Hammering out a Fifty Turn Rule proposal

Hello all,

This thread is an offshoot of the Endless Battle Issues thread, with the purpose of more thoroughly fleshing out and discussing a fifty turn rule. The rule was first proposed by Eien here. There appeared to be some significant support for such a proposal judging by the likes on Ciele's later post here. The original proposal can be seen below:
If we do not implement a hard turn limit, is there any way to possibly implement a system similar to the fifty-move rule from chess? Fifty-move rule states that if no piece has been captured or no pawn has been moved in 50 turns, then a tie is forced. From the wikipedia page on this rule:

"The purpose of this rule is to prevent a player with no chance of winning from obstinately continuing to play indefinitely or seeking to win by tiring the opponent."​

Here's an example of a disgustingly complicated system for Pokemon (I guess all endless battle systems are complicated, though?):

If:
1. No Pokemon has received direct damage (where direct damage means damage taken from the opposing Pokemon by any move)
2. No Pokemon has received net passive damage barring Pokemon with Regenerator (where net passive damage means final HP - initial HP < 0 in any one turn. self-induced damage, entry hazard damage, and ability- or item-caused damage would count)
3. No Pokemon has received net healing outside of Regenerator (where net healing means final HP - initial HP > 0 in any one turn)
4. No Pokemon with Regenerator has received greater than 30% net passive damage in any one turn
5. This is valid for X turns in a row (where X is the decided turns)

Then:
Tie is forced.

I haven't spent a ton of time on this, so if you find a loophole sorry, but assume loopholes are filled. This is also horribly complicated and I'm not advocating for implementation in this really ugly current form, but you get the general idea. The question is whether any variant of the fifty-move rule is feasibly possible in Pokemon.
I believe that eien's post serves as a decent starting point for the framework on a proposed fifty turn rule.

Zarel, and I would guess some others, would like to see a fifty turn rule proposal hammered out more thoroughly before considering options to resolve the Endless Battle Issues thread. Note that this thread's existence does not necessarily mean we are going to move in the direction of a 50 Turn Rule as a solution to the Endless Battle Issues. Personally, I am neutral with regard to which solution we choose to resolve that thread. I would, however, like to see some sort of solution implemented before the start of OLT.
 
I'm not completely sure how this would work, but I'd like to request a system where in if both players mutually agree to continue (probably press a button) the game may go on for another 50 turns, and on and on. This would help close games go on if both players were enjoying them or at least thought they had the time. For this to work there'd probably need to be a dq timer for pressing the button as to not keep the game in limbo, and I think some tours could either ban continuing or make it required a certain amount of turns, etc. Just a little suggestion that could be implemented, wondering if it could work.
 
Why do people oppose a hard limit rule for forced ties? People merely clicking 1 move every 50 turns can still let them stall to a draw for hours upon hours. You may be preventing "endless games" but you aren't preventing 10+ hour games.

Why can't we implement a request draw button? As long as we don't let it happen immediately and super frequently then there seems to be no issue.

I spoke about these topics in the other thread:
Regardless of whether or not we create a forced draw based on "x amount of turns with no move clicked", there seem to be no drawbacks to a request draw button.

If it's possible on link or wifi battles or whatever to tie in some way then this shouldn't be an unappealing option for the sim. Even if there is no precedent, there is still the ability to outright forfeit so there shouldn't be any real harm in allowing a mere request to draw. If you fear abuse of a draw request, I recommend having like a 100 turn minimum and maybe every 100 thereafter. Those numbers don't have to be exact but you get what I'm saying. If I'm missing some downside of this feature then please let me know.

As far as forced draws go, this gets tricky. It's very distasteful to break any sort of game mechanics, so it really just becomes a question of whether or not we decide it's worth it to do so. As far as I know there's no link/wifi precedent for such a thing so we would have to create our own new thing. I think the most feasible is a turn cap that is exceedingly high (think 1000+), so we don't need to create some sort of super arbitrary mechanism such as "game draws when neither player clicks a move in 75 turns". A raw turn cap is favorable because it doesn't unnaturally force players to use PP "before the 75th turn". Rather, with a raw turn cap, players can still play at their own natural pacing, just within a very distant cap that doesn't affect shorter games.

tl;dr: no apparent downsides to an abuse-free request draw button; forced draw should be a raw turn cap (1000+)

If we go with some variation of the 50 turn rule, it should definitely be closer to 25 turns to prevent real life time stalling, but then we're messing with some potentially shorter games with a brief stalemate. And is there a reason why we can't simplify this rule to "no pp used by either player in 50 turns"? Going by passive damage and whatnot seems awfully complex and ugly as a rule to implement, while focusing solely on pp usage is simple and still seems to get the job done.
 

