Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[WanderingWinder]

Here are my stats on this
Card           Effect With              Effect Without
Platinum      -3.71                      0.45
Gold           -0.81                      1.64
Silver          -0.04                      -1.03 
Copper        4.15                       -2.00
Colony        -3.71                      0.90
Province      -0.85                      2.71
Duchy         -0.93                      2.18
Estate         -0.32                      0.11
Curse          0.95                       -0.45


And I'm considered a fairly extreme big money player amongst higher-levelers.

[DG]

Good players have big 'effect without' stats on victory cards as they win by depleting three piles.

I consider the effect with/without stats to be generally unreliable anyway as they're dictated by not only how you play the card, but by the nature of the card and how everyone else plays it. The win rate without silver is surprisingly high overall so all the good players are probably going to have big minus for 'effect without' silver. It doesn't say anything about how they play with silver.

[Razzishi]

Good players have big 'effect without' stats on victory cards as they win by depleting three piles.

I consider the effect with/without stats to be generally unreliable anyway as they're dictated by not only how you play the card, but by the nature of the card and how everyone else plays it. The win rate without silver is surprisingly high overall so all the good players are probably going to have big minus for 'effect without' silver. It doesn't say anything about how they play with silver.

The effect with/out takes into consideration both your overall win percentage and the population's specific win percentage in determining what it thinks your specific win percentage for a card should be.  It then states the number of standard deviations different from that calculated amount that your results have been.  Your average (geometric perhaps, but some sort of average) Effect should be 0 for both, regardless of your win percentage.

[DG]

The effect with/out takes into consideration both your overall win percentage and the population's specific win percentage in determining what it thinks your specific win percentage for a card should be.  It then states the number of standard deviations different from that calculated amount that your results have been.  Your average (geometric perhaps, but some sort of average) Effect should be 0 for both, regardless of your win percentage.

Whatever that means, it doesn't happen.

[WanderingWinder]

Good players have big 'effect without' stats on victory cards as they win by depleting three piles.

I consider the effect with/without stats to be generally unreliable anyway as they're dictated by not only how you play the card, but by the nature of the card and how everyone else plays it. The win rate without silver is surprisingly high overall so all the good players are probably going to have big minus for 'effect without' silver. It doesn't say anything about how they play with silver.

The effect with/out takes into consideration both your overall win percentage and the population's specific win percentage in determining what it thinks your specific win percentage for a card should be.  It then states the number of standard deviations different from that calculated amount that your results have been.  Your average (geometric perhaps, but some sort of average) Effect should be 0 for both, regardless of your win percentage.

Essentially effect with takes your win percentage when you buy that card, then subtracts your overall win percentage, and divides by the standard deviation, then does the same for the overall group of players, then it subtracts those two results. Effect without is the same, except for games where that card was available and you did NOT buy it. So you really wouldn't expect these things to average out to 0 overall unless you played the cards relatively as well as the isotropic population as a whole. But that would mean you would have no strong or weak cards compared to the average player, which is a bit outlandish.

[tlloyd]

Witch Doctor
Action - Attack
Cost: 5

+1 Action
Reveal the top 5 cards of your deck. Put the revealed Curses into your hand. Put the other cards on top of your deck in any order.
Every other player gains a Curse.

Here's another version:

Witch Doctor
Action - Attack
Cost: $6

"+2 Cards
Choose One: trash any number of curses from your hand, gaining 1 VP [token] per curse trashed, OR each other player gains a curse and returns 1 VP [token] to the supply."

I think this card is interesting by itself, but creates some very fun interactions both with cursers and with other VP token cards. Thematically, I like the idea of a card that can dish out curses and get rid of them - hence Witch Doctor.

[DStu]

Good players have big 'effect without' stats on victory cards as they win by depleting three piles.

I consider the effect with/without stats to be generally unreliable anyway as they're dictated by not only how you play the card, but by the nature of the card and how everyone else plays it. The win rate without silver is surprisingly high overall so all the good players are probably going to have big minus for 'effect without' silver. It doesn't say anything about how they play with silver.

The effect with/out takes into consideration both your overall win percentage and the population's specific win percentage in determining what it thinks your specific win percentage for a card should be.  It then states the number of standard deviations different from that calculated amount that your results have been.  Your average (geometric perhaps, but some sort of average) Effect should be 0 for both, regardless of your win percentage.

Geometric average is the squareroot of the product. So m=sqrt(a*b). This won't be zero unless a or b is zero, and is not really a good idea if one of the factors might be negativ. Unless you want to have imaginary means.
I guess what should be zero (if at all) is some weighted average, where the weights are the standarddeviations on winrate with/wo, as the effect with/wo are somehow scaled with the standarddeviation.

[Razzishi]

Essentially effect with takes your win percentage when you buy that card, then subtracts your overall win percentage, and divides by the standard deviation, then does the same for the overall group of players, then it subtracts those two results. Effect without is the same, except for games where that card was available and you did NOT buy it. So you really wouldn't expect these things to average out to 0 overall unless you played the cards relatively as well as the isotropic population as a whole. But that would mean you would have no strong or weak cards compared to the average player, which is a bit outlandish.

I don't expect it to be zero by any sort of calculable average, I meant in the sense that you'll always have cards you play better and cards you play worse.  I suppose I stated it a bit more forcefully than I meant.  Because all the stats boil down to when you win or lose, there should be balance between positive and negative Effects of the same type.  You can't do better (or worse) than expected with every single card, because then you'd just have done better overall. 

And I certainly didn't mean to say that a card's Effect With was related in any way to its Effect Without, which I feel as though I might have been interpreted as meaning; I meant that if there are some positive Effects, there have to be negatives of the same type.

[WanderingWinder]

Yes, I think that the way everyone was reading you was that you were saying that for any given player and card, the average effect with and without that card is 0. Whereas what you meant to say (and correctly so) is that if you take every card for a given player, that player's effect with across every card should average out to 0, and the same with the effect without.

[Donald X.]

The set has been officially announced.

http://www.riograndegames.com/games.html?id=396

[Thisisnotasmile]

3 