You are viewing the MafiaScum.net Wiki. To play the game, visit the forum.

Callforjudgement's Micro Deadlines: Difference between revisions

From MafiaWiki
Jump to navigation Jump to search
(it's about time I put this on the wiki rather than trying to summarise it in every game; let's have a summary so that I can link it in game threads, and an explanation for people who are interested)
 
(some minor changes to the policy based on experience over several games)
 
(One intermediate revision by the same user not shown)
Line 4: Line 4:
|Author=callforjudgement}}
|Author=callforjudgement}}


This page describes the deadline policies I, [[callforjudgement]], use in Micro games.
This page describes the deadline policies I, [[callforjudgement]], use in [[Micro Game]]s.


== The Deadline Policy ==
== The Deadline Policy ==
Line 10: Line 10:
* Prods happen '''more often''' than in most games, but have a '''lower impact'''.
* Prods happen '''more often''' than in most games, but have a '''lower impact'''.
** If you do not post for 32 hours, you will be [[prod]]ded.
** If you do not post for 32 hours, you will be [[prod]]ded.
** Failure to respond to a prod causes you to be force-[[replaced]]. The amount of response time you are given will be lower each time.
** Failure to respond to a prod causes you to be [[replaced]]. The amount of response time you are given will typically be lower each time. (Note that unlike other activity replacements, responding to a prod late (but before you are replaced) allows you to continue playing in the slot.)
*** ''First prod'': 32 hours before the prod, plus another 48 hours to respond to it
*** ''First prod'': 32 hours before the prod, plus another 48 hours to respond to it
*** ''Second prod'': 32 hours before the prod, plus another 36 hours to respond to it
*** ''Second prod'': 32 hours before the prod, plus another 36 hours to respond to it
*** ''Third prod'': 32 hours before the prod, plus another 24 hours to respond to it
*** ''Third prod'': 32 hours before the prod, plus another 24 hours to respond to it
*** ''Fourth prod'': 32 hours before the prod, plus another 12 hours to respond to it
*** ''Fourth prod'': 32 hours before the prod, plus another 24 hours to respond to it
*** ''Fifth prod'': immediate force-replacement
*** ''Fifth prod'': immediate force-replacement, even if you subsequently come back to the thread.
** A prod response post must: a) express at least one read on a living player; b) ask a question of a player; and/or c) answer a question asked by another player, in order to count.
** A prod response post must: a) express at least one read on a living player; b) ask a question of a player; and/or c) answer a question asked by another player, in order to count.
** Being prodded three times ''in a Day'' also causes you to be force-replaced.
** Being prodded three times ''in a Day'' also causes you to be force-replaced, even if you subsequently come back to the thread.
** It is envisaged that even players who are keeping a reasonable activity schedule may end up being prodded now and again; in particular, there isn't an obligation to avoid prods (merely an obligation to stay sufficiently active).
** It is envisaged that even players who are keeping a reasonable activity schedule may end up being prodded now and again; in particular, there isn't an obligation to avoid prods (merely an obligation to stay sufficiently active).
** Players will not be prodded during a declared V/LA. However, as usual, the moderator can disallow excessively long V/LAs.
** Players will not be prodded during a declared V/LA. However, as usual, the moderator can disallow excessively long V/LAs. (The moderator will typically send activity prompts to players who are on V/LA and not posting, but these are not official prods.)
* Day deadlines are '''initially short''' but '''can be extended'''.
* Day deadlines are '''initially short''' but '''can be extended'''.
** The ''initial'' length of each Day deadline will be 96 hours. Note that this is just an ''initial'' length; it's typical for the deadline to be repeatedly extended (meaning that the game will have a pace more like a [[Newbie]] game than a [[Blitz]] game).
** The ''initial'' length of each Day deadline will be 96 hours. Note that this is just an ''initial'' length; it's typical for the deadline to be repeatedly extended (meaning that the game will have a pace more like a [[Newbie]] game than a [[Blitz]] game).
Line 47: Line 47:
Unfortunately, tests with 24 hour prods turned out to be a little ''too'' fast for players to be comfortable, even when the actual consequences of the prods in question were minimal (and even when they were given a name other than "prods"). 32 hours appears to be a good compromise, meaning that players can post at different times each day and yet still avoid getting prods at all (even though getting prods, under a system like this, is not something to be ashamed of).
Unfortunately, tests with 24 hour prods turned out to be a little ''too'' fast for players to be comfortable, even when the actual consequences of the prods in question were minimal (and even when they were given a name other than "prods"). 32 hours appears to be a good compromise, meaning that players can post at different times each day and yet still avoid getting prods at all (even though getting prods, under a system like this, is not something to be ashamed of).


