Talk:First-move advantage in chess

Problems with the recent NCO edit on Nimzovitch-Larsena and Owens

I have several issues with the recent NCO edit on the Nimzovitch-Larsen and Owens. Most importantly it ascribes views to Nunn that aren't explicit in his writing, based on the editor's interpretation of the symbolic evaluation of a very small number of lines in NCO. (Each opening gets only a single column [although NCO layout uses rows for variations], with a small number of side variations.) I think the editors interpretation is unwarranted. Nunn's evaluation of Owen's is "White is slightly better". White is slightly better is the most common evaluation in the book, and Black frequently adopts variations in which White is slightly better for reasons that are obvious to any experienced chess player. Translating +/= to "leaves black at a disadvantage" is very misleading in this context, and I think wrong. If addition is to remain (and I think it should be removed), the language must be changed to match precisely what Nunn indicates (White is slightly better), not the editor's spin. There are other issues too, such as calling Nunn's 15-year-old work "the modern view". It's a single more modern view, but that isn't enough evidence that it is the modern view. Staunton's opinion is attributed to Staunton, but symbolic evaluations of two obscure lines in a book that devotes little time to either are elevated to the modern view, which is also wrong. Finally evaluations of openings can change frequently, and I do not agree with editors scanning down the columns of NCO or MCO or similar sources and then making pronouncements on entire openings based on the evaluation symbols they find at the end. This is basically an original research or undue weight issue. As a final side note, I was disappointed that when I asked for page numbers in the cite the editor restored his edit without including the page numbers in the reference. I thought that to be rather rude. Quale (talk) 15:50, 14 June 2015 (UTC)

I don't think it is inaccurate to call 1.b3 = and 1.e4 b6 +/= the modern view. How many sources do you want to back up this tangential point? There are references given in the Owen's Defence lead for starters. Cobblet (talk) 00:38, 15 June 2015 (UTC)

"The current view" would be a more NPOV phrasing, "current" doesn't have the emotional implications of "modern". Also agree that "slightly better for White" is a more accurate reflection of the source. MaxBrowne (talk) 01:28, 15 June 2015 (UTC)

That isn't what the edit says, Cobblet. The source says "White is slightly better", the edit says "black is at a disadvantage". That isn't even close to an accurate reflection of the source. Second, the edit claims the modern view is different than Staunton's view. I think that's simply incorrect. Staunton did not say Black was equal in Owen's, he said "playable". As far as I know, +/= is often playable for Black. In this specific instance, I think Owen's qualifies as playable. If Black were not able to play any opening that might leave White with a slight advantage, there wouldn't be many defenses left. Third, Staunton also said 1.b3 is worse than other lines for White because it is essentially defensive, and I think that is precisely the modern view as well. NCO couldn't possibly be used to rebut the opinion that 1.b3 is defensive, because "=" says absolutely nothing about whether an opening is defensive, aggressive, neutral, or anything really. So the source doesn't contradict Staunton at all, despite the way the editor slanted the edit. Finally, unlike Staunton, NCO does not compare Nimzovitch-Larsen and Owen's directly at all. It's pure WP:SYNTH to concoct a comparison between these openings that the source does not make. The edit is poor. If you want to know how many sources I want, how about a source that 1) directly compares the two openings in question rather than being the product of editor synthesis, and 2) actually says something qualitative about the openings rather than the essentially quantitive and bloodless = or +/=. Quale (talk) 01:44, 16 June 2015 (UTC)

I think MaxBrowne's latest edits (the ones made 24 hours before your latest reply here) have made the sentence more neutral in tone, and I believe both parts of the sentence as it stands now are verifiable and putting them together does not constitute WP:SYNTH. I agree though that NCO is not an appropriate source for either part of the sentence. Cobblet (talk) 11:31, 16 June 2015 (UTC)

Max Browne's edits were made an hour after your odd initial comment, so unless you were commenting on what you knew would be the future state of the article rather than the edits that existed at that time, I don't understand what you're complaining about. Forgive me if you can or condemn me if you must, but I read your comment as applying to the article that existed at the time you made it, not the article an hour later after another editor smoothed some rough edges. I still think that the insertion doesn't improve the article at all, although it isn't terribly harmful. Not every verifiable statement is helpful or appropriate at every point in an article, and it can be a synthesis violation to combine two true statements (1.b3 = and 1.e4 b6 +/=) into a claim that the source doesn't make (NCO doesn't say that Staunton's view is incorrect.) The edit uses synthesis to imply that the modern view is different than Staunton's, and I don't think this is true. Even if it is true, you simply can't demonstrate that by looking at variation evaluations in NCO. I think sometimes Wikipedia editors try too hard to rescue bad edits. Sometimes the articles are better when bad edits are simply be removed entirely. Quale (talk) 13:16, 16 June 2015 (UTC)

