Comments/Ratings for a Single Item

I agree. Having no draws means the game is decided from the beginning, and this doesn't seem a good feature. A work-around would be to play it as a shuffle game, where equally many possible initial shuffles are winning as losing. But with perfect play the game then becomes a lottery., and you might as well flip a coin in cases that would otherwise be a draw, to decide who has won.

A popular game without draws is mini-Shogi. In case of repetition (which is not a perpetual check) gote wins. Because captured pieces are dropped back in play, a game doesn't naturally peter out to a dead draw like in Chess, so rules that in an unnatural way decide an equal position in one way or the other are not needed; positions where both players have insufficient material do not exist. (Even with a bare King you can in theory still win.) It seems that the initial position is a win for gote, though, and this is a bit unsatisfactory.

Should chess variants that must always end in a decisive result (for a given game played of them) be regarded with at the least some dislike by purists?

This would depend on why this happens. If it is because the game is decided for one player from very early on, then maybe. But if it is because the game's rules steer it away from, or declare a winner for, drawish situations, then maybe not. Thanks to captured pieces reentering the game, Shogi is less drawish than games with regular piece attrition like Chess, but it can still become drawish if each player gets his King to the other side of the board. The official rules deal with this by declaring a winner based on points. Shogi with Impassable Kings simply prevents this from happening. If stalemating your opponent is also considered a win, then the game is very decisive, and I think in a good way.

Indeed, stalemate = win is a natural rule. In games with piece drops stalemates only occur when one player is almost out of pieces, where it would also be very easy to checkmate.

Trying to force a decision in a King-vs-King end-game seems very artificial, though.

Trying to force a decision in a King-vs-King end-game seems very artificial, though.

I can't think of a game where that happens. My example was in Shogi, which would never have a King-vs-King endgame.

one side or another must inevitably win in a [even nearly{?}-]perfectly played game, even in any otherwise error free game played by each of the opponents involved

That's true for every deterministic game. If draws are possible, the perfect result might be a draw. How is that any better though? Nobody's going to play such a complex game perfectly. What matters are the results of actual imperfectly played games.

I can't think of a game where that happens.

It would be a consequence of forbidding repetition (or declaring it a loss, which is equivalent), as a means to eliminate draws, in combination with stalemating being a win. And I just made a Diagram for a variant that presents decisive K-vs-K end-games as a feature, and made it quicker by adopting a King-facing ban (Nacht Schach).

One other case I can think of that H.G. didn't mention (in his own reply to this post of yours) might happen is if the board is not rectangular/square, such as being made possible due to being played on an irregular shaped board (e.g. this could even happen as a possible result of a CV being played where a number of board cell[s] are laid down in alternative turns by each side, before each side's pieces are then alternatively put down, then-completing the setup phase for such a CV). Then, in a bare K vs. bare K situation that may happen much later, one side's K might be able to move to a cell, as if out on a limb, and then a move in reply, by the other side's K, stalemates that K, immediately.

Here's an example of a CV I played a few times decades ago, where irregular board shapes can routinely happen during a setup phase, as I alluded to above:

https://www.chessvariants.com/rules/choiss

To try to put things into a clearer perspective on that score, if I can, I'd note that are physical team sports leagues, where decisive results must happen in a given game (except for exhibition games, for example). For those, there are unavoidably significant (and undeniably) 'chance-based' events that occur when mere humans participate. Such as in baseball games played in North America's Major League Baseball, where one team must win - there are no draw results allowed, in the regular or playoff season, there, but at least the losers (and their fans) can at times with more truth (than in a board game of more arguably 'pure skill', like for many CVs) console themselves with the thought that they more or less lost because of an unlucky occurance, especially if the teams in question were closely matched by ranking/odds-makers.

In (FIDE) chess itself, games resulting in draws are rather commonplace, as is very well known. Many [potential] fans of chess (maybe more so at lower rating levels, if only due to mistakes being more frequent, increasing the chance of a decisive result for a given game between such players?!) lament, sometimes, that perhaps too many (percentage-wise, in a given database, for example) draws happen as a final result of a given game, at elite level, these days.

However, I recall that I've read somewhere in a chess book long ago, that was written by a Danish grandmaster (I think his name was Lars Bo Hansen) words to the effect that he was not very concerned by draws being possible in chess. He argued they are allowed by the scoring system of chess, and that they seem a perfectly natural and logical result for games that are well played by both sides. Plus, why should he need to risk a loss (if forced to, too routinely) by taking big chances?