Another potential issue is players who seem to stall intentionally, taking advantage of the entire prod timer and/or not posting unless prodded. It's important to make those players post (and more so towards the end of the game) to give other players a chance to read them. As such, I prefer to have my prod timers get shorter over time (which also allows for a natural "OK, you've been prodded too much, I'm force-replacing you" cutoff once the prod response timer reaches zero). Because 32 hours for the initial prod is a relatively fixed time in a system like this, that only gives room for the prod response timer to get shorter.
Another potential issue is players who seem to stall intentionally, taking advantage of the entire prod timer and/or not posting unless prodded. It's important to make those players post (and more so towards the end of the game) to give other players a chance to read them. As such, I prefer to have my prod timers get shorter over time (which also allows for a natural "OK, you've been prodded too much, I'm force-replacing you" cutoff once the prod response timer reaches zero). Because 32 hours for the initial prod is a relatively fixed time in a system like this, that only gives room for the prod response timer to get shorter. (The prod response timer cannot, however, go below 24 hours, in order to prevent timezone issues with when prods are sent.)


Such players also have a habit of repeatedly asking for deadline extensions (even in a game with more normal deadline rules). Extending deadlines to get more discussion is one thing, but extending deadlines to get more stalling is another. Active players are more likely to use their deadline extensions for the good of the game, but if a player's being repeatedly prodded, that's less certain. So it makes sense to give players who are avoiding prods more scope to control the deadline, which in turn means that prods should make the deadline extensions marginally less effective. This also means that if the game extends for a long time, that at least means that it should be full of activity, whereas slow days are cut off faster.
Such players also have a habit of repeatedly asking for deadline extensions (even in a game with more normal deadline rules). Extending deadlines to get more discussion is one thing, but extending deadlines to get more stalling is another. Active players are more likely to use their deadline extensions for the good of the game, but if a player's being repeatedly prodded, that's less certain. So it makes sense to give players who are avoiding prods more scope to control the deadline, which in turn means that prods should make the deadline extensions marginally less effective. This also means that if the game extends for a long time, that at least means that it should be full of activity, whereas slow days are cut off faster.

Latest revision as of 05:15, 29 January 2019

Type:
Author:

This page describes the deadline policies I, callforjudgement, use in Micro Games.

The Deadline Policy

  • Prods happen more often than in most games, but have a lower impact.
    • If you do not post for 32 hours, you will be prodded.
    • Failure to respond to a prod causes you to be replaced. The amount of response time you are given will typically be lower each time. (Note that unlike other activity replacements, responding to a prod late (but before you are replaced) allows you to continue playing in the slot.)
      • First prod: 32 hours before the prod, plus another 48 hours to respond to it
      • Second prod: 32 hours before the prod, plus another 36 hours to respond to it
      • Third prod: 32 hours before the prod, plus another 24 hours to respond to it
      • Fourth prod: 32 hours before the prod, plus another 24 hours to respond to it
      • Fifth prod: immediate force-replacement, even if you subsequently come back to the thread.
    • A prod response post must: a) express at least one read on a living player; b) ask a question of a player; and/or c) answer a question asked by another player, in order to count.
    • Being prodded three times in a Day also causes you to be force-replaced, even if you subsequently come back to the thread.
    • It is envisaged that even players who are keeping a reasonable activity schedule may end up being prodded now and again; in particular, there isn't an obligation to avoid prods (merely an obligation to stay sufficiently active).
    • Players will not be prodded during a declared V/LA. However, as usual, the moderator can disallow excessively long V/LAs. (The moderator will typically send activity prompts to players who are on V/LA and not posting, but these are not official prods.)
  • Day deadlines are initially short but can be extended.
    • The initial length of each Day deadline will be 96 hours. Note that this is just an initial length; it's typical for the deadline to be repeatedly extended (meaning that the game will have a pace more like a Newbie game than a Blitz game).
    • Each player has a one-shot deadline extension, which can be used by posting "Extend deadline" (or similar) in the game thread. The effectiveness of this depends on the number of prods the player using the extension has received:
      • No prods: deadline will be extended by 64 hours
      • 1 prod: deadline will be extended by 56 hours
      • 2 prods: deadline will be extended by 48 hours
      • 3 prods: deadline will be extended by 40 hours
      • 4 prods: deadline will be extended by 32 hours
    • If a player replaces in, they will replace in with no prods and a deadline extension active. Because of this, there is no automatic extension when a player replaces in (but they can if they wish immediately use their extension to give themself time to catch up).
  • Night deadlines are 48 hours long.

