Roulette

[Lxndr]

The Wheel: A Game System Using Roulette

   

[*]Characters
   

      

[*]Have some sort of statistics (need to be defined) ranked by number (what is average?  needs to be defined).  This number defines the maximum number of chips a character can bet during any single conflict.
      [*]Have a "bank" which is their pile of chips.
      

         

[*]Each PC has his own colour of chips, as in real roulette
         [*]They bid chips out of their bank for conflict resolution (see below)
         [*]They also spend chips out of the bank to improve their characters (costs need to be worked out)
         [/list:u]
      [*]Have some sort of income-generation system(s)
         

            

[*]Lucky Number:  Every character has a lucky number that is automatically bid on every spin.  It is a "free" chip, but is otherwise treated as any other bid.
            [*]Some sort of "Flaw" system, where when that flaw comes up and impacts the story, chips flow into the bank. (Still considering)
      [/list:u]
   [/list:u]
   [*]The GM
   

      

[*]Has only a bank.
      

         

[*]Does not bid; just spends chips directly, setting a "target number."
         [*]Generates income solely from players losing bids.
         [*]Has no maximum
      [/list:u]
   [/list:u]
   [*]The Wheel
   

      

[*]Used for conflict resolution.  This is the entire conflict, rather than taking things "blow-by-blow."
      [list=1]
         [*]A conflict and its participants are defined.
         [*]Each participant (sas the GM) declares their maximum bid, and removes that number of chips from their bank.  Meanwhile, the GM removes a number of chips equal to the target number.
          [*]Any static bids are placed on the board with the proper colours.
         

            

[*]Lucky Numbers
            [*]other conditional chips (as yet undefined)
         [/list:u]
         [*]The wheel is spun, and the GM reveals his chosen target number.  Until the ball stops bouncing, all involved may place or change bids.  On my wheel at home, this takes about forty-five seconds.
         [*]After the ball stops bouncing, the number is called.
         

            

[*]Winnings are collected.
            [*]Chips whose bets failed to win are added directly to the GM's bank (including chips like Lucky Number) - these are not winnings
            [*]Chips whose bets were won are payed out using standard roulette rules
            [/list:u]
         [*]Success is determined.
         

            

[*]Success against the GM is determined by direct-spending.
            [*]Success against another PC is determined auction-style.
            [*]Only one's winnings may be spent; the "bank" is meaningless at this time.
            [*]Chips spent on success, by any individual, are lost, regardless of outcome.
            [*]Cooperating individuals may add their chips together against an opponent.
            [*]Any remaining chips are placed in the appropriate "bank".
         [/list:u]
         [*]Highest # of chips narrates their victory and the dispensation of everyone else.
         

            

[*]Ties resolve in favor of the first bidder
            [*]Yes, this means if the GM sets a target of 1, you have to spend TWO.
         [/list:u]
      [/list:u]
   [/list:u]   
[/list:o]
*Although the odds change somewhat, this game is equally useful for an American Wheel (two zeroes, 5.26% house advantage), a European Wheel (one zero, 2.7% house advantage), or a straight die roll (no zeroes, 0% house advantage).

[Lxndr]

As Mike Holmes said to me a few days ago (and I'm paraphrasing here) there's a "best combination" between a certain # of chips you want, and the # of chips you are able to (or are willing to) commit to your bid.

I'm still wanting this idea to turn into a game, so I spent some time crunching numbers.  SIMPLE numbers.  Bids between one and six chips, mostly.  And I discovered some interesting trends, and made some observations.

[*]Crunching these numbers is HARD.  I brute-forced only one through six and I'm not sure if I managed to find the "best combinations."  There are 179 different possible spots to place a bet, per chip, on an American table, broken down as follows:

[*]38 straight bets (including 0 and 00) - 1/38 chance, 36 chip payout
[*]59 split bets (including 1/0 and 3/00) - 2/38 chance, 18 chip payout
[*]13 street bets (including 2/0/00) - 3/38 chance, 12 chip payout
[*]46 square bets (no zeroes) - 4/38 chance, 9 chip payout
[*]11 sixiannes (no zeroes) - 6/38 chance, 6 chip payout
[*]3 different dozen bets - 12/38 chance, 3 chip payout
[*]3 different column bets - 12/38 chance, 3 chip payout
[*]2 different color bets - 16/38 chance, 2 chip payout
[*]2 different high/low bets - 16/38 chance, 2 chip payout
[*]2 different odd/even bets - 16/38 chance, 2 chip payout
[/list:u]
If anyone has any hints on how to figure out the "best" combination, especially assuming multiple chips (179 turns exponential mighty fast), please let me know.
[*]Because crunching the numbers is so hard, I'm once again tempted to keep the numbers low.  Making the "average" human 2 or 3 seems to speed up handling time, compared to having it be ten.  This also keeps difficulty numbers low.
[*]As long as you have more than one chip on the table, on an American wheel you have a 95% chance of breaking even, if that's your goal.  Two chips can usually bring back two chips.  Ten chips normally can bring back ten.  And so on.  But unless the player wants his bank to slowly shrink, or I change my current game framework, the aim will always be HIGHER than what's desired (as the minimum # of chips that needs to be spent to succed is one, regardless of bid).
[*]Getting only ONE chip higher is interesting, with what little I'm aware of.  With one chip on the table, the best bid only has a 47% chance of bringing back two.  With two chips on the table, the best bid's chance is 63% to bring three chips home.  But (and here's where I find things interesting), with three chips, your best chances of bringing in 4 are smaller (though your choices are "have a higher chance of bringing in SOMETHING" or "instead of 4, I'll bring in six").  This interesting hobgoblin continues up and down the line.
So, looking at it sideways, it's possibly your best decision might be to bid LESS than all the chips in your arsenal, once you're aware of the target number.  Then again, bidding more can increase either your chances of success even without breaking even, and/or your total winnings.
[*]Even knowing what the "best bid" is, a player still has to make a number of choices (do I want to maximize the chances of winning, the chances of breaking even, or both?).  So the current plan of "making the target number known before bidding" will remain.  This also seems to argue for the "fixed target number" as opposed to making "spins" with the GM opposed.
This, by the way, also has the GM making some difficult choices, since he's got to withdraw from HIS bank to set a target number at all, and he sets it without knowing how many chips they will bid (only how many chips they CAN bid).  Does he set it high, or low?  Too high, the player might not bid anything at all - he'll probably lose, unless his lucky number comes up, and the GM's bank is now that much smaller.  On the other hand, if the GM sets it too low, the player might just cake-walk all over it.  
Since the player has to spend one MORE chip than the GM, that means that even if the GM has nothing, the player has to bid and win at least one chip.  Most difficulties, I imagine, are going to be like attributes - low, between zero (routine) and five (difficult).
[*]The "lucky number" seems to be a pretty good equalizer, overall.  It ensures that a player can bid at any time, even if he has no chips.  It also allows the GM to collect one chip into his bank every time the player DOESN'T win it.  I still think the game needs some sort of income system other than the lucky number, though.  
[*]I am, on the other hand, pretty well convinced the game does NOT need any other sort of bonus chip - besides the lucky number, a player will never get any sort of bonus to an action.  Although it removes the fine control a GM has over such bonuses, it's seeming more inappropriate all the time - it feels more APPROPRIATE, for some reason, that PCs succeed or fail based on their own luck.  Though perhaps narrating the RESOLUTION of an action could generate income?  "Dude, that was cool," etc.
[*]On the other hand, perhaps sometimes bonus chips would make sense?
[/list:u]
(Still need to define:  Type, and # of statistics; player income system; a possible additional "loss" condition for losing all chips; whether or not there ARE multiple success levels and how they'd work; and chip costs for character advancement)

Any input on these comments, or anything else?

The idea I'm most concerned about, the most insecure about, is giving the GM a bank from which target numbers are SPENT.  If the players are lucky in their spins, then the GM might be the one wallowing in the welfare line.  But it has its advantages - the bookkeeping for conflicts is much simpler.  The biggest issue I can imagine is, well, more than one separate non-PC faction involved in a conflict.

The GM HAVING a bank (which is replenished by lost chips) is at least one reason why I'm not wanting the GM to also give out bonus chips at all - my gut feels like it'd make the GM talk out of both sides of his mouth.

Oh yeah, and Mike>  Why did you want to know if I could get chips stamped with a particular logo?

[Mike Holmes]

I still have very strong misgivings about all this. We don't seem to be communicating well about it, however, so I'm going to suggest that you playtest the game. I think that if you do so, you'll better be able to see the problems I'm talking about, and better yet, be able to see what might make for good solutions.

Custom logos just seemed like a cool idea. Could market them for the game, or have special kinds of chips, etc. For example, all your chips could be the same color, but you could have some chips that could only be used in certain kinds of conflicts, or something.

Mike

[Lxndr]

You're right that custom logos are a cool idea.  I'm definitely considering marketing them right along with the game (and handing them out at the ends of demonstrations, too - "Keep the character, and keep the chips too.").  I'm also considering selling personal-sized roulette wheels from my site, too, once this game is completed.

I'm also still having misgivings, to be honest.  I know my numbers aren't quite right (but I don't understand why you think larger stats and banks are a necessity for things to flow properly - you mentioned that in passing in an earlier post), and what little numbers I have crunched seem to be advantaging the GM to quite a drastic extent, in the long haul.

Typing this stuff out helps, which is why I'm still making posts and poking ideas with pointy sticks.  And reading other's results helps.  

Right now I'm enamored with two ideas in creating this setup, which may or may not be compatible, but certainly seem to be butting up against one another:

[*]The GM has a bank of chips, which is replenished on a 1:1 basis by a player losing any bid, including the "free" lucky chip
[*]A character's attributes defines only how much "effort" they can expend, not how much effort they DO expend (which is tracked through chips)
[/list:u]I've considered removing the "reduce the winnings to pay for success" portion - just compare the winnings directly against the target number instead, and then throw the whole wad back in the bank.  This slows bank reduction quite a bit.

This could still be interpreted as "effort" - it just seems that, if you have enough chips to secure a success, you don't NEED to worry about effort.  Which seems reasonable and fair to me, but makes things less fuzzy.  I think I might have fuzzed things up TOO much with that, though.

I'm still not sure exactly how to handle ties, though.  Not sure how often they'd come up (probably mostly in PC/PC conflicts).

I still want some sort of player-income system, so that their bank can be replenished, but I'd rather the bids themselves come only out of currently available funds.  So I'm hoping to be able to design this system such that the ONLY special chip is the lucky number.

Besides, since the GM gets every chip the player loses, it's of dubious intent if he gives the player a bonus chip, especially if the GM decides where to place it.  And tracking bonus chips separately seems... problematic.  Since you'd need to keep track of bonus chips separately for each player, I imagine the possibility of twice as many colors on the board.  Ouch.

So... without spending for success (but still spending on character improvement), players right now are only breaking even, on average.  My current musings for an income mechanic dances along the lines of the Nobilis "Restrictions."

(For those reading this who might be unaware, a Nobilis Restriction is a "Disadvantage" in GURPS terms.  Whenever it comes up AND impedes you or inconveniences you in some significant-to-the-story way, the character in Nobilis gets more miracle points.  It doesn't seem that much a stretch to apply the same principle to this game.)

[simon_hibbs]

Maybe I missed discussion of this somewhere along the line, but there's a major problem when you have people playing against each other on the same turn of the wheel. Whoever places their chips last can choose to cover, or duplicate part or all of the other player's bets, seriously skewing their probable relative results. That is, if I place my chips after you and we have the same number of chips, I could bet in exactly the same way as you, guaranteeing a standoff however the ball rolls. If I have slightly more chips, I can guarantee I will beat you if I place last.

Alternating placing chips doesn't work, because whoever alternates second can still cover the other player's bets. You need to have completely blind bidding, which probably means more than one table with results on each determined from the same roll of the wheel.


Simon Hibbs

[Mike Holmes]

We did cover this. That's part of why Alex is going with the GM setting target numbers instead of competing. When we were talking about actual on board competition, I had suggested that the bets be placed under cups so that the amounts would be concealed (for exactly the reasons you state).

Mike

[Lxndr]

We discussed that (though a lot of that discussion was on indie-netgaming, not here.  Right now, a big part of my answer is, "The GM doesn't make any bids himself."  So the only worry is player-vs-player action (cooperative actions don't count, for obvious reasons).

My current proposed solution is the following:

Each involved bidder declares how many chips they are able to bid, and the Lucky Number chips are placed on the wheel.  When the GM begins the wheel spinning, everybody may simultaneously place chips, move chips around, et cetera, until the ball stops bouncing.

With this, whoever "bids last" is really up to the players (and the "clock").  If I place on red, and YOU place on red, then I can move my chip.  And so can you.  So bids sort of alternate, but you can move bids after they're placed, and so on, until the ball stops bouncing, at which point you're stuck wherever the chip was.  This could become a game of "chase the chips around the table" if someone was resolutely determined to match every bet, guaranteeing a standoff, but it removes the "turn-based" structure entirely.

Another proposed solution involved the use of opaque cups, overturned.  Chips would be placed beneath the cups in secret, and the cups themselves would then be placed on the board.  In most cases, some of the cups placed on the board would be empty... "bluff cups" if you will.  So the person with more chips could only try to cover the other side's bets, but wouldn't know exactly where the cups would be placed (though he'd likely have a good guess).

An idea bouncing around in my head right now is to have different coloured roulette balls, such that player X and player Y would collect on two different numbers.  Thus, it's not as simple as covering.  I'm also toying with several other numerical solutions, so far without anything promising.

I'd have to playtest before I'm sure, but perhaps the one-spin resolution ideal is flawed?  On the other hand, perhaps it's not BAD that the game rewards the higher-ranked so resolutely?

Food for thought, either way.

[simon_hibbs]

The 'two balls' solution seems like the most elegent - much better than shuffling chips about. I'd hate that, I'm here to roleplay not play dumb hand-eye co-ordination games.

I'd have to playtest before I'm sure, but perhaps the one-spin resolution ideal is flawed?  On the other hand, perhaps it's not BAD that the game rewards the higher-ranked so resolutely?

The concept already involves a lot of effort for each spin, having multiple spins for a contest is a lot of game mechanical overhead on the game. Alos if the outcome is pretty deterministic, surely you'd want a lightweight mechanism?

Simon Hibbs

[Lxndr]

"Two balls" is rather elegant, I guess.  It was easy to come up with because my little roulette wheel came with two balls standard (though they're not different colours, I suppose a marker could clear that little problem up).

But I have to wonder how potentially bizarre it might get in three-cornered or four-cornered competitions (and how long a wheel might be willing to stand that many balls on it).  Also, I worry about telling potential players, "sorry, you don't have enough balls to play this game."  (Also, I'm not sure how well a cheap roulette wheel could stand up to multiple balls whacking on it at once).

So... I'm hoping to stick with one ball.  Perhaps there's a way to get multiple numbers out of one ball...?

[iago]

So... I'm hoping to stick with one ball.  Perhaps there's a way to get multiple numbers out of one ball...?

Well, if you only cared about numbers ranging from 0 to 9, you could look at the results as a two-digit number with the digits considered separately, but I imagine that throws everything all to hell.

[Lxndr]

I've had an afflatus, though whether it is a good one or not remains to be seen:

"Offset."

Before putting down chips, each side secretly writes down, and hands to the GM, an "offset," which can be a positive number, a negative number, or zero.  This determines how many steps away from the ball, on the wheel, their "real number" actually is.  Positive numbers are clockwise, negative numbers are counterclockwise, zero is (of course) the same number.

Since neither side would KNOW each other's offset while bidding, they could be fighting over the same spot (if their offsets are identical) or different spots (if their offsets are different).

This would allow a potentially infinite quantity of contestants in an action, and is a much easier "secret" to keep track of than actual bids.

Thoughts?

(I'm tempted, sometimes, to go with the method offered in The Pool - "player vs. player conflicts are never resolved with a roll" but that seems to go counter to the "philosophy" that the mechanics themselves seem to express.)

[Mike Holmes]

Beat me to it. I read Fred's post, and it occured to me that you could roll a die and add that many spaces for each character. But your method is better, I think (doesn't require a die for one).

In either case, however, players might still shadow each other, on the off chance that the randomized result comes up the same.

How about if the challenging player just has an offset of three? Once the wheel is spinning, there's no way to check what the number in question is for every bet that the non-offset player makes (and in any case, certain bets won't be shadowable in this way). Yes, the red and black bets would still be easy to shadow, but the players will know this and can do odd-even instead.

Mike

[Lxndr]

Well, as long as players aren't AWARE they're shadowing one another, while bidding, they can't strategize with that as a guarantee.  So my "reveal the offset only after the ball stops bouncing" seems to stop the major worry of shadowing (i.e. players purposefully strategizing in the "cover the other player's chips" approach).

I'm tempted to stick with the "randomly and/or player-generated offset" method, in part because it's expandable to three, four, six, twenty-three, etc., while a fixed offset is more difficult to expand beyond "2."

(And, of course, using the optional "dice instead of wheel" rules, you don't have to worry about any of this.  Each player would just roll his or her set of dice simultaneously after placing bids.)