Note that, even if one regularly wanted to take big chances to win when starting from a given position in a given game of chess, that may sometimes be none too easy to arrange (i.e. to attempt to sufficiently sharpen the position on the chess board), so as for a chess player to have a realistic chance of successfully avoiding a drawn result in the end (say against a reasonably skilled opponent). Especially so if the given chess position that has already been reached is already quite drawish (though an overly ambitious opponent might be hoped for, even in such a case, sometimes). That's especially noteworthy when a draw would be otherwise fully satisfactory to a player like Lars Bo (perhaps a fairly common situation during a given elite level player's chess event, if he were playing a given game with the Black army, when it's not a must-win situation, for such a player).

In my variant Spiral Chess one bare King can even force checkmate on another.

I'm glad I'm not so good at chess that I think trying to win is risky. If that's how high level chess players think, then chess with the current rules is a poor vehicle for high level competition.

Imagine you were designing chess for the first time. Would you really think it a good idea to design the rules such that it would make sense for the second player to go into a game with the goal of trying to draw? I don't know the statistics on this, but it sounds like size of the first move advantage also contributes to that problem.

Back to your original question; when I submitted Expanded Chess here at first, I included drawless end game rules, but changed that part because it was suggested to be too different. I don't really disagree with that perspective myself. Draws have historically been part of chess. Perhaps what makes drawless rules feel off is that it's hard (if not impossible) to remove draws without introducing a new victory condition or changing the pieces. And those are what really make chess feel like chess to normal players who rarely ever see a draw.

Last I saw in a chess book, around 55% of the time results favour White (i.e. White wins, with the [half-]value of White draws factored in, too), for any kind of a monster size database (though ratings could be low, besides high ratings, for all of the players involved).

Whether a (FIDE) chess player of any strength that is participating in a chess event (of any kind) decides to try to play for a win or a draw may depend on one or more factors, such as their desire to gain/risk rating points of whatever kind (or if possible cash prize $ are taken into account, especially if a game is being played towards the end of such an event).

Such factors are regularly taken into account even when contemplating what to possibly play (if allowed, perhaps anticipating a position that might arise well into the move/turn count) right from the setup position in a given chess game. If a player knows well enough in advance that they are to be paired against someone, they might look over any available previous game(s) that that future opponent has played vs. anyone (such as in a large database, similar as for anyone who plays on Game Courier on this CVP website might choose to research other players' previously played games in the GC Database - more worth the effort, perhaps, if their and their opponent's GC ratings are fairly close in value).

A situation that is more of real concern at elite players levels (the worst cases regarding extreme difficulty, in trying to win in a given game, even with either colored chess army, occur in high level correspondence chess play, or in high level computer engine matches, even, for a number of years now - the percentage of draws in such cases are downright depressing, to me personally - fortunately these two cases are of little interest to chess promoters, or to chess fans in general) is that winning in a game of chess when starting out with the Black army, even if Black may be able to thoroughly prepare for a given opponent in advance, is if Black is none too sure that White won't try to steer for a drawish/dull/solid position possibly right from the setup (or, to some extent, even if Black knows that that sort of approach by the player to play White will be his intended strategy).

Then, it may be that if Black needs to win at all costs, he may well be advised to hope to get to play a (at least) somewhat risky chess opening (even if it's a well-tested defense, say against 1.e4 with Black), or try to, e.g. grind down (i.e. play on against, for many, many moves as long as needed/[reasonably justifiable] vs.) a player of the White pieces if he plays quite unambitiously, but still sufficiently solidly (for example), i.e. by adequately avoiding a sharp tactical opening that Black may have hoped for. In such a case, e.g. in a high level chess chess match, it can be like in a boxing match where wearing down an opponent (i.e., as if to try to eventually induce him to make mistake[s] that prove to be too costly to him, before a position happens to be reached in the game where a drawn result more or less cannot be avoided), say by having greater stamina, may prove successful at least some of the time.

If I recall right, even the late Bobby Fischer may have had at least sometimes tried that approach in a given FIDE chess game of his, at least very long ago, in his younger days (these days, elite players may have a bit too much respect for each other's skills, as I somewhere read words to that effect, some years back - I also recall at the least one world chess championship, not too many years ago, had all 8 games at slow time controls drawn, when by the match rules, the players then resorted to playing games in a tiebreaker phase, with faster time controls - Magnus Carlsen won that particular match [among others, a number of years back], though he was significantly younger then, when as his match strategy he concentrated on trying to play cautiously, just so as to get past the first 8 games, to try to win with his quick thinking/reflexes in the tiebreak phase - there are too few slow time control games with that sort of format for a world chess championship match, for my liking, and I'm unsure if Magnus' opponent tried to take much risk in playing in the opening stages of those 8 games, even when playing in his 4 games with the White army).