Martin

A monoid in the category of endofunctors
is a Smogon Discord Contributoris a Forum Moderator Alumnusis a Community Contributor Alumnusis a Contributor Alumnus
I disagree with something wishy-washy like what is proposed in the Drew's post. Either we have a hard turn rule (whether it be turns without damage, turns without PP consumption, turns with either of those on one side of the field, or just overall turn count) or no rule at all. Whether it is 50 turns or 100 turns or 500 turns or 1000 turns or whatever number we eventually decide on, I think that number should be an absolute end point which is consistent across all formats and settings without any silly options which undermine the purpose of implementing a limit such as this: especially when the ladder is used for tours like OLT, further blurring the line between tournament and ladder play.

Personally, I support a total turn limit at which point a stalemate is called (i.e. regardless of damage dealt at any point in the match) as it is the only method of preventing 10 hour games without implementing the one-hour limit present on cart. If we were to implement this, it would need to be a high number (e.g. somewhere in the region of 750-1000 turns) to minimise the possibility of stalemates being called where they aren't applicable. This isn't perfect, of course, but, like ABR said, the system can't prevent 10+ hour games if it's an activity checker because "1 move every 50 turns" isn't hard to maintain and, as such, fails to serve the intended purpose adequately. I oppose a request draw feature if only because it is much easier to abuse in tours like OLT at the cost of people lower on the ladder; I would be all for it if we didn't have to take OLT into account, but we do and, as such, it isn't a viable solution (not to mention that it can be abused outside of sitting on ranks by forcing a forfiet out of an opponent who is in a rush in order to net free ELO by simply rejecting all draw requests).
 
I disagree with something wishy-washy like what is proposed in the Drew's post. Either we have a hard turn rule (whether it be turns without damage, turns without PP consumption, turns with either of those on one side of the field, or just overall turn count) or no rule at all. Whether it is 50 turns or 100 turns or 500 turns or 1000 turns or whatever number we eventually decide on, I think that number should be an absolute end point which is consistent across all formats and settings without any silly options which undermine the purpose of implementing a limit such as this: especially when the ladder is used for tours like OLT, further blurring the line between tournament and ladder play.

Personally, I support a total turn limit at which point a stalemate is called (i.e. regardless of damage dealt at any point in the match) as it is the only method of preventing 10 hour games without implementing the one-hour limit present on cart. If we were to implement this, it would need to be a high number (e.g. somewhere in the region of 750-1000 turns) to minimise the possibility of stalemates being called where they aren't applicable. This isn't perfect, of course, but, like ABR said, the system can't prevent 10+ hour games if it's an activity checker because "1 move every 50 turns" isn't hard to maintain and, as such, fails to serve the intended purpose adequately. I oppose a request draw feature if only because it is much easier to abuse in tours like OLT at the cost of people lower on the ladder; I would be all for it if we didn't have to take OLT into account, but we do and, as such, it isn't a viable solution (not to mention that it can be abused outside of sitting on ranks by forcing a forfiet out of an opponent who is in a rush in order to net free ELO by simply rejecting all draw requests).
I'm still not seeing how a "request" can be abused in this form, though. If someone is asking for a draw when the game can be completed, then a smart opponent can simply reject the request, and only has to do so once every 100 turns or so. If someone accepts a draw when they can otherwise win, that's their fault. I think a request option is potentially very useful among players who can quickly identify when a game is going to eventually lead to a draw, so they don't have to wait 1000+ turns for an outcome they know is coming.
 

Bughouse

Like ships in the night, you're passing me by
is a Site Content Manageris a Forum Moderator Alumnusis a CAP Contributor Alumnusis a Tiering Contributor Alumnusis a Contributor Alumnus
In chess, the 50 move rule comes into play since the most recent capture of a piece (or movement of a pawn). Exceptions were found of forced win endgames that took longer than 50 moves and these were carved into the rule... then later carved back out and they said tough shit that you /could/ win if you had more time - the rule is 50 moves.