My initial comment was in response to you questioning the facts presented, which I believe are verifiable; I did not dispute your other concerns with regards to NPOV and SYNTH. And I'm not complaining about anything, just pointing out that some of those concerns have been resolved since the initial edit. As it stands right now, I don't think there's any implication that the modern view is markedly different from Staunton's. Cheers, Cobblet (talk) 21:42, 16 June 2015 (UTC)

Thanks to all those who have contributed to this discussion. There are some important points that should answer the sometimes slightly misguided questions that arose.

Firstly, NCO (and its predecessors such as Korn) is the analog of an encyclopedia to chess openings. MCO is another (multi-volume) example, but not on my shelf. Its contents are largely a summary of the conclusions of thousands of books on chess openings, supplemented by a database of all master games.

What NCO (i.e. its four authors) "says" is the evaluations it attaches to the lines. A strong chess player will realise that English is not necessary.

Secondly, considerable experience and use of NCO and others over the years means I realise something that some people here don't: that the large majority of critical opening lines are assessed by Nunn and his colleagues as equal. What this means is that although many lines give advantage to white, there is in most cases some choice of variations that gives black equality against all white alternatives. In some cases, black needs to tread a very narrow path, but it is only the dubious opening lines that leave black worse by force (in the sum total of the lines studied). [Checking this is made easier because the footnotes never overturn the analysis in the main lines in NCO: they rather offer alternatives that lead to the same or a worse evaluation for the side varying]. This is also what is found when all lines are analysed with modern engines.

As a result, the fact that black has no choice of lines in the Owen's where he reaches full equality (against best play by white) is significant, and identifies why top players do not consider the Owens a reliable defense. Elroch (talk) 13:52, 16 June 2015 (UTC)

External links modified

Hello fellow Wikipedians,

I have just modified one external link on First-move advantage in chess. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

• Added archive https://web.archive.org/web/20090325220009/http://www.spectrum.ieee.org/print/5379 to http://www.spectrum.ieee.org/print/5379

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

• If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
• If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 19:45, 6 October 2017 (UTC)

The Lede v The Article...

The lede opens with the sentence, "The first-move advantage in chess is the inherent advantage of the player (White) who makes the first move in chess," which is a definitive declaration that White has an inherent advantage. However the article which follows then makes it quite clear that there is no definitively proven, inherent advantage for White, with many theorists and players - some of them former World Champions and 2 of them the greatest players of all time - disagreeing with the conjecture. The second sentence immediately starts to back down from the initial declaration by saying that "Chess players and theorists generally agree that White begins the game with some advantage." We've gone from a definitive, inherent advantage for White, to a wishy-washy general agreement about "some advantage." Even that is deceptive though because the implication that all chess players and all theorists agree on this advantage is patently false; as a full reading of the article demonstrates. The third sentence backs away even further and makes an unproven, cause and effect claim that the supposedly definitive, inherent advantage for White is proven by the statistics when in fact:

1. The statistics can be interpreted in a different way - as is invariably the case with statistics - to show that this definitive, inherent advantage leads to White winning a little more than a third of the time; which certainly doesn't support the claim of a definitive, inherent advantage at all.

2. As shown in the main body of the article, there are other possible explanations for the statistics, such as the definitive, inherent advantage for White being nothing more than an unfounded, self-fulfilling prophecy which is nothing more than players believing White has an advantage and having that unfounded bias affect their play.

