Forum changes: Editing of posts has been turned off until further notice.

Main Menu

A Snowball's Chance in Hell (Pool Variant)

Started by Lxndr, July 18, 2003, 05:58:42 PM

Previous topic - Next topic


(Also posted to Indie-Netgaming)

Thought y'all might get a kick out of this.


(Snowball is a variant rules system based on 'The Pool' and 'The Questing Beast,' as well as other variants on the same.  It has only been playtested once, but I think it's a good idea, and worth sharing.  The name came from a desire to give this variant a water-related name, as well as the mental image of a snowball rolling down a hill, growing bigger as it reaches the bottom of the slope)

What is Snowball?

Snowball was created in the indie-netgaming irc channels, mostly by me (Alexander Cherry, Lxndr) but with suggestions and help from Shreyas Sampat and jrice_blue (not sure who the real name is), who were also the first two guinea-pigs.  The costs for traits have changed from the game last night, as I feel that the cost structure below is a bit stabler, however the characters in the current game will not be changed.  Other changes, clarifications and the like can also be found below.

Snowball was created to roughly mimic and emulate the style of story-creation seen in the movie "Memento," starting with the last scene and moving back towards the first in discrete intervals.  However, there's no reason that snowball couldn't be modified for any other construction, including forward-moving stories, and stories whose scenes are entirely out of order.  I'll try to address such things later, but the basic Snowball construction assumes the ending scene first, with everything else following after.

Like its progenitor games, Snowball is a role-playing game system geared towards narrative collaboration between a GM and one or more players.  Characters in Snowball are not defined by a list of carefully balanced attributes that match up to some objective standard.  Instead, they are defined with a series of traits whose only rating is, effectively, how important such traits are to the story.

In Snowball, the influence players have on events is not limited solely to the actions of their characters.  By invoking traits, suggesting ideas, and gambling dice, players can actively direct the story, briefly taking on the role of not just the GM, but sometimes the other players as well.

A story can be seen as a hike up a hill to get to the top.  In snowball, we start at the top, and roll down, gathering the story as we travel.

Before you Begin

One person in your group needs to be the Game Master, or GM - this is the person responsible for setting up scenes, creating situations, playing other characters and generally ensuring not only that the game runs smoothly, but that the story being told remains engaging and enjoyable to all participants.

You will also need a bunch of six sided dice, probably about ten per player, including a handful of "GM dice" that look different from the rest.

Snowball can be used in any sort of setting, but there should still be some sort of consensus.  Before the first game of Snowball, there should be a general idea as to the setting, as well as the opening of the first scene.  Do not be too detailed!  Both the setting and the story will be discovered and detailed in play, by players and GM alike.

One: Character Creation

Each character begins a name, a Trait (see below), and four dice in their Pool.  Everything else is mutable, waiting to be created.  Creating a backstory in Snowball is discouraged - as the scenes march backwards, the backstory of everything is what you'll be discovering.  In addition, adding Traits can be a risky business.

Two: Traits

A Trait is a short, definitive statement about your character, something important, something that makes your character who and what they are.  Traits can be anything - aspects of personality, flaws, skills, abilities, allies, possessions, beliefs, destinies - anything that helps you bring your character to life.  Make sure your Traits are specific enough to avoid conflicts over vagueness.  Be specific.

A starting character has a single Trait at level 3.  At the beginning of the story, this is the only thing that defines the character - further revelations will be added during play.  As this is the only Trait available to the character during the opening scene (the one which ends the story), and must be invoked at least once during that scene, it is destined to be a pivotal and meaningful part of your character's story.

Adding a new Trait costs dice out of the character's Pool.  Traits may only be raised one level at a time, at a cost equal to the new level.  Thus, purchasing a new Trait costs one die out of the Pool.  Raising a trait from two dice to three dice costs three dice, and so on.

Adding and changing Traits can only be done at certain times - between scenes, and during appropriate moments in play.  Every scene change, a character may add or raise one Trait one level without risk; they must still pay the dice.  Any Trait added in this manner must be invoked at least once during the upcoming scene, just like the starting trait.

A player, during a scene, may choose to add a new Trait, or increase an existing one, during an Idea or Conflict.  In a case like this, the character rolls the dice spent to add/raise the Trait instead of taking a Trait bonus (they may still gamble or be given additional dice by the GM).  If the character gets an MoV, they get the Trait added or raised and the dice are spent.  If the character gets an MoD, their Trait is corrupted or altered, but they still get it, and the dice are spent.  If the character gets a Guided Event, they may not try again to get or increase that Trait until the next scene.