The more I think about it the less I think the event that should start a 50 move rule type thing is PP usage. Those games would last over a day still lol if someone wanted to. I think a total turn limit since the most recent Pokemon fainted is probably the right place to go. That is after all the most analogous thing to a capture in chess. It's also coincidentally probably way easier to keep track of in the coding.

I'm not sure what turn numbers should be selected. But we can even have different amounts of turns for different game states. So maybe a 6-6 game from turn 1 has to get to turn X before it's a draw, but once it's 6-5 it's only Y turns, and if it's 5-5 then it's Z turns etc. Could be based on individual game states (6-4 and 5-5, etc) or on total Pokemon available (12, 11, 10, etc).
 

Zarel

Not a Yuyuko fan
is a Site Content Manageris a Battle Simulator Administratoris a Programmeris a Pokemon Researcheris an Administrator
Creator of PS
The problem with "request draw" is that it doesn't really solve the main problem. As long as one of the players doesn't want to draw, the game can still go on indefinitely. The other player and spectators wouldn't have any choice in the matter. This is why Go and Chess have various threefold repetition rules even though you could also just agree to a draw in many formats.

There's also the other possible issue (brought up in the other thread) that draws can be coordinated in ways that subvert the ladder system.

I am not implementing "request draw" unless it's forced on me by tier leaders. I'd rather flesh out a 50-turn rule.

We can have both a 50-turn rule and a 1000-turn limit; I think that'd be best.
 

Zarel

Not a Yuyuko fan
is a Site Content Manageris a Battle Simulator Administratoris a Programmeris a Pokemon Researcheris an Administrator
Creator of PS
If the rule is on PP usage (or a proxy thereof, like damage), I think it should max out at 10 turns. That'd be roughly equivalent to the threefold repetition rule (since I'd guess three doubleswitches is a full repetition).
 

UltiMario

Out of Obscurity
is a Pokemon Researcher
I know it's just nitpicking, but far as implementation goes for allowing players to continue if both players want to: If this criteria is met for "endless battle clause" both players get a Force Draw button. They'd individually be able to press it to end the game if they want, but if neither player wants to they can continue. I think this is the most fair way to treat the issue, rather than just forcing the tie or creating dialogues that bug you every X turns.
 
The problem with "request draw" is that it doesn't really solve the main problem. As long as one of the players doesn't want to draw, the game can still go on indefinitely. The other player and spectators wouldn't have any choice in the matter. This is why Go and Chess have various threefold repetition rules even though you could also just agree to a draw in many formats.

There's also the other possible issue (brought up in the other thread) that draws can be coordinated in ways that subvert the ladder system.

I am not implementing "request draw" unless it's forced on me by tier leaders. I'd rather flesh out a 50-turn rule.

We can have both a 50-turn rule and a 1000-turn limit; I think that'd be best.
I didn't mean that request draw should replace forced draw in any way. I just want at least the option to request a draw, which can exist in addition to some forced draw option.

Requested draws don't subvert the ladder system in any way if 2 people consent to it (they have to - it's a request) and is only available after 100 turns + 15-30 minutes let's say. How does it hurt anyone if 2 people agree to draw, regardless of the reason? One player can't even bug the other until another 100 turns pass.

Now, in regards to a ~50 turn rule or large cap (1000), there are still many flaws to the style of the 50 turn rule. It's either a short number of turns (like 5) that alters the way normal, brief games are even played. If you do a higher number closer to 50, then it doesn't prevent games from going on for hours. If it's used in addition to a large cap, then it might as well not exist at all, because the large cap will always come into play first. The only feasible solution is a large cap and a large cap alone. And because it's so distant and endless games can be identified early on, we should also allow people to request draws.

e: an extending option sounds good so games like me vs tele can exist if the players want it to
 
Last edited:
TL; DR: pp counting to determine stalled games is not feasible with a simple heuristic. It would need to be more complicated.

---

Details: If pp usage forestalls the automated draw, any player may delay this for a pretty long time. In that, it isn't much different and still requires both players to agree for the game to end.

