draw

How do you draw on chess.com?!  My opponent repeated a position 50 times, but there was no draw button and typing "draw" or "/draw" on the console did nothing.

Did they really not implement draws here?  If so, this isn't real bughouse.

They did not implement draws in bughouse.  I believe this was intentional.  Playing OTB, we never allowed draws.

That's ridiculous.  Positions can be repeated over and over and over again for 2 minutes while you wait for pieces.  It completely changes the game.

Yes, but if the position on the other board is changing, is it really repetition if only one board is repeating?

Not rigged.

Rigged is when one person is favored over another. Here, you can either take advantage of it or Lowe to it.

The problem with bughouse is that there is no universally-agreed rule set.  Some allow draws; others don't.  Some allow promoted pawns to keep their promotions after being captured (I've played this OTB), others don't.

From the wikipedia page:

"The match can be drawn by agreement or when two players run out of time or are checkmated simultaneously. Depending on (local) rules threefold repetition applies, in which case the reserve of pieces is not taken into account."

There is a footnote which says "For instance, the threefold repetition applies on FICS but not on Internet Chess Club."

Of course, simultaneous checkmates or losses on time are rarely an issue in online play, because computers can identify who was first even if the difference is a fraction of a second.  So the only remaining possible draws would generally be by agreement or three-fold repetition.  I can't really think of a situation in which an agreed draw makes sense in bughouse (it's not like you can build impenetrable fortresses when pieces can drop anywhere and both sides are getting reinforcements all the time), and three-fold repetition is not universally recognised.

You realize a draw by agreement isn't even possible here either?  It's not just draw by repetition that isn't implemented.

Draw by agreement makes sense when there is a simultaneous mate on both boards and rather than rolling the dice on who is quicker the players decide to do the honorable thing and recognize the draw.  For example: taking a queen on board 1 gets you mated, but lets your partner mate on board 2.

Here's a rule-set that doesn't include 3-fold draws: http://www.nwchess.com/articles/rules/bughouse_rules.htm

Here's one that does: http://bughousechess.wz.cz/CompleteBughouseChessRules.htm

There is a universally-agreed set of rules for standard chess.  Everyone knows how bishops move.  Bughouse doesn't have the same level of standardisation right now.  Chess.com had to pick one set of rules, because it's not really practical to have multiple sets of rules for a single variant.  If you want to campaign for a switch to a set of rules that includes 3-fold, by all means go ahead, because that's just as legitimate a set of rules as one that doesn't.  But don't expect universal agreement; there are people who believe there should be no draws in bughouse under any circumstances.

In your first position, I would assume that your opponent had pre-moved the mate and that therefore your partner would have no time to play out the mate on his board, and therefore resign (or just play it out on the off chance that my opponent was slow).

In the second position, unless they're so close on time we can't see who's ahead, the player with less time should resign.  A draw would only make sense here if we can't see who's ahead and there's more than 30 seconds or so to wait to see which clock flags first.

There are also some people who play that you can't drop pieces for mate or can't drop pieces for check.  Just because some people play crazy rules doesn't excuse ignoring the rules used by 99.9% of bughouse players and all top players.  You can check the top bughouse tournaments around the world, other chess servers that have bughouse, all bughouse books that have been written, all top players, etc.  The notion that draws are a controversial subject is absurd.

The thing about bughouse is that there is no universally agreed rule set.  You may have always played with a 3-fold rule, but others are used to no draws (except perhaps by agreement or simultaneous checkmate/timeout).

Since there is no universal rule set, chess.com can implement whatever they want, so long as it's reasonably consistent with what many people are likely to be familiar with.  You're upset that they don't have draws.  Others feel strongly that bughouse should never have draws.  What's chess.com supposed to do?  People play it both ways, and they can't really implement both.

Until we get FIDE (or an alternative Bughouse organization) to release a rule book for bughouse that everybody follows, there are always going to be variations in the rules.

FICS allows 3-fold draws, ICC does not.  Both allow draws by agreement.

I don't see any reason not to allow draws by agreement, although all four players should need to agree.  It's the 3-fold rule that is not universal.

And, I'm not strongly opposed to a 3-fold rule (although I personally would vote against it).  But it should be decided recognising that there are two legitimate variations, rather than assuming the way FICS does it is the only way most people have played it.  A poll on the subject might be interesting.

You keep talking about FICS as if that was all I pointed to.  In fact all of the top bughouse players, all bughouse books, and all of the top tournaments use 3-fold draw.  I've played over 50,000 bughouse games, I've been to the world championship in Berlin many times, I've played more simul bughouse games than anyone else in the world, and I've probably met more bughouse players in person than anyone else in the world.  The notion that the 3-fold draw rule is contentious is utterly ridiculous.  The only outliers are people who have never played competitive bughouse.

ICC doesn't have a 3-fold rule - is that a stupid implementation on their part too?

(It's possible, I'm not an expert here, but everything I find online seems to indicate that there's more than one legitimate rule set)

Regarding ICC & FICS: Despite ICC having more players & chess games played than FICS, it has (and I may be being generous here) about 1% of the number of bughouse games played. Until recently, thanks to chess.com, FICS has been the lone home of internet bughouse, despite it being technically possible to play elsewhere.