If a character wants to reduce a Trait, they get dice back equal to the levels lost.  Reducing a level 3 trait to a level 2 trait gives back one die, and reducing a level 4 trait to a level 1 trait gives back three dice.  A character can only remove a Trait entirely during a scene change, and then only with the GM's permission, but otherwise Traits can be reduced at any time if extra dice are needed.  Generally, if a Trait is removed, it should be addressed somehow in the upcoming scene.

Three: Casting the Dice

Dice are cast to determine the general outcome of all conflicts and ideas.  This is not the same as rolling when you want to take an action - the effects of dice in Snowball are much broader than the swing of a sword.  

Anyone can call for a roll whenever a conflict is apparent, or someone wants to introduce a new conflict, complication, or other idea.  Before you roll the dice, you must state your Intent.  This is merely what you want to accomplish - it should be simple, and really short.  You must also state a Calamity, or a general short sentence about what happens if you fail - this must somehow complicate, confuse, muddle or corrupt your Intent.  If the GM calls for a roll from a player, the player is still responsible for stating Intent and Calamity.

You always roll six sided dice.  The GM will give you 0 to 5 dice depending how much he likes your idea and wants it to succeed, you can use a Trait bonus (if applicable), and you can gamble up to nine from your Pool.  You can invoke a Trait of any character, but only one Trait bonus may be used in any roll.

There are three possible results:

* If you roll a 1, it's good.  You get a Monologue of Victory, which must generally follow your stated Intent.
* If you roll a 6, with no ones, it's bad.  You make a Monologue of Defeat, which follows along your Calamity, and you lose any dice you may have gambled.
* If you roll neither a 6 or a 1, it's a Guided Event.  The GM may narrate the outcome any way he wants, following your Intent, your Calamity, or going off another way altogether.  You also get to add a die to your pool.

As you can see, adding dice to your roll greatly increases your chances of getting a 1.  But if you fail a roll, you lose all the dice you gambled.  A bad throw can instantly reduce your Pool to nothing.

A GM can declare a given conflict roll to require the creation of a new Trait.  See above for how that would work.

Four: Success or Failure

During a Monologue of Victory or Defeat, the player in question is entirely in control of the game for a few moments.  He can describe his character's actions, the actions of those around him, and the outcome of those actions.  He may also focus less on the direct elements, and describe things such as what's happening in the next room, or who's entering the scene.

You can do just about anything.  In fact, these are the only real limitations you must observe:

* Don't make alterations to the characters of other players.  You can add complications to them, and affect the things around them, but don't intrude on the creation of a fellow player.
* Keep your narration in synch with the established facts and tone of the game, as well as your Intent or Calamity.
* Keep your narration reasonably short.  The GM may end your Monologue at any time.

Five:  Scenes and the Story

The first scene in a Snowball scenario is always the final scene of the story.  Generally, this first scene should be established by a single paragraph, somewhere around 50 words in length, and leaves more questions than answers.  Keep in mind the Trait each player has chosen for their character, as well as anything agreed upon such as genre, setting, and so on.  An example from the one game run so far is below (the story seed was "The Death of the Green Man"):

Quote"Everyone's ears were ringing.  The pistol lay smoking on the ground where it had been thrown.  Blood was everywhere, the hired men scattering in fear.  And, lying amidst the sand on the slabs of the ruined temple, The Green Man's blood oozed from his ruined face, a hole in the base of his neck."

The first scene should be set up to provide some closure.  Each successive scene has the goal of "ending where the previous scene began." Successive scenes should be short and, like the first scene, start in a quickly-sketched situation that demands some sort of immediate action, like a bunch of mini-Kickers.  So the previous scene could involve a character being chased (or chasing someone), or them just arriving at the scene, or otherwise in a dramatic pose, in media res.  GMs should keep in mind any facts already established in the previous narration, as well as any Traits the players may have purchased between scenes.

In general, all the characters are to be together during the opening scene, though they are likely to be broken down into smaller groups as the scenes progress steadily earlier.  Of course, do with this bit of advice what you will.

Before each scene after the first, all characters who are to be involved in the next scene add a die to their pools.  Yes, this die can be spent immediately on a trait if the player so desires.