The bottom line here is that it is by no mean proven that White has a definitive, inherent advantage and the article should reflect that, rather than starting with the assertion that White has an advantage - as though it is a fact, which it is not - and presenting contrary views as being dissenters; which they are not. You wouldn't start an article about Vaccination with a claim that it causes autism, then present the evidence that it doesn't as a dissenting view; an example I use because Vaccination is an excellent example of statistics being used to prove a cause and effect relationship where none actually exists, exactly as the statistics are used here to 'prove' White's advantage. Neither should this article - or any article for that matter - start by presenting an unproven claim as true, then provide evidence to the contrary as a dissenting view. If a conjecture is unproven then it should be presented as such, with evidence for and against it then presented equally so that the reader can make their own determination. Doing otherwise, as is the case with this article, immediately predisposes the reader to believe the unproven claim is factual, then biases the evidence so that the dissenting view must provide stronger evidence just to be equal; which is ironic, given the nature of the article!

;-)

A patient, dispassionate, objective reader will come away from this article with the understanding that there is no definitive proof that White has an advantage in Chess. However a less patient and objective person who only reads the first few sentences will come away with the erroneous viewpoint that White has an advantage in Chess. That viewpoint is unproven and therefore wrong and, as such, the article needs to change. FillsHerTease (talk) —Preceding undated comment added 10:47, 4 December 2017 (UTC)

I intend to start changing the article to reflect that White having an advantage is an unproven conjecture - with arguments on both sides - if no one objects. I don't want to waste my time if it's just going to be reverted obviously, so I will leave this for 1 week and if no one has objected I will start making updates after that. FillsHerTease (talk) 13:51, 7 December 2017 (UTC)

I agree. The lead did appear to over-state the case that White has a definite advantage, at least when judged by the available references. For the time being I slightly rephrased the opening statement, and added two citation-needed tags.—LithiumFlash (talk) 03:16, 25 December 2017 (UTC)

I don't agree. The article was better before your changes. The claims in the lede are more than adequately referenced in the detail sections of the article. Quale (talk) 05:45, 25 December 2017 (UTC) BTW, "drawn with best play" and "White has an advantage" are not mutually exclusive claims or in conflict with each other. The article explains this in detail. Quale (talk) 05:47, 25 December 2017 (UTC)

This article should not have statements synthesized by Wiki editors, and bold statements require reliable references. The article currently does not have any references to support the statement "...the consensus has been that a perfectly played game would end in a draw."—LithiumFlash (talk) 12:47, 27 December 2017 (UTC)

Really? Did you read the article? The First-move advantage in chess#Drawn with best play section has citations saying chess is a draw from Steinitz, Lasker, Capablanca, Fischer, Fine, Rowson, Adjoran, and Watson (also quoting Kasparov). If you're not going to read the article you should not edit it or tag it. Quale (talk) 07:03, 29 December 2017 (UTC)

The article expresses a sampling of other opinions but it is certainly not a consensus. The lead should not state that there is a consensus (chess being a draw) when there are already sections in the article expressing other possible endings of perfect play. The lead should be both concise, and accurate. It is very misleading to tell readers that there is consensus when in fact there is none.—LithiumFlash (talk) 05:36, 30 December 2017 (UTC)

I'm afraid you don't understand what "consensus" is. Consensus does not require unanimity, and opinions held by a small minority do not prevent consensus. Also, you are just wrong here. Adams and Berliner thought White had a decisive advantage, but they knew that their opinions were in the minority and were against the consensus that the game is probably a draw. They recognized that there was a consensus, they didn't agree with it, and so they said so. Quale (talk) 05:53, 30 December 2017 (UTC)

And while I'm here, Stack Exchange is not a reliable source for claims about chess. It's essentially a wiki and has the qualities of a self-published source. Stack Exchange answers may cite published sources to bolster the writer's argument, and those sources may be useful in Wikipedia. But really I wish drive-by editors would stop trying to fix this article because it ain't broke. Nearly every "improvement" to this article in the past four years has made it worse. I'm not saying the article is perfect and can't be improved, but virtually none of the people who have felt compelled to put their stamp on this article is as knowledgeable about the subject or as skilled a writer as the original author. The result is that their help has not helped.