Explanation

Different sizes of games have different issues from a moderator's point of view. In Large Games, a major problem as of 2018 is spamposting; while a player is asleep, the other players can post so much that they have no real chance of catching up back to the new end of the thread by the time they wake up. Micro Games, however, often have the opposite problem, in which a large amount of moderator intervention is necessary to make anything happen at all, with all the players waiting for each other or the like. I've seen various ad-hoc solutions tried to this over the years, but think it's best to have an objective policy.

Deadlines

Bankable deadlines are one approach to discouraging players from stalling out a game to near deadline, but they have two problems. One is that the "default" behaviour if players ignore the deadline is to burn up the time from the future and to use it on the current day, which is pretty much the last thing you want when the game is stalling. The other is that they can lead to absurd results sometimes when there are quickhammers or the like. A similar approach is to give Town a reward when they end the day early; the last time I played in a game like that, it effectively degenerated into a hyper-Blitz game in which players attempted to end every Day within 24 hours, which was far too stressful to really be enjoyable.

The opposite approach is to start the deadlines short but give the players control over extending them; the deadlines then effectively become "bankable" in that if you don't extend a deadline on one day, you have the opportunity to extend a different day's deadline instead. I've experimented with a points-based system for this in the past, but the simplest method is to give everyone a one-shot deadline extension. Another advantage of this is that "banked" deadlines become self-correcting; there cannot possibly be more than 3 deadline extensions remaining in 3p lylo (and may well be none left; this is reasonable, though, as very small endgames benefit from having short deadlines anyway). The biggest advantage, though, is that players actually have to post the deadline extension in thread when the deadline runs low (and because the extensions are individually fairly small it's always running low), meaning that the thread gets repeatedly bumped and giving players something to discuss when the game stalls (why was it that player who used the extension).

Prodding

However, games stalling out because the players have nothing to talk about is only one problem. Another problem is players who don't devote enough attention to the game; it can be that there are lots of players there ready to play, but they're waiting on one player in particular. In my standard rules, I have "you should aim to post almost every day". On the other hand, typical prod timers work quite differently from that (letting players repeatedly wait much longer between posts, but putting large penalties on them if they go too far over the limit). So the logical thing to do is simply to enforce that rule; count the days on which a player doesn't post (i.e. prods), and replace a player who lets too many of those occur.

Unfortunately, tests with 24 hour prods turned out to be a little too fast for players to be comfortable, even when the actual consequences of the prods in question were minimal (and even when they were given a name other than "prods"). 32 hours appears to be a good compromise, meaning that players can post at different times each day and yet still avoid getting prods at all (even though getting prods, under a system like this, is not something to be ashamed of).

Another potential issue is players who seem to stall intentionally, taking advantage of the entire prod timer and/or not posting unless prodded. It's important to make those players post (and more so towards the end of the game) to give other players a chance to read them. As such, I prefer to have my prod timers get shorter over time (which also allows for a natural "OK, you've been prodded too much, I'm force-replacing you" cutoff once the prod response timer reaches zero). Because 32 hours for the initial prod is a relatively fixed time in a system like this, that only gives room for the prod response timer to get shorter. (The prod response timer cannot, however, go below 24 hours, in order to prevent timezone issues with when prods are sent.)

Such players also have a habit of repeatedly asking for deadline extensions (even in a game with more normal deadline rules). Extending deadlines to get more discussion is one thing, but extending deadlines to get more stalling is another. Active players are more likely to use their deadline extensions for the good of the game, but if a player's being repeatedly prodded, that's less certain. So it makes sense to give players who are avoiding prods more scope to control the deadline, which in turn means that prods should make the deadline extensions marginally less effective. This also means that if the game extends for a long time, that at least means that it should be full of activity, whereas slow days are cut off faster.

Conclusions

The final system is shown above: fast prods, initially lenient prod response times that get stricter as a player gets prodded more often, initially short deadlines (but long enough to make gameplay possible in an emergency), and 1-shot deadline extensions that are more effective for players with fewer prods. A nice side effect of this is that it handles "replace-in extensions" in a way consistent with everything else, without needing a special rule for them. (It should be pointed out, though, that since I started using these rules I've hardly needed to replace people!)

I'm unsure if deadline rules like this would work in larger games. I'm guessing that they'd continue to solve the problems that they're designed to solve, but on the other hand, they'd fail to solve the problems that larger games have and smaller games don't. Some other sort of deadline/posting rules would likely be helpful for that. (That said, games under the Geriatric Ruleset, which is designed to help prevent the spam-posting issue, sometimes stall out in much the same way that Micros do. So maybe using elements from this scheme would help in larger games after all.)