Appendix: A Bonus Suggested Pool Variant

I like the idea of anti-pool, as proposed by Mike Holmes.  I also like the setup in The Questing Beast, where there are three possible events instead of two (specifically, I like MoDs, but only with Guided Events to balance them out).  This particular setup combines the two, and has been nicknamed anti-beast.  I sort of like the name Whirlpool, though.  I'm not sure yet how it'd interact with the above rules - the phrase "a snowball's chance in hell" keeps running through my mind.

Anyway, flipping the beast, as far as I can tell from number-crunching, works best if done as follows:

* If you roll a 1, you get an MoV.  Any dice you gambled are lost.
* If you roll a 6, and no 1s, you get an MoD.  In addition, you get as many dice as the GM gave you for your nifty action (between 0 and 3).
* If you roll neither a 1 or a 6, it's a guided event.  Your dice remain the same.
Alexander Cherry, Twisted Confessions Game Design
Maker of many fine story-games!
Moderator of Indie Netgaming

James V. West

Cool! This is going up on the Pool Variations page as soon as my site it back up.


I've got a more-expanded pdf written.  It'll probably get revised a few more times, but let me know after I get back from gencon and I'll send you that instead.

Glad you enjoyed it, though.

Anyone (James or someone else) got any constructive criticism?
Alexander Cherry, Twisted Confessions Game Design
Maker of many fine story-games!
Moderator of Indie Netgaming

Shreyas Sampat

A Variant for Tainted Traits, by Jeremy Rice, with extension by me:

When you get an MoD on a Trait roll, your Trait is Tainted.  Whenever you recieve a Monologue with that Trait, every 6 you roll represents one manifestation of the Trait's Taint, with magnitude approximately equivalent to a Donjon 'fact.'

My revision to this:
Allow different and more digits to create Taint facts.
Dissociate the mechanic from Taint; the added facts are simply things out of the Monologuing player's control.

Lxndr is where it can be found, if you want to post a link to it, or a copy, James.  I'll be throwing it on the Twisted Confessions page, once it gets up and running, and that'll be its permanent home.

It still needs work, though.  

As for Tainted Traits:

My original thought, which I'm still somewhat enamored by, is just to change the text of the Trait slightly, and between that and the facts established in the MoD that results from the Taint, just fill in the gaps.  This, in my opinion, requires the least amount of bookkeeping.
Alexander Cherry, Twisted Confessions Game Design
Maker of many fine story-games!
Moderator of Indie Netgaming


That is the new, "permanent" home of Snowball, if you're still interested in linking to it on the Pool Variations.  Snowball now officially flips the beast.

I also have another small Variant that I humbly request to see added to the page, what I call "The Beast With Two Backs:"

Instead of losing dice on a Monologue of Defeat, you lose dice on a Guided Event, and gain a die on a Monologue of Defeat.  This way, the players only pay when the GM has to narrate instead of them.

The reason why I came up with it was, I believe, somewhat petty, but I think it makes sense, and the odds are good.
Alexander Cherry, Twisted Confessions Game Design
Maker of many fine story-games!
Moderator of Indie Netgaming

James V. West

I'll add those asap!

Why do you say the reason you came up with the idea is petty?


I say petty tongue-in-cheek, somewhat.  The Beast With Two Backs was created because I was thinking of a way such that the players were penalized when the GM actually had to do work (i.e. narrate a Guided Event).  That seems vaguely petty to me.  :)

(The other reason I came up with it was so that I'd have a perfect trilogy, adding Flipping the Beast and The Beast With Two Backs to the regular Beast.  But that's not nearly as petty.)
Alexander Cherry, Twisted Confessions Game Design
Maker of many fine story-games!
Moderator of Indie Netgaming


having tried out the puddle, i decided to try something more challenging & convinced people to try a snowball game. good move.

i obviously learned much from my previous attempt at running a pool variant, but most interesting was the fact that this was the most linear game i've run in years.

it started with a bang, in media res, then cut back to an hour before, scene three two hours earlier, carrying on for 5 scenes until shooting back to the end of the opening scene for the scene 6 denouement. the big twist happened in scene 4.

after the first scene i introduced scene two (previous action) also with a bang. so we knew how the scene started and how it finished. the game was linking the two together. from my POV when writing the scenario i had to know how to start scene 2 (& scene 3 etc), so it really was a linear adventure. if we'd put it in chronological order it would not have been nearly as much fun. the fun came from the players confusion as to what was going on (my response, 'you don't know, none of you know. perhaps you've been hit by a mindblast spell.')

so i scripted out a 5 scene linear scenario (or 5 opening scenes at least, linked together by a single plot), wrote 3 bangs to fill each scene and off we went. brilliant, simple and fun. in a fortnight i'll be running this game at a convention, so i'll report back on how a more varied bunch of people adapted to the set-up.

main thing the players and i agreed was the importance of getting things straight in the first scene. the mechanics become more comfortable in scenes two through six, but at first the players are disoriented and confused and need a good push into the action. so frame the first scene hard, give them pregen characters and don't give them any time to worry about stuff. take longer over each successive scene as they get more comfortable with the mechanics and the context of the scenario itself becomes established...

excellent fun.