Also, once again, please read the article before trying to improve it. There is already a section in this article on solving chess, and it is titled "Solving chess". Astonishing, I know, you could never be expected to find that. That section is well cited. The lede is supposed to summarize the article, so don't put new claims in the lede. Look to the article body and summarize it in the lede. If something about solving chess needs to go in the lede, it should be a very brief summary of the most important parts of the Solving chess section. But while we're here, the lede is already a good summary of the article and it doesn't need to say anything about solving chess. First, it's obvious in context that chess hasn't been solved or there would be no debate over whether White or Black has an advantage. The mere existence of this article implies that chess hasn't been solved. Also, we already have a whole article for that: see Solving chess. Quale (talk) 05:54, 30 December 2017 (UTC)

LithiumFlash, it's beyond ironic that you would tag the article as having unreliable sources. The only unreliable source that was in the article was stack exchange, and you were the one who added it. The article was better before you insisted on trying to "fix it". Quale (talk) 05:57, 30 December 2017 (UTC)

Agree w/ Quale in all respects. --IHTS (talk) 01:06, 31 December 2017 (UTC)

Chess has two players, White and Black

If the article has a section on perfect play, then it should explain the status of play for both White and Black (i.e. who wins, or a draw, or unknown). To not mention this is a serious deficiency in the article. I've elaborated slightly in this respect with references. (The other option is to remove any discussion of perfect play, since that is not the main theme of the article anyway).—MeixiangKazuki (talk) 07:34, 31 December 2017 (UTC)

It has been explained many times that Stack Exchange is not a reliable source, consisting as it does of user generated content, yet you continue to edit war and attempt to insert it into the article. It is unclear what the relevance of your other source (the Ph. D. paper by Victor Allis) is; I have examined it and see nowhere that it addresses the question of what the result of a perfectly played game of chess would be. 222.153.250.135 (talk) 08:29, 31 December 2017 (UTC)

Not only that, MeixiangKazuki's post above makes no sense. (He says discussion of the result of perfect play is missing from the article, that's untrue, then he says to remove said discussion.) --IHTS (talk) 11:32, 31 December 2017 (UTC)

Please read my comments again carefully. It reads, "If the article has a section on perfect play, then it should explain the status of play for both White and Black..." and also "The other option is to...". It does not say what you purport it to say. Your work on this article is certainly not being conducted in good faith.—MeixiangKazuki (talk) 13:48, 31 December 2017 (UTC)

The article Solving chess is actually more directly related to the topic of perfect play. In that article, the very first sentence in the lead paragraph states "Solving chess means finding an optimal strategy for playing chess, i.e. one by which one of the players (White or Black) can always force a victory, or both can force a draw." (bolding added)

Neither player (White or Black) is excluded from the article there, and there is no debate about it. It is properly sourced, as I have also properly sourced it here.

Since Solving chess is the prime article which discusses perfect play, the content in this article (on this subject) should match the text there (but perhaps be stated more concisely), and without elaboration.

Due to the relation of the these two articles, I also added "Solving chess" in the "See Also" section of this article.—LithiumFlash (talk) 15:33, 1 January 2018 (UTC)

Yet another disruptive edit

For goodness sake please stop inserting Stack Exchange! We do not source anything to forums or wikis per WP:UGC. Also, "consensus among many" is a tautology. Consensus does not mean unanimity, it means a clear majority. It is an accurate statement that the consensus among strong players and theorists is that a perfectly played game of chess would end in a draw. The Allis source does not address this question, and it's unclear whether the Nowakowski source does either (no page references are provided). Only the Shannon paper is relevant, and on page 3 he agrees that the game is most likely a draw. 222.153.250.135 (talk) 00:59, 1 January 2018 (UTC)

Yes, we have tried to explain that repeatedly but they won't listen. Also do not insert new claims into the lede as the lede must summarize the most important points from the body of the article. Read MOS:LEDE and note its important instructions, including "Apart from basic facts, significant information should not appear in the lead if it is not covered in the remainder of the article." The issue that you seem to have with the article hardly qualifies as one of the "most important points", so I don't think it belongs in the lede anyway.

Also these disruptive edits to the lede have now been reverted by three different editors. At this point I think if you want to change the lede your best approach would be to propose the change you think should be made on this talk page and explain why you think it would improve the article, then see if other editors have any comment on the change before you edit the article. The edits have been kind of all over the place. First the claim on this talk page that "the statistics can be interpreted a different way" (that White doesn't have an advantage, which is an idiotic argument), then repeated edits that seemed to dispute the idea that the consensus view is that chess is probably a draw, then repeated edits added stack exchange as a reference even though it isn't a WP:RS, then an "unreliable sources" tag was added to the article even though no unreliable sources were identified, and finally it seems that there's some attempt to explain what "perfect play" means again using stack exchange as a reference (although I'm actually not sure what the point of the recent edits is supposed to be). Quale (talk) 05:47, 1 January 2018 (UTC)

