Miscellaneous > Mafia Game Threads

f.ds Mafia Game Creation Guide(s)

(1/4) > >>

ashersky:
Inspired by the rinkworks manifesto regarding fan cards for Dominion, here is my own game creation guide for f.ds mafia.

The text below is my own input, with additions from other experienced mods.  I will also link to contributions made in the comments that I either haven't assimilated below or may differ enough from my advice to fit better as a separate post.

For faust's thoughts, see here.
For scott_pilgrim's input, see here.
For silverspawn's argument against multiball (with which I may not agree), see here.

The first rule of mafia game design is you don't talk about...wait, no.  The first rule of mafia game design is that you can ignore all the rules.  Forum mafia is an unregulated genre of games that we all play for enjoyment and nothing else.  Even "officially" "sanctioned" "tournaments" are just regular mafia games played with a more diverse set of players.

So why write a guide at all?  Because for a lot of players, the amount of enjoyment garnered from a game is at least somewhat related to the design, balance, and theme of a game.

And why me?  At this point, cameos from golden oldies excepted, I am the longest tenured player/mod on the forum.  I've been honored to be chosen by other members of the forum as one of the best players, mods, and game designers around.  And lastly because I'm enamoured with the theory and meta of forum mafia, possibly even more than playing!

*These are my thoughts and opinions, and you don't even have to read them.  Other great mods will assuredly comment on them, both in agreement and disagreement.  You should definitely read those comments!*

A final note: this post is only about game design, not moderation.  As such, it does not discuss vote counts, in-game rulings, etc.

ashersky's tips, tricks, and warnings!

1. Know the game. Put in the simplest terms, you need to have experience as a player and a moderator before you can successfully design an original, spectacular game.  Ideally, you'll have played multiple games as both town and scum, as both vanilla and power roles, in the category of game you want to design.  There are multiple standard setups on f.ds that a player can moderation to gain experience.  The more familiar you are with the way games work, the better you can flesh out your own ideas.

2. Focus on the big picture. Mafia games need to be designed in full.  A designer's decisions about a role, power, etc. need to be made with every other role, power, etc. in mind.  Something that is cool in isolation may interact badly with a different part of the game.  Be ready to kill your darlings.  That one awesome idea you had might have set you on the path for this great setup, but maybe you need to excise that original thread for it to be successful.

3. Balance does not mean fair. Defining balance in the context of mafia games can be difficult.  A designer can think that a balanced game would result in every faction having an equal chance to win.  The reality of game design is that the addition of even a tiny amount of complexity will increase the instability of a game to the point of making it impossible to guarantee a specific win rate.  Just remember that that's okay.  Note: this is a contested topic.

4. Consider game-breaking strategies. Follow-the-Cop is the cautionary tale that all designers must memorize.  If you include roles or mechanics that can guarantee a faction a win, you need to think long and hard about the potential consequences.  A setup being closed does not mean the players won't figure out a way to break your game.  Keep in mind that often a strategy or interaction that causes a large enough swing in the favor of one faction, even without guaranteeing the win, is enough to ruin a game.

5. Ask yourself this question when considering a mechanic/role/power: What percentage of the time will a player choose to do X, given the consequences? Some of the most interesting, awesome roles or powers ever imagined are never included in games because the reward of using it never outweighs the risk.  A standard role that illustrates this important point is the Desperado: essentially a Day Vigilante, but making the wrong choice (shooting town) ends up in not just that townie's death, but your own.  The reward is high (unblockable kill, possibly taking out scum, settling a claim/counterclaim situation, etc.), and the risk is high (two town deaths).  The role has been successfully used in multiple setups.  This concept is often discussed in terms of positive and negative utility to a faction.  There are design choices to be made that include purposefully adding negative utility to a role, power, or faction.  Just be sure it is needed and thought out.

6. Be honest. Not just in ensuring the game you design doesn't stray into Bastard Mafia territory, but make sure your players can trust you.  Some of my own biggest failures in design have been the inclusion of roles or powers that the players honestly did not expect could be included.  Whether it was a recruitment power for scum in Dune or a ninja kill for scum in Modern Community without a complimentary town role, players expressed some level of unhappiness with the design feeling dishonest.  I don't regret the decisions I made, but I include this point for future game designers to consider -- think about the reaction of your players when they learn of a mechanic during or after the game.

7. Consider the open vs. semi-open vs. closed labels. Open  setups are the hardest to create in general, since the amount of balancing issues you have to consider increases exponentially with the complexity of the game.  Semi-open games can often act as templates to use (i.e., asher9++, which creates a different game each time and therefore have lots of replay value. Closed setups are generally the easiest to successfully run because even if you mess up balancing, it's unlikely to matter because people won't figure it out.