Let's imagine a situation where both players have the same two pokemon:
Toxapex: Haze - 48pp, Recover - 16pp, Scald - 24pp, t-Spikes - 32pp
Amoonguss: Giga Drain - 16pp, Sludge Bomb - 16pp, Spore - 24pp, HP Fire - 24pp

Sum: Toxapex - 120pp, Amoonguss: 80pp.

"Use one pp every 10 turns" isn't a strong enough strategical disadvantage that player 1 can punish player 2 for delaying like that and get a win instead of a draw. Like they would be in chess if their opponent threw away pieces to prevent the 50 turn rule trigger. Player 2 gets to delay the draw by 10 * 200 = 2000 turns.

I don't think a simple pp counting solution is feasible.
A human can count only the pp that matter and imagine how the game will go on to determine if it's drawn. But a computer can't do that unless you teach it how to play, or give it a complicated heuristic.
 

Zarel

Not a Yuyuko fan
is a Site Content Manageris a Battle Simulator Administratoris a Programmeris a Pokemon Researcheris an Administrator
Creator of PS
The other reason is because draws represent a failure of game design. You at Smogon are game designers - the minute you make a rule that changes the game from the official rules (and I'm not just talking about removing the built-in link battle time limit, I'm also talking about stuff like banning abilities and items), you are designing your own game based on Pokémon, and it's your job to design a good game.

If you want an "offer draw" button, it means you've designed a game so bad that neither player wants to play it. And my response is: Don't take the cop-out, go back to the drawing board and make the game actually good. At the very least, make a tie rule that doesn't require player intervention.
 
The other reason is because draws represent a failure of game design. You at Smogon are game designers - the minute you make a rule that changes the game from the official rules (and I'm not just talking about removing the built-in link battle time limit, I'm also talking about stuff like banning abilities and items), you are designing your own game based on Pokémon, and it's your job to design a good game.

If you want an "offer draw" button, it means you've designed a game so bad that neither player wants to play it. And my response is: Don't take the cop-out, go back to the drawing board and make the game actually good. At the very least, make a tie rule that doesn't require player intervention.
Pokemon singles as a whole is what failed in this regard, not smogon. It is because of options like Regenerator, Leftovers, and similar attack/recovery PP (instead of a lot more attack) that prevents games from ending. If games can theoretically be drawn in the long run anyway, why not allow players to agree to it after a substantial but not massive amount of time?
 
Requested draws don't subvert the ladder system in any way if 2 people consent to it (they have to - it's a request) and is only available after 100 turns + 15-30 minutes let's say. How does it hurt anyone if 2 people agree to draw, regardless of the reason? One player can't even bug the other until another 100 turns pass.
2 people can maliciously consent to draw a game at the expense of a third one, or to give unnecessary help in some tasks like suspect tests.

Example 1:
OLT or any other ladder tournament; Player Grass, Player Fire and Player Water are competing for bottom spots of top 8 qualification for playoffs; Player Grass is at the 7th spot with, say, 10 points more than Player Fire, which is at the 8th spot with, say, 30 points more of Player Water, which is at the 9th spot. 30 points can be obtained easily facing high GXE/ELO players (idk how this properly works so every number is assumption) in, say, 3 battles, or either facing medium-low GXE/ELO players in 10 battles, with any risk of the game (be it hax, matchup, disconnection, or anything else).
Player Grass is a close friend of Player Fire; both players are ensuring their playoffs spot so they do just have to keep their position. If Grass/Fire draw their games, maybe with the help of some other players in mid-high position that already gave up in taking the spot, Player Water will have way less chances to qualify, because all other players can simply choose to not fight against themselves. Players in mid-high position can start battling against Player Water in order to make him lose points, and if they instead face Grass/Fire, they draw. This example is very unlikely on a huge extent, but this sort of "circlejerk" could be very likely to some littler ones.

Example 2:
Player A and Player B are both doing a suspect test; they face each other and they are both close to desired COIL. They draw. While I can accept a situation like this, someone else can call this a "gaming of the system" since you cannot have the power to choose who are you facing on a ladder competition (and that can be somehow proven by the fact that, while tie is possible via link games, it is not possible on any Battle Spot ladder or any sort of Ranked battle on-cart).

