think "many" is more accurate than "most" because many endgames have been solved.
I am not sure why the lead talks about what "should" happen, instead of what actually happens.
I would like to make a few other changes as well. The claim that most tournament players can't mate with bishop and knight is just false. The longest forced win section is outdated; the new record was discovered more than two years ago.
Your complaint about mentioning chess variants in the first sentence, but not elsewhere, is reasonable. One could strike the phrase "and other similar games" from that sentence. It would probably be a good idea, however, to mention the chess variants in another place, in the body of the article.
The article claims that "most endgames are not solved", and this is likely true, but there is not a reliable source to support it, so the question is whether we can get away with that. Just my opinion, I think it's OK to say that. Do you not believe it? The "official" solved endgame databases go up to just a few pieces total, but that has to be a tiny percentage of all actual endgames (even excluding endgames that would never happen, such as quadruple pawns, etc.).
The reason the lead says that the king "should" stay hidden (during the early stages of the game) is that players often fail to hide it. Is there some better way of phrasing this that you have in mind?
We don't claim that most tournament players can't mate with bishop and knight, we claim that many players can't. As a matter of fact, I am pretty sure the most would be true here, but again, I would not expect to find a reliable source to back this up. Or do you believe that it's not true? In my experience, most tournament players are at the scholastic level; and even in adult tournaments, I have seen some very strong players who did not know that mate. Bruce leverett (talk) 23:11, 24 October 2023 (UTC)[reply]
Ok, I've been making a lot of changes to this article. I believe its much better now, but please take a look over it and let me know if you or anybody else have any objections. Jack234567 (talk) 05:34, 30 October 2023 (UTC)[reply]