Also Shannon's opinion on whether chess is most likely a draw isn't of very great import. Shannon wasn't a strong chess player and has no reputation in chess theory. Steinitz, Lasker, Capablanca, Fischer, Kasparov, etc., have that covered and we don't need more sources in favor of this proposition unless they are recognized as world championship strength or noted experts in chess theory. Quale (talk) 05:58, 1 January 2018 (UTC)

The Stack-Exchange is not user-generated open forum. It is a moderated forum written by mathematicians and computer scientists who have established themselves to have a good reputation in their field. It is subject to peer-review, and unsubstantiated material is removed from their repository.—MeixiangKazuki (talk) 06:10, 1 January 2018 (UTC)

You could try raising this at the Reliable Source Noticeboard, but I'm pretty sure you'd get the same answer as last time. 222.153.250.135 (talk) 07:46, 1 January 2018 (UTC)

See section above "Chess has two players, White and Black" for necessity of including status of both players in the game of chess when discussing perfect play.—LithiumFlash (talk) 15:35, 1 January 2018 (UTC)

LithiumFlash You still haven't addressed the point - how is stackexchange a reliable source? Galobtter (pingó mió) 15:43, 1 January 2018 (UTC)

I support and agree with what has been already said (see above). It is a forum moderated by experts, and receives significant review by peers in the field. Unreliable information is quicky removed from stated conclusions. On this topic, it supports what other sources already say, but is germane because it often presents the most-up-to-date knowledge in regards to the topic at hand. Also, in this Wiki article, it is not being used as the sole citation - it supplements the others.—LithiumFlash (talk) 15:51, 1 January 2018 (UTC)

Regardless of the fact that the accepted answer essentially says "I've never heard of it" therefore it must not exist, and one of the comments quotes wikipedia, how does that stackexchange even support the statement? Galobtter (pingó mió) 15:59, 1 January 2018 (UTC)

Which statement are you referring to?—LithiumFlash (talk) 16:04, 1 January 2018 (UTC)

The statement that chess theorians have long debated etc which you added black wins to. Galobtter (pingó mió) 16:23, 1 January 2018 (UTC)

To be clear, I am not supporting text that says "Black wins in chess". I am supporting text that says it has not been answered which side wins in a game of perfect play. The article Solving chess says it pretty well in its opening paragraph:

"Solving chess means finding an optimal strategy for playing chess, i.e. one by which one of the players (White or Black) can always force a victory, or both can force a draw...No complete solution...is known, nor is it expected that chess will be solved in the near future."

These opening statments are not regarded as dubious, and have not been the subject of debate at Solving chess.

I actually don't think it's important which set of references we included here. Stack exchange is fine but can be excluded if some have an issue with it. But any statements that say only White can win in a perfect game should not be left standing in this article because that has not been proven anywhere.—LithiumFlash (talk) 16:41, 1 January 2018 (UTC)

The article on solving chess is not a reliable source. It actually cites stackexchange (though later). Also, there's a difference between black possibly winning and it being a long-standing debate, which is what your sentence says. Find some source. Then add. Galobtter (pingó mió) 16:44, 1 January 2018 (UTC)

Claude Shannon in his paper makes it crystal clear that he considers a possible win by either side, then follows by concluding that we don't know the answer. His paper is from 1950, and to my knowledge nothing has changed the conclusion. If his paper considers that either side can possibly win, then so should this article.—LithiumFlash (talk) 16:58, 1 January 2018 (UTC)

There's a difference between that and saying that they have long-debated on that. Galobtter (pingó mió) 17:01, 1 January 2018 (UTC)

I am not arguing if anything is long-debated or not. I am expressing that both White and Black have been considered, and are still considered, to possibly be the winner of a perfect game. The article (Solving chess) says it this way:

"Solving chess means finding an optimal strategy for playing chess, i.e. one by which one of the players (White or Black) can always force a victory, or both can force a draw...No complete solution...is known, nor is it expected that chess will be solved in the near future."

It has been variously included in this article such as (which I propose):

Chess players and theoreticians have long debated whether a perfectly played game is a win for White, draws, or wins for Black.

My proposed revisions do not make a change to the word "debate". That stands as is. Let me know if you have another preferred way to say this. The key is to not mislead readers into believing that only White can win.—LithiumFlash (talk) 17:22, 1 January 2018 (UTC)