I believe that ending those sort of almost-endless games must be definitely an option, but situations I've described are definitely something to consider when we're talking about draw-button abusing.

Making request draw button at a determined turn (300+?) can help in some way if it let the simulator assume you are going to request a draw for lengh-battle issues, or either to let "people-playing-mafia" waste time and/or having more chances to get caught, given the chance. It's just a more complicated situation than it seems, thus it would require a carefully thought solution, taking in consideration every single backdraw.


Smogtours does not run any official ladder tournament, neither suspect test, and it runs many few battles than the main simulator so any abuse can be caught with ease, therefore I think this feature can be put on it without any dramatic issue.
Since we are here talking about smogtours, can someone fix that pixel thing plsty
 
2 people can maliciously consent to draw a game at the expense of a third one, or to give unnecessary help in some tasks like suspect tests.

Example 1:
OLT or any other ladder tournament; Player Grass, Player Fire and Player Water are competing for bottom spots of top 8 qualification for playoffs; Player Grass is at the 7th spot with, say, 10 points more than Player Fire, which is at the 8th spot with, say, 30 points more of Player Water, which is at the 9th spot. 30 points can be obtained easily facing high GXE/ELO players (idk how this properly works so every number is assumption) in, say, 3 battles, or either facing medium-low GXE/ELO players in 10 battles, with any risk of the game (be it hax, matchup, disconnection, or anything else).
Player Grass is a close friend of Player Fire; both players are ensuring their playoffs spot so they do just have to keep their position. If Grass/Fire draw their games, maybe with the help of some other players in mid-high position that already gave up in taking the spot, Player Water will have way less chances to qualify, because all other players can simply choose to not fight against themselves. Players in mid-high position can start battling against Player Water in order to make him lose points, and if they instead face Grass/Fire, they draw. This example is very unlikely on a huge extent, but this sort of "circlejerk" could be very likely to some littler ones.

Example 2:
Player A and Player B are both doing a suspect test; they face each other and they are both close to desired COIL. They draw. While I can accept a situation like this, someone else can call this a "gaming of the system" since you cannot have the power to choose who are you facing on a ladder competition (and that can be somehow proven by the fact that, while tie is possible via link games, it is not possible on any Battle Spot ladder or any sort of Ranked battle on-cart).

I believe that ending those sort of almost-endless games must be definitely an option, but situations I've described are definitely something to consider when we're talking about draw-button abusing.

Making request draw button at a determined turn (300+?) can help in some way if it let the simulator assume you are going to request a draw for lengh-battle issues, or either to let "people-playing-mafia" waste time and/or having more chances to get caught, given the chance. It's just a more complicated situation than it seems, thus it would require a carefully thought solution, taking in consideration every single backdraw.


Smogtours does not run any official ladder tournament, neither suspect test, and it runs many few battles than the main simulator so any abuse can be caught with ease, therefore I think this feature can be put on it without any dramatic issue.
Since we are here talking about smogtours, can someone fix that pixel thing plsty
If people have to wait a couple hundred turns AND a fixed amount of time, people won't be agreeing to a draw every time they see their friends. If people are really going to mindlessly switch for a couple hunfred turns and then sit there for 30 minutes - 1h just to draw their friend, and that's assuming they're playing someone who wants to draw as much as them, then sure let them do that. But I know most people would rather just play the game in front of them than do this long dance so they can draw.

You also have to realize that 2 people can simply agree to mindlessly switch rapidly or use pp once in a while to artificially "force draw", so you might as well give them a button.

And if ladder abuse is the most feared component, which is silly anyway, then enable this feature for smogtours please.
 
Last edited:
Posting on behalf of UBERPYRO :

Why don’t we take a simpler approach, such as no faints after, say, 100 turns?

The suggestion in the OP would be very hard to keep track of by players. One player may be in control of a game, but play too safely and *bam* a tie is enforced. Even with a warning, it would be difficult for players to learn and remember. It’s complex. And lastly, while this can be enforced on the server, it would be impossible for players to keep track of it in-game, if that matters at all.

