[Regency Park] Playing Card Resolution and Statistics -Help?

[Trevis Martin]

Hi Folks

I've been working in bits and pieces on my game Regency Park.  It is a Game of love, marraige and social drama set during the English Regency period and heavily influenced by the novels of Jane Austen.  It asks questions about what you would do for love, specifically how far will you risk your social status and that of your family.

I'm working on a card based mechanic inspired by the many card based mechanics around but I don't know how to statistically analyze it.  This is how it works.

When a conflict is declared the player announces what thier intent is and the ability they will use in the conflict.
Abilities are rated Poor, Average, Good, Great, Exceptional and Extraordinary  (though I'm thinking about limiting it to the first four.)

Each player draws cards according to the ability they are using,

Pr = 1 card
Av = 2 cards
Gd = 3
Gt = 4
Exc = 5
Ext = 6

Each conflict has a trump based on what kind of conflict it is.  The trump for social conflicts is Spades, for Emotional is Hearts, for physical is Clubs and for Mental is diamonds.  (I'm also thinking of stripping this down to only Social = Red/ Physical = Black)

You show all your cards. High card wins the conflict with the exception that trumps beat everything not trump. 

Narration goes to high card regardless of trump.

Does this show any significant statistical problems?  I know only the basics of statistics and I don't know how to analyze the probabilities of winning a conflict here

Can anyone help me out with that?

best

Trevis

[Thededine]

What happens in the case of a tie?

[Trevis Martin]

Hi

In case of a tie those who are tied draw a tiebreaker card.  The other  answer I'm kicking around is to rank suits in reverse alphabetical Spades Hearts Diamonds Clubs.  Spades highest, but I think a tiebreaker would be easier.

best

Trevis.

[MikeSands]

In case of a tie those who are tied draw a tiebreaker card.  The other  answer I'm kicking around is to rank suits in reverse alphabetical Spades Hearts Diamonds Clubs.  Spades highest, but I think a tiebreaker would be easier.

I think it's generally better to reduce the actions required - setting suit order means that one play always results in a winner. Extras can bog the flow down, so you should only use that if you *want* play to slow a bit in these cases. Unless there's some game effect of a draw and tiebreaker, which would make it worth the extra work. Especially as you would get second ties every now and again.

Also, I'd say that the suit order should match that of a card game that's in period - as I recall, bridge had not been invented then but whist might have been played. In any case, a small amount of research should turn up what the suit order in a relevant popular game was, and gives an extra little bit of colour to the mechanic.

[Warren]

I'm really interested in this, it sound fascinating, although the card & trumps system reminds me of Castle Falkenstein, IIRC. I like Mike's idea of making the suit order match work in the same way as a period game to add extra flavour. If you do use trumps, I'd say swap Spades & Diamonds around, mainly due to the fact I think Diamonds = Money = Social Status.

Anyway, Bridge wasn't around back then, but Whist - it's forerunner - certainly was. In that, the trump for any given hand is determined randomly by the suit of the last card remaining in the deck after all the other cards have been dealt. I quite like that idea in the game too, or maybe determining the trump suit by the arena of the conflict, i.e. "Do I take the rejection from Mr.Hansforth with good humour?" could be an Emotional conflict, and hence Hearts are trumps.

For general period card game information, this resource might help: An Evening Card Party in Regency England.

I'd like to know more about this project - it sounds facinating.

Warren

[Trevis Martin]

Hey Thanks for the link there Warren,

I'm very fond of Castle Falkenstein with the exception that I never liked the way the card system worked with all the numbers.  What I did like about it is the sort of period flavor that cards give for a game like this.  A purely aesthetic thing, but I like it.  In fact I've been gearing up to play CF again with the original system and it's feeling a little weird.  I'm spotting some huge 'how to play' holes in the system that I'm not used to, haveing played all these tight indie games.

I like the trumps but I don't want them to have to play a whole game to resolve stuff of course.  The GM could randomly draw a card to determine trump or we can stick with categories.  It really has to wait until the abilities/traits/resources type thing gels in my head.  I'm not thinking there is going to be a lot of those. I think as an idea, once people draw, they can just play their stongest card.  That will give them the option of playing to win the conflict or playing to narrate it.

Hm, after the whist link I'm getting other wierd ideas.

Mike, point taken about the extra step.  I was trying to decide weather it would be simpler to pull an extra card or remember some suit order.   Warren, I'll put together a post that has some more info about what I'm trying to do with this b/c there's some other questions I'd like to ask.

best

Trevis

[Warren]

Small point, but in Whist the trump card is the one at the bottom of a deck because that's how the cards are dealt initially. In your game, it would be the top card off the deck after all the 'ability' cards are drawn.

i.e. I've got Average 'stubborness' vs. your Good 'charm' trait. The Dealer hands the two cards to me and three to you face down, from the top of the deck. He would then take the top card from the deck and turn it face up to indicate the trump suit in this conflict. After the conflict is resolved all those cards are returned to the deck - which is then shuffled, maybe?