Anyway, whatever strata is chosen to define what is an elite (or ultra-elite!?) level (such as for chess itself), as you have alluded to such a level is something that most of us can only dream of reaching, but at least before reaching such a level, we have less problems in winning our games, percentage-wise, rather than our having what might be perceived as too many draws (most especially with Black). So, possibly having great difficulty, on a somewhat regular basis, in trying to win (with either army) is not such an issue for us, unless, e.g., we worry a bit too much about whether our chess opening repertoires (as if they were never to change very much, in our own 'chess career') could in theory help us to win as needed if we ever were to reach such an elite level.

I wonder how decisive chess would be if repetition were forbidden.

A statistical study of a large chess games' database might shed some light on what a typical percentage of chess games are drawn due to the longstanding 3-fold repetition of position rule (resulting in a drawn result). The trouble is, quite a few of those chess games might have later ended in a draw, anyway, such as by the 50-move draw rule of chess, or simply by the players involved agreeing to a draw, or there being insufficient material on the board for either side to conceivably deliver a checkmate, perhaps far into the endgame stage in some cases.

However, something to consider is that draws by repetition being forbidden (normally/altogether? - not quite sure which, without my checking) being a part of the basic rules of Chinese Chess may hint that that sort of evidence-based 'study' just might have been carried out long ago, at least to some extent, in the case of that particular CV.

In the case of (now longtime standard 9x9) shogi, initiating that sort of 'study' may have occurred to that CV's own theoreticians, also, but, on the other hand, it may have been deemed unnecessary to make such (a study and/or rule) long ago, due to a high frequency of decisive results, rather than of draws, possibly being readily observed in a relatively modest sample size of finished games that were played between even highly skilled players, back then.

It's funny, though, that I had one shogi game offline, that I played against the friend who had introduced me to that CV, end in a draw by repetition result relatively early in the move count (even) from that CV's setup position - it's all the more funny because (I'm guessing) we had played far less than 150 games of that CV, in total, against each other.

Naturally, various attempts have been proposed to try to decrease the number of drawn results in chess games by (only slightly?!) modifying the basic rules of chess, such as to award 3/4 of a point for stalemating an opponent, or perhaps even for baring an opponent's K - any such attempts have yet to gain much traction, as far as I know (unless in the case of such things that are possibly viewed as a bit more 'radical', like 'chess without castling being allowed' [at the moment I forget the exact name of that proposal, which has had some testing, in high level events, at the least, in recent year{s}] are included, perhaps).

In my own case, as an aspiring CV inventor, I proposed, on a Canadian chess message board, long ago (then later had published here, on CVP site) what I thought was a quite/(perhaps too?) 'modest' departure from the rules of chess, to be used in a CV of mine I called 'Throne Chess'. That's where, if one player gets to legally move their own K to their opponent's K's starting square in the identical setup as for (FIDE) chess, they then win the game immediately, as an extra winning condition added to the basic rules of chess.

After I had first proposed 'Throne Chess', long ago, I noticed elsewhere, online, a couple of hundred of games (of that CV idea of mine) were play-tested, quite soon after I had first proposed that CV idea, and there was a reported increase of decisive results of 2%, over what would normally be expected for such a database sample size of finished (FIDE) chess games. Such a reported increase of only 2% more decisive games than what would be expected for a similar size database of (FIDE) chess games played is somewhat disappointing to me, and, in any event, this CV of mine, if played and studied more extensively, would I assume alter the expected result of any number of (perhaps) 'basic' chess endgames, if there is sufficiently adequate play, in terms of skill-level, by each side - something to consider for who knows how many 'modest' ways that may be proposed to tinker with the basic rules of (FIDE) chess.

A similar sort of (i.e. possibly unwelcome/unexpected) 'blowback' for aspiring CV inventors wishing to try to only 'modestly' tinker with the basic (FIDE) chess rules can happen even right in some otherwise quite standard chess opening variations, that have long been deemed perfectly acceptable to practical/professional chess players, perhaps most commonly if when preparing their opening repertoire with the Black army side, at home (such as, depending on if considering the rather commonplace case of if a decisive result is not strictly needed, in a given chess game that they might ever play, later on).

That, at least sometimes, might well include chess opening variations that have well-known draws by repetition that may even be otherwise unavoidable (e.g. if White wishes to indicate his possible willingness to go towards taking such a drawn result, maybe just for the sake of sometimes even letting Black be concerned about whether to avoid such a possible result), perhaps even lest an otherwise quite playable chess opening variation is cruelly pruned, e.g. due to having such a variation otherwise result in an outright loss for Black, due to there being draws by repetition ruled out if such a change to the basic rules of chess were to be made to that effect.