If we determine an endless battle by if an attacking move has been used, or if damage has been dealt, then this is easy to circumvent. Someone could use an attacking move every 49 turns, and stall for a very long amount of time. Making a more complicated system to prevent this issue will produce a clause difficult for players to follow and understand, similar to the first suggestion.

By basing it off Pokémon faints, the system is extremely simple. It’s relatively easy to keep track of, easy to understand, and it’s similar to the taking of a piece in chess. If it’s 100 turns per faint, it sets a hard limit of 600 turns to the Pokémon battle, ensuring ludicrous 1000+ turn games don’t occur. Additionally, it ensures progress is being made since 1 Pokémon must faint periodically. Battles are also likely to speed up turn-wise as Pokémon begin to faint (due to holes in coverage). While much progress can be made in a Pokémon battle without any Pokémon fainting, there’s no easy way to tell this with a simple clause. Progress from damage alone is not black and white, like fainting. Lastly, if one player is in control of the battle, they should definitely be able to get at least 1 KO on the opponent within a reasonable turn limit like 100.

I know this hurts stall vs stall matches which may be seen as great displays of skill between skilled players. It’s true that the system I’m suggesting may interfere with battles that aren’t truly endless, but I think it is reasonable to assume that pp stalls that go beyond stalling the recovery move of 1 Pokémon isn’t a game that most players want to play, and I definitely think that somesystem needs to be put in place to prevent that from happening.

Simply put, a turn limit based of KOs would be easy for players to follow and would guarantee that long battles don’t exist. I would make this turn limit 100, but I would also consider slightly lower numbers like 80.
 
Posting on behalf of UBERPYRO :

Why don’t we take a simpler approach, such as no faints after, say, 100 turns?

The suggestion in the OP would be very hard to keep track of by players. One player may be in control of a game, but play too safely and *bam* a tie is enforced. Even with a warning, it would be difficult for players to learn and remember. It’s complex. And lastly, while this can be enforced on the server, it would be impossible for players to keep track of it in-game, if that matters at all.

If we determine an endless battle by if an attacking move has been used, or if damage has been dealt, then this is easy to circumvent. Someone could use an attacking move every 49 turns, and stall for a very long amount of time. Making a more complicated system to prevent this issue will produce a clause difficult for players to follow and understand, similar to the first suggestion.

By basing it off Pokémon faints, the system is extremely simple. It’s relatively easy to keep track of, easy to understand, and it’s similar to the taking of a piece in chess. If it’s 100 turns per faint, it sets a hard limit of 600 turns to the Pokémon battle, ensuring ludicrous 1000+ turn games don’t occur. Additionally, it ensures progress is being made since 1 Pokémon must faint periodically. Battles are also likely to speed up turn-wise as Pokémon begin to faint (due to holes in coverage). While much progress can be made in a Pokémon battle without any Pokémon fainting, there’s no easy way to tell this with a simple clause. Progress from damage alone is not black and white, like fainting. Lastly, if one player is in control of the battle, they should definitely be able to get at least 1 KO on the opponent within a reasonable turn limit like 100.

I know this hurts stall vs stall matches which may be seen as great displays of skill between skilled players. It’s true that the system I’m suggesting may interfere with battles that aren’t truly endless, but I think it is reasonable to assume that pp stalls that go beyond stalling the recovery move of 1 Pokémon isn’t a game that most players want to play, and I definitely think that somesystem needs to be put in place to prevent that from happening.

Simply put, a turn limit based of KOs would be easy for players to follow and would guarantee that long battles don’t exist. I would make this turn limit 100, but I would also consider slightly lower numbers like 80.
This was a WCoP game this year: http://replay.pokemonshowdown.com/smogtours-gen7ou-296106

The first death was on turn 163, and 1 of the players wasn't even using stall. This proposal seems very dangerous to totally legitimate games played. I'm sure there are many more examples like this.

Additionally, there's nothing "hard to keep track of" in regards to a large cap, and we could always implement a counter or a battle warning saying "game draws in x turns if y doesnt occur" if something like the 50 turn rule exists.