8. Provide as much information as you can.. Related to #s 6 and 7, you should endeavor to provide your players with as much information as possible.  While it may feel good, cool, smart, ingenious, or otherwise amazing to hide mechanic/role/setup information from the players, if only for the big reveal moment, it generally makes for a better game if players know what they are doing.  It is a harsh reality to realize, but often, those ah-ha moments only feel good for the designer, and leaves a bad taste in the mouths of the players.

*A Note on Flavor*

If you design a game around a specific flavor, a common theme here at f.ds, remember that others will know the flavor as well as you do.  Always take into consideration the extra care required to pull off thematically chosen or assigned powers, roles, and alignments -- not only for the game breaking possibilities someone with a lot of flavor knowledge could uncover, but also for the disagreements you might face from other die hard fans.  If flavor could give away a key part of the game, ensure there is a workaround, such as provided fake claims, or disclaimers in the setup information to dismiss the possibility.  To be clear: ensure there is no way a player is automatically cleared as town due to flavor name (unintentionally).

I think this is probably a work in progress, and I plan to add to it in the future.  I look forward to my fellow experienced mods adding to this reference guide, which I will reiterate is here not as a rule book, but a resource.

faust:
This is a very good idea!

I think probably a lot more can be said on the topic; if you go into the details, you can really get a guidebook the length of rinksworks' Fan Card Creation Guide.

Some comments and suggestions for additions:

- I don't think (3) is right. You should go for a game the has roughly equal win rate for each faction. A game having high variance doesn't mean that it cannot be fairly balanced from the start.

- (4) is probably one of the more crucial things. If you were to extend this guide, you'd spend a lot of time on different game-breaking things. This is especially important for an new mechanics you introduce into the game.

- I think ou should make a distinction between open/semi-open/closed setups. Open ones are the hardest to create in general, since the amount of balancing issues you have to consider increases exponentially with the complexity of the game. Semi-open games are basically templates to use like asher9++, which create a different game each time and there have lots of replayability value. Closed ones are fairly easy in that even if you mess up balancing, it's unlikely to matter because people won't figure it out.

- one could talk more about role creation. Like, when designing a new role, be sure that you have in mind how that power interacts with others. In general, you should be clear on how any two powers interact before the game starts. This is something I relatively often see made up on the go, which is a sign that you didn't think about the setup enough beforehand.

- we could have some links to well-designed games. This may be controversial I guess. But for example for open games, scott_pilgrim's Yoshi Island is the gold standard. Designing an open RMM that doesn't break and isn't boring is the hardest thing to do.

- For flavor, you should always make sure that some players aren't getting IC status due to their flavor name, unless you specifically want to have that interaction. I guess this is already implied in your post, but it could be clearer.

- there's also the usual "kill your darlings" advice. You may have started the design with a really cool role/interaction/mechanic in mind, then the game took shape and suddenly your amazing idea doesn't fit in. Maybe it just wasn't supposed to work! If the rest of the game looks fine, then that's worth a lot more.

- generally, one should aim to provide the players with as much information as possible. Mysterious roles are fine every once in a while, but it's usually better if the player knows exactly what their role does. It's better to be prepared for hypothetical interaction questions with hypothetical other roles that may not be in the game than to say "sorry; I cannot tell".

I'm sure there's more to say, but this is what came to my mind.

ashersky:
Great points!  I incorporated stuff, so please re-read!

Jimmmmm:
+1


In the spirit of (4) and (8), I would say unless you have a good reason not to, Publish the VT PM. Maybe all Town roles or even all roles, but the VT one is the most important. The worst is Day 1 arguments and creating semi-ICs based on PM wordings, and it seems to happen pretty regularly.



--- Quote from: faust on October 19, 2016, 05:04:09 am ---- I don't think (3) is right. You should go for a game the has roughly equal win rate for each faction. A game having high variance doesn't mean that it cannot be fairly balanced from the start.

--- End quote ---

I think it's somewhere in the middle. For example, SKs rarely win, but that's balanced (I guess in terms of expected utility) by the fact that it's a big achievement to win by yourself. It probably wouldn't be considered fair for a single player to have a 1/3 chance at winning. Obviously SKs are something of a special case and the more standard you want a game to be the more you should strive for equal chances for Town and scum. But I think in less standard games it's fine to not have equal winning chances if for the players with lower chances it feels less unfair and more High Risk, High Reward.

Roadrunner7671:
Not to be mean Ashersky, but I feel like you start a bunch of Mafia projects then they sort of die off. Are you sure you have the time to create a guide like this?

Navigation

[0] Message Index

[#] Next page