I do think there is a flaw in the FICS repetition rule. It counts piece cycling as repetition, e.g. with a black pawn on e6 1. P@f5 P@f7 2. fxe6 fxe6 3. P@f5 P@f7 etc.  when probably the player with fewer pawns to drop should lose this battle of attrition rather than being able to claim a draw.

I don't mean to rehash the whole discussion from the "notes" earlier, so I'll just stick to a few main points, and maybe more people will see them here.

1) Bughouse is supposed to be fun and moving back and forth 50 times while the other board doesn't move at all isn't much fun.

2) I regularly play with a group of fairly strong players, and we initially didn't play with repetition. As we got better and the annoying situations became more common, we felt compelled to add the rule. What chuckmoulton says is no coincidence. Bughouse is new here at chess.com, so there probably aren't enough players strong enough to cause these situations to arise often enough to be annoying... yet. But it will happen eventually. Be warned

3) I support a rule that considers a draw claim valid only if A) The repetition doesn't involve drops and B) The other board is either not moving at all or is also repeating

Also, over the board, when both teams will mate on the same move, we call the game a draw.

As MGleason mentioned, online, the server can determine who was a fraction of a second faster. This is an acceptable solution. However, we should be aware that this may lead to the winner being determined by who has the faster internet connection or even quirks in server programming. On FICS, for instance, a premoved drop mate will tend to be ordered over a premoved chess move mate if both moves are received at the same time (due to being premoved). This was not an intentional decision by FICS - just something that was uncovered after premove was invented in the late 90s.

An alternative solution that might be worth experimenting with online would be that when one side mates, the result of the game is not declared until a very short amount of time later (let's say maybe 1/5th of second). If the other team also mates on the other board during that time, the game would be drawn.

I am not sure if this would be an improvement over current rules or not, but I'd be curious to see it tried somewhere at least.

What do you all think about that one?

I'd be content with near-simultaneous checkmates/timeouts being draws, with a tolerance of 200 milliseconds.  The frequency of such a draw would be very low, but not impossible.

I also wouldn't mind draws by agreement, so long as it requires all four players (i.e. two players on one board can't inflict a draw on the other board).

For 3-fold repetition, I think if you count pieces in hand as part of the position (which they technically are), it addresses the issue of counting drops as repetitions even though one side is running out of pawns.

If the 3-fold repetition must be on both boards (because the other board either is not moving or is also repeating), I think that's reasonable.  If you don't want a draw, wait to repeat until your partner or their opponent has moved.

"An alternative solution that might be worth experimenting with online would be that when one side mates, the result of the game is not declared until a very short amount of time later (let's say maybe 1/5th of second). If the other team also mates on the other board during that time, the game would be drawn."

Nice suggestion. This goes a long way in replicating OTB play. I like it!

There is a lot here, and, while I am interested, I see a couple of absolute conditions:

1.) A game being played (actively) on a team member's (and your team member's opponent) board has progress. (It is not a draw, no matter how many moves you repeat on your local board; especially if like Crazyhouse, you can repeat take/drop on the same board, but across board conditions.)

1.b.) Detection of repeated take/drop (over two boards) across time-control constraints over (potentially) latent network connections has an implicit timeout requirement irrespective; in which case the latency in first detecting, and then enabling an automatic "draw" button or condition becomes somewhat prohibitive in nature. (Some of this deals in software development and networking aspects of playing online.)

2.) Rule #1 especially comes into play when pawns are being pushed (i.e. standard, traditional chess rules when it comes to 50-move rules of play in semi-stalled games).

3.) Rule #2 means that timed games (especially close time constraints) will eventually time-out, or, become repetitious (draw); this considering observation [1.b], network latency (even if you were to track absolute client receipt times; which would increase latent processing and dissemination to relative clients such that clocks would lag > 1sec across even the most distant network connection).

Finally:

Drawing in games with less than a minute remaining (irrespective of original time control), and with an increment of 1 or 2 or 3 seconds is prohibitive.

Summary:

If everyone played games > 5 minutes (average per-board time), and with <= 3 second time increments, I would consider, as a programmer, implementing a "draw" (automatic or suggested and agreed upon) feature. But, unless or until these specific and explicit conditions come to pass, or, upon creating games <= 5 minutes with <= 3 second increment, I would not even begin to consider a "draw" type of feature.

OTB:

I've NEVER seen an OTB Bughouse draw. Never. It's always somehow conceded or called. (This said, being unfamiliar with competition-level Bughouse matches or tournaments.) And I am only familiar with Crazyhouse draws from playing online with repeated take/drop scenarios, which would then need to span across two boards detecting this across a total of 4 players, 2 boards, digitally, with latency, which would eventually cause a time-out.

Just my $0.02. (Paid in full.)

"Draw" in (typical) online Bughouse matches, consisting of short time controls and increments, seems like a prohibitive factor in terms of the time and cost involved to first implement, test, and then administer. (Though, I could be wrong, and, if the Chess.com dev's find a way to implement this, I'll never ever complain about them again.)

If you somehow question my experience or observations, then do yourself a favor and design your own live, real-time Foreign Currency Exchange application; this according to your local country and/or jurisdictional laws; and this given availability and accessibility of access to exchanges and laws in respective countries where you host servers, data exchanges, and updating of information across more constrained information pipelines beyond basic inter-connected latency from core data centers, etc. (If you question my experience, go gain some of your own as it pertains to live real-time data transaction processing.)