UBERPYRO
 

Zarel

Not a Yuyuko fan
is a Site Content Manageris a Battle Simulator Administratoris a Programmeris a Pokemon Researcheris an Administrator
Creator of PS
Pokemon singles as a whole is what failed in this regard, not smogon. It is because of options like Regenerator, Leftovers, and similar attack/recovery PP (instead of a lot more attack) that prevents games from ending. If games can theoretically be drawn in the long run anyway, why not allow players to agree to it after a substantial but not massive amount of time?
Yes, but I think the correct solution here is to have an automatic draw condition, instead of letting players decide.

"Offer draw" is a symptom, and while it means you can't get games BOTH players don't want to play, you can still get games that ONE player doesn't want to play. Which can lead to various negative externalities: now, a player can explicitly refuse to draw if the other player needs to sleep or go to the bathroom, etc, etc.

It's much better to have a rule that is always and consistently enforced, so it's at least not the players deciding whether or not the game is fun enough that they want to continue playing.
 
Yes, but I think the correct solution here is to have an automatic draw condition, instead of letting players decide.

"Offer draw" is a symptom, and while it means you can't get games BOTH players don't want to play, you can still get games that ONE player doesn't want to play. Which can lead to various negative externalities: now, a player can explicitly refuse to draw if the other player needs to sleep or go to the bathroom, etc, etc.

It's much better to have a rule that is always and consistently enforced, so it's at least not the players deciding whether or not the game is fun enough that they want to continue playing.
Request draw, again, is not a replacement for force draw. It's an addition. Both can exist simultaneously.
I didn't mean that request draw should replace forced draw in any way. I just want at least the option to request a draw, which can exist in addition to some forced draw option.

Requested draws don't subvert the ladder system in any way if 2 people consent to it (they have to - it's a request) and is only available after 100 turns + 15-30 minutes let's say. How does it hurt anyone if 2 people agree to draw, regardless of the reason? One player can't even bug the other until another 100 turns pass.

Now, in regards to a ~50 turn rule or large cap (1000), there are still many flaws to the style of the 50 turn rule. It's either a short number of turns (like 5) that alters the way normal, brief games are even played. If you do a higher number closer to 50, then it doesn't prevent games from going on for hours. If it's used in addition to a large cap, then it might as well not exist at all, because the large cap will always come into play first. The only feasible solution is a large cap and a large cap alone. And because it's so distant and endless games can be identified early on, we should also allow people to request draws.

e: an extending option sounds good so games like me vs tele can exist if the players want it to
 

Zarel

Not a Yuyuko fan
is a Site Content Manageris a Battle Simulator Administratoris a Programmeris a Pokemon Researcheris an Administrator
Creator of PS
Request draw, again, is not a replacement for force draw. It's an addition. Both can exist simultaneously.
Either it's useless, in which case we shouldn't have it, or it's useful, in which case your force draw conditions are bad.

Design a game that does not require "Offer draw". You said yourself that your force draw conditions require probably drawing things out to 1000 turns. Make better conditions.
 

Zarel

Not a Yuyuko fan
is a Site Content Manageris a Battle Simulator Administratoris a Programmeris a Pokemon Researcheris an Administrator
Creator of PS
It'd actually be not hard to just extend Endless Battle Clause's "no damage" detection, and just make it ignore PP.

Game ties after 10 turns of PP usage, or 50 turns of no net damage.
 
Last edited:

p2

Banned deucer.
Why don't we trial a request draw button and if issues are still persisting, follow up with a forcedraw scenario?

It seems weird to immediately go for the most drastic route instead of what seems to be the most simple one, it's also specifically catering to stall vs stall which we're gonna see a lot of during OLT and many players are gonna really really appreciate the option to go for draws instead of dragging games out for hours just to not lose elo.

If you want an "offer draw" button, it means you've designed a game so bad that neither player wants to play it. And my response is: Don't take the cop-out, go back to the drawing board and make the game actually good. At the very least, make a tie rule that doesn't require player intervention.
Chess has the option to offer draws, and that's clearly a game that isn't "so bad that neither player wants to play it".

I also don't see request draw being an easily abused thing, I remember the idea of it being brought up that people could use it to dodge their friends and such, but this is only really applicable in olt or suspect ladders, the former which can get you infracted on here(and possibly a lot more), while the latter is actually punishing to both players due to the game limits enforced (not gaining elo means you're not increasing your gxe, and therefore hurting your overall coil gain almost every time). Either way, players can still be punished and banned from suspect voting if they're caught abusing the system, so I don't think avoiding friends is a good excuse to not implement request draw.
 