Glad to hear you had fun, and that someone managed to use Snowball besides myself.  :)

Did you run it from what's in here, or from the pdf?

If the pdf, what "version"?  The rules themselves aren't being updated very much, but I'm still tinkering away with it.

Also, I'd like to include you in the playtesting credits, since that is basically what you're doing - what name should I credit you as?
Alexander Cherry, Twisted Confessions Game Design
Maker of many fine story-games!
Moderator of Indie Netgaming


i used the pdf via indie-netgaming. version 1.2 i believe.

one thing i would suggest, if you are still tinkering with the rules, is to work on clarity of presentation. its pretty good, but my players got into a muddle about the difference between outcomes for Ideas and Conflicts. having a tablular breakdown of event resolution for each would help, like:

1. state general idea, including Intent and Calamity & trait used
2. receive dice from narrator
3. enter gambled dice from pool
4. roll 'em
5. on a MoV do x
   on a MoD do y
   on a neutral result do z

for ease of quick reference would be valuable.

my name is guy jobbins - i keep forgetting i'm here under an alias!




Well, it's now at version 1.3 or so.  I've uploaded a new version to indie-netgaming (which I totally forgot about updating) and there's also a newer version at Twisted Confessions (which usually has the newest revision at any given time).

I had a couple non-gamer friends look Snowball over, and I've updated it to hopefully be more easily interpreted.  My next step is to include an actual example of play (probably a fictional one).

Your suggestion for a tabular breakdown is a good one, though, and I'll definitely make sure to include that in a future version.
Alexander Cherry, Twisted Confessions Game Design
Maker of many fine story-games!
Moderator of Indie Netgaming


Alexander Cherry, Twisted Confessions Game Design
Maker of many fine story-games!
Moderator of Indie Netgaming


Quote from: LxndrGuy,

How did the convention play go?

apologies for the delay in reporting back, but i'm happy to say that the convention game went really well. i just ran the one game, and we sat with a full complement of 6 players plus myself. it was a nice mixture of people: a couple were real livewires who went off on great tangents, there were a couple of quiet thoughtful types who pulled everything together, and couple somewhere in between. none had played snowball or pool variants before, although two had heard of the pool. it was a game set in glorantha, and all save one (one of the livewires) had a fair knowledge of that setting.

i didn't explain anything about rules before play began, other than assuring people it was really simple. i said that if anyone had an Idea during the first scene they should speak up and we'd go through an example of how the rules worked.

the first scene opened in media res, as usual. at first the players looked a little out of their comfort zone. i had about four bangs prepared for the first scene and led them through these when it looked as though the players wanted guidance. everytime they asked 'how did i get here', 'who are these people', 'what is going on' type questions i simply replied 'you can't remember'. a couple of Ideas came up during the first scene, non-critical ones, and this gave the players the chance the get used to the mechanics. the first scene lasted about 10 minutes.

as soon as we switched to scene 2 one of the players said, 'ah, memento!' and everyone got it. after that play went very smoothly indeed, and was a great deal of fun. the players really went off on one and did stuff i hadn't considered, and they managed to have it all tie up at the end.

the game was advertised as 3 hours of play, and we finished bang on the button, a real rarity in convention games which people were grateful for as they all had other things they were signed up to. despite the relative shortness of the game it ran a natural length and told a whole story with a great denouement. the players appeared very happy and decided not to tell others too much about it other than it was great fun, because they want others at other conventions to get to play it and not have the action spoiled.

i've uploaded a draft of the scenario to if you fancy a look




I know next to nothing about Glorantha, but damn if that's not a cool adventure.  An interesting twist on "the story going backwards" - the Snowball games I've played have just assumed that the characters know what's going on, but the players don't.  For a con, I can see how you'd want to tie in the non-linear, oddly-structured story into the character's perceptions as well.  Very nicely done.  :)
Alexander Cherry, Twisted Confessions Game Design
Maker of many fine story-games!
Moderator of Indie Netgaming