I was thinking if any move that would repeat a previous position were not allowed, the only game end conditions needed would be stalemate and checkmate. Presumably, players could recognize a long stalemate endgame and agree to a draw without having to play it all out.

Another less intrusive idea: give black a higher score for a draw than white, with the exact numbers adjusted to balance out white's expected score advantage.

Note that Chinese Chess doesn't forbid repetitions per se; it only forbids forcing of repetitions (e.g. by perpetually checking). In the latter case the player that does the forcing moves (checks or attacks on more valuable or unprotected pieces) loses, irrespective of who repeated first. Voluntary repetitions are still scored as a draw.

It is generally perceived as very unsatisfactory if a player can force a win by forcing a repetition.

Hi Daniel

Regarding your first sentence, I suspect that, regrettably, for the time being it just might take me a bit too much thought/effort to enter a fully satisfactory reply (though anyone else is welcome to attempt to do that). Meanwhile, whenever you have some free time, you might look up the particular rules of chess that list all of the various possible situations that are described with words and/or diagrams, there, where a drawn result for a game of chess might conceivably happen.

To try to put my dilemma, as I perceive it (whether it is real or imagined [I apologize if it's the latter]), into perspective, I'll try to give you a (hopefully valid) analogy, namely: for those who are first learning chess, it's perhaps a bit like learning to drive. When one has become a lot more experienced at driving a car, many years later, obeying the precisely stated rules of the road becomes a bit too instinctive to recall with precision, as if to think over any single rule of the road in a sentence of English. It might be all the more so if such a person were to be asked to give a somewhat lengthy talk, about any number of the rules of the road, to students at a driving school, without some preparation (and maybe even significantly more preparation to give even some sort of an educated guess about the feasibility and/or consequences of changing perhaps just one of those rules).

Your second sentence, however, I may be able to answer a bit more easily, though I don't know if it'll be an answer that seems fully satisfactory, for those who want more decisive results in chess, most especially in elite level play. It is that there has been, at least starting some years ago (and maybe it's still in use, perhaps) a special match tiebreak condition set (though for world chess championship cycle events, only, as far as I know) where (say, due to limited time available to the playing facilities available that host such a match) a match must end in the space of just one (final) game, e.g. if a previous fast time control phase used to break a (still) tied match still did not achieve that result.

Then, the special (i.e. single game only) match tiebreak condition must be used. Somehow (I don't know if there is one particular rule always used for this) it is decided which player takes which army (i.e. White or Black) in that single game, which is called the 'Armageddon Chess' tiebreak phase (due to it being 'the final battle' - I hope Fergus, as being CVP webmaster/(editor-in-chief), will not edit out [at least part of] that standard phrase that is used as the name for it [as is still the case in chess playing circles], due to a certain word that's part of that phrase).

Anyway, a chess clock is used for the play of the game in question (notably with no increments ever added to either player's available time during play), and, as I recall, it is with the player of the White army getting 5 minutes to play the entire game, and Black getting just 4 minutes (except, if the result is a draw, Black in effect wins the game, and thus the match - otherwise, White wins the match [I don't know what happens, rules-wise, for any possible unusual cases that might ever happen]).

Personally, I think this particular special match tiebreaking method (and the name of it) is pretty crappy (e.g. I wonder if the player of the White army is to be generally, in theory, favored [over an opponent of equal chess rating playing the Black army] to become the winner of such a single game [perhaps even significantly] more than usually would be the case [that is, by the normal rules of a {5 minutes per side} speed chess game still being applied, though where a no-decision {i.e. normal drawn game stands as just that} kind of result would {naturally} still be possible], by this special format, due to White's having an extra minute on his own clock). Note, though, that I don't know if there has ever been such a high level match where this special tiebreaking method has been employed, due to its being necessary.

In any event, a fellow chess-playing master friend once gave me his own appraisal of such an 'Armageddon Chess' format by saying, 'at least it's chess' (e.g., for the sake of comparison, many years back, the winner of an important match in such a world chess championship cycle was decided completely at random, due to the match still being tied past its previously planned tiebreak stage [by the match conditions that were already set in advance, I think], and because the match was played in Monte Carlo, which [I presume {also}] has within its city limits a world famous casino, a roulette wheel was used to in effect act as the match's tiebreaking 'method', to decide which of the two players would then advance without any further delay to the next level of that year's world chess championship cycle).