Zarel

Not a Yuyuko fan
is a Site Content Manageris a Battle Simulator Administratoris a Programmeris a Pokemon Researcheris an Administrator
Creator of PS
Chess has the option to offer draws, and that's clearly a game that isn't "so bad that neither player wants to play it".
Chess is fun without the option to offer draws. The threefold repetition and the 50-turn rule are, if anything, erring on the side of "making games short", and make some otherwise-winnable endgames unwinnable.

Smogon OU clearly isn't, judging by how you guys are begging me to add support for it.

Go back to the drawing board. Fix it properly.

It seems weird to immediately go for the most drastic route instead of what seems to be the most simple one, it's also specifically catering to stall vs stall which we're gonna see a lot of during OLT and many players are gonna really really appreciate the option to go for draws instead of dragging games out for hours just to not lose elo.
Having a game that people need to constantly stop and ask themselves "do I want to keep playing?" should be the "most drastic route".

Like I said:

> "Offer draw" is a symptom, and while it means you can't get games BOTH players don't want to play, you can still get games that ONE player doesn't want to play. Which can lead to various negative externalities: now, a player can explicitly refuse to draw if the other player needs to sleep or go to the bathroom, etc, etc.

Fix. Your. Game. "Offer draw" is a cop-out, which would be fine but it's a cop-out that doesn't even fix the problem. If it's "drastic" to expect you to actually fix the problem, something is seriously wrong here.
 

p2

Banned deucer.
Smogon OU is a good tier for the most part but lets recap the main reason we want request draws.

Stall vs Stall games are nearly the entire reason we want request ties introduced, there are some other rare cases like the ABR/tele game (this game was needlessly dragged out for so long anyway), but the real main thing demand for request draw is Stall vs Stall and given OLT is starting up in the next week or so where Stall is going to be extremely common. We want request draws to prevent having to play through Stall vs Stall matchups for extremely long times - this is seriously the only relevant matchup, other than GSC, where there is a problem and the answer to that shouldn't be determining a wincon through the other players patience and ability to play for hours.

As for "fix your game", I don't see how we possibly can, due to the way stall is structured this gen, you would possibly need to ban up to 3 things to ensure these mindless endless double switching scenarios stop occurring. I don't see why this is an issue we should ban around when 1. it has incredible collateral damage and 2. request draw caters to an uncommon matchup, but is going to be common in OLT, and we (the players) want to have an available solution that doesn't punish each other (not being punished for not wanting to play for 3 hours vs other stall).

Hope I cleared the issue up a little regarding why fixing the game is not an option at this point. Seriously, please just give us the ability to request draw, we lose nothing from it and it is massively beneficial for everyone. If we still run into only one player wanting to play and abusing the system, why not take further measures to prevent that after there are relevant cases that come up afterwards instead?
 

destinyunknown

Banned deucer.
2 people can maliciously consent to draw a game at the expense of a third one, or to give unnecessary help in some tasks like suspect tests.
This can already be abused on the ladder currently. If both players agree not to click any move, the battle room expires after a certain amount of time, and the game isn't counted, it's like it never happened. Some people had to resort to this during the Dugtrio suspect test, because there was no way of ending stall games (or it wasnt worth the time) so agreeing to that was a better solution.

Chess is fun without the option to offer draws. The threefold repetition and the 50-turn rule are, if anything, erring on the side of "making games short", and make some otherwise-winnable endgames unwinnable.


Fix. Your. Game. "Offer draw" is a cop-out, which would be fine but it's a cop-out that doesn't even fix the problem. If it's "drastic" to expect you to actually fix the problem, something is seriously wrong here.
But you can offer draw in chess too. Seeming as you're so intent in pushing the chess comparison, in chess there's both a rule that forces the draw, plus an option for players to agree to a draw when they so please, and it is in fact a common practice.

Why is an offer draw a 'cop-out' for pokemon, but not for chess? Your argument doesn't make any sense, it's clearly a good option for both games. In fact, offering draws in pokemon would be a much less common situation than in chess, so you're again wrong here.
 
Last edited:

Users Who Are Viewing This Thread (Users: 1, Guests: 0)

Top