You could even rotate the Dealer position every conflict, and give them the GMs roles and responsibilites for that scene, if you really wanted to. Just a thought, anyway.

I'm looking forward to hearing more information about Regency Park, and I hope I can be of some help.

Thanks,

Warren

[Black Iris Dancer]

I love card mechanics *^.^* I like how their handling and revelation are different from dice; partially, I especially like that cards have connotations which can be drawn upon. In short, I'm liking your system so far. Are the jokers still in the deck (I assumed they weren't, in the calculations below, but nothing should go too far wrong even if they are), and if so, do they do anything special? Shakespeare's a bit earlier than the period you're going for, but the Fool crops up all over the place, so it seems like there might be interesting ground to explore there. Also, is there anything particular about how the cards are presented, or are they laid down for all to see immediately? Having them start face down and slowly get turned over seems like a good potential tension-builder, if it works with the other mechanics and feel of the game.

Anyway, um, you asked about card statistics. I cooked more statistics than you probably care about...

 
 
 
 
 
 
 

	If I am Poor...	Average	Good	Great	Exceptional	Extraordinary
And my opponent is Poor...	50%	67%	75%	80%	83%	86%
Average	33%	50%	60%	66%	71%	75%
Good	25%	40%	50%	57%	62%	67%
Great	20%	33%	43%	50%	56%	60%
Exceptional	17%	28%	38%	44%	50%	55%
Extraordinary	14%	25%	33%	40%	45%	50%

If my program's crunching is to be believed (and I'm pretty sure it is), the percentages in the table are indicative of the chance an ability with the score listed in the column has in conflict with an ability listed in the row. These numbers are based on the assumption that you have a suit ordering. Trump then doesn't actually have an effect on probability of victory, but I like the mechanic's colorful, and it does have an effect on narration. How much of an effect is twisting my brain a bit at the moment, but I'd guess the loser narrates something like a quarter of the time in evenly-matched conflicts, and the likelihood falls off with a binomial curve much like success as the conflict gets more unbalanced.

Hope that's of some help...

[Trevis Martin]

Wow, thanks Black Iris Dancer.  That is very helpful.

Trump makes no difference at all, statistics wise?  Hm, had no idea.

[Walt Freitag]

Yeah, trump makes no difference.

The statistics here are very easy to understand. In any group of 2 or more cards, one card will be the winner. With or without trumps, with or without suit ranks or redraws(1) for breaking ties, with or without wildcards or jokers. The figures in BID's table will remain true as long as your system gives each card a value independent of other factors such as matches with other cards drawn or relationships between the card and character stats.

So, if A draws two and B draws three, that's five cards drawn, one of which is the winner. A's two cards each have one chance in five of being the winner. Each of B's cards has one chance in five of being the winner, hence A wins 40% (= 2/5) and B wins 60% (= 3/5). Hence, the entires in the table for Average vs. Good. All the other combinations can be analyzed with similar ease.

- Walt

(1) One fiddly technical detail: the odds will be affected slightly if the tiebreaker procedure is that each player redraws a single card regardless of how many tied cards that player had originally. So for example, suppose one player has a non-trump jack and the other has two non-trump jacks and neither has any higher-value cards. If the procedure is that each player draws one tiebreaker card, then the overall advantage of drawing more cards is very slightly reduced. If the procedure is that the player with the two jacks would get two redraws, then that effect goes away and the odds are once again exactly as given in the table.

[Warren]

Don't trumps make a difference if one player chooses the Trump suit after he has seen his cards? And, if the rules are that you have to follow the lead suit unless you have no cards of that suit (as in Whist and Bridge, for example), the person who leads would normally have an advantage.

i.e. I am "Good" and get 3 cards. I get given JC, 9H and 5C. If I get to choose the trumps, I'd pick "clubs" and lead the Jack, there are only three cards (Q,K,A of clubs) out of the remaining 49 cards left that can beat me. (I win 94% of the time).

If, on the other hand, I choose "clubs" as trumps and the other person leads, then I could be in trouble. If they lead a diamond or a spade (53% of the time), I cannot follow suit, and would play a trump card and win. If, on the other hand, they led a heart, I would have to play my 9, which may or may not not win (wins 58% of the time). If they led a club, I'm a good favorite with my Jack (72% winner). That means my odds of winning overall have dropped to 84%. (Assuming, of course, that my opponent leads randomly).

Just to add a bit more input into all of this.

[Walt Freitag]

Sure, Warren. If you have a player naming the trump suit after seeing his cards, and play in which one player leads and the other must follow suit, then trumps have major effects. But the mechanism Trevis described doesn't have any of those elements.

Leading and following suit make things a lot more complex, and introduce some odd effects. For instance, if one player declares the trump suit and the other player leads, then the declarer might (depending on how many cards the other player has) be more likely to win with one card than with two, because that guarantees he can play trump.

- Walt