Some notes, by the way, taken from a discussion on #indierpgs last night.  These are somewhat sloppy:

Statistics!  Finally!

[*]Sobriety - both mental and social seriousness and clear-headedness
[*]Guile - both physical and social treachery and sneakiness
[*]Nerve - both physical and mental strength and resilience
[*]People - connections, "who you know" (and how much they like you)
[/list:u]
(There is also "Life," a trait one can take multiple times.  Your life can be your job, your family, a hobby, a lover, whatever it is that makes your character say "X is my life."  Life is allowed to stack with any attribute, but if your opponent uses it against you, it subtracts what it would normally add.)

Attributes shouldn't be too static - they should move up and down as the story progresses.  Thoughts on this (still in progress):

[*]Whenever your lucky number is spun, the attribute being used goes down by one.  Conversely, whenever a zero is spun, the attribute being used goes up by one.  (A variation that makes attribute fluctuations much more common:  Whenever a contest succeeds using a particular attribute, that attribute goes down by 1.  Conversely, every failure increases it by one).
[*]Players can "sell back" any of his character's attribute values and get back the full chip value invested.  On the other hand, they can only raise an attribute value with winnings, before they're returned to the bank.
[*]Allow players, after failing a bid, to pay down the target number on a 1:1 basis out of the Bank, with the knowledge that EVERY chip used goes to the GM's Bank.  If the target number is reduced below the player's winnings (or is reduced to zero if the player had no winnings), the player wins.  Yes, this includes selling down attributes to do so.
[/list:u]
Another random thought:  how much SHOULD attributes cost to raise?  2:1 or 3:1 are my current thoughts.  linear attribute improvement!!!

[Mike Holmes]

Well, as long as players aren't AWARE they're shadowing one another, while bidding, they can't strategize with that as a guarantee. So my "reveal the offset only after the ball stops bouncing" seems to stop the major worry of shadowing (i.e. players purposefully strategizing in the "cover the other player's chips" approach).

If I have two options to bet on, each that have the same odds, but one of which shadows you which creates slightly better odds, why wouldn't I go for the shadow?

I'm tempted to stick with the "randomly and/or player-generated offset" method, in part because it's expandable to three, four, six, twenty-three, etc., while a fixed offset is more difficult to expand beyond "2."

??

Player 2 is offset 2, player 3 is offset 3, player 4 is offset 4, etc. How's that any different?

And this way is simpler because it doesn't require the recording. The GM just assigns each player his offset. Also, interestingly, with a garunteed offset, the player can still shadow another player meaning that he's creating an "all or nothing" sort of environment with respect to that player. Which is an interesting gamble itself.

So, I see your stats - what's the action of the game about?

Mike

[Lxndr]

Well, as long as players aren't AWARE they're shadowing one another, while bidding, they can't strategize with that as a guarantee. So my "reveal the offset only after the ball stops bouncing" seems to stop the major worry of shadowing (i.e. players purposefully strategizing in the "cover the other player's chips" approach).

If I have two options to bet on, each that have the same odds, but one of which shadows you which creates slightly better odds, why wouldn't I go for the shadow?

Well, if you have two options to bet on, each that have the same odds, and you're unaware which one (if either) shadows the other person, which are you going to pick?  Am I missing something?  ::is puzzled::

I'm tempted to stick with the "randomly and/or player-generated offset" method, in part because it's expandable to three, four, six, twenty-three, etc., while a fixed offset is more difficult to expand beyond "2."

??

Player 2 is offset 2, player 3 is offset 3, player 4 is offset 4, etc. How's that any different?

Ack.  And the simple solution eludes me.  I thank you.
(Though I'd say player 1 is offset 0, 2 is offset 1, etc.)

And this way is simpler because it doesn't require the recording. The GM just assigns each player his offset. Also, interestingly, with a garunteed offset, the player can still shadow another player meaning that he's creating an "all or nothing" sort of environment with respect to that player. Which is an interesting gamble itself.

But if you assign each player his offset, then it's easier to strategize around it and shadow someone, if that's their goal, nu?  Hm.

So, I see your stats - what's the action of the game about?

I really don't know how to answer that question.  It's still a work in progress, plus I'm not sure quite what you're asking (honestly).

I'm imagining the game is "about" romantic, cinematic, free-wheeling, life in the fast lane, where luck is capricious and it's as easy to crash and burn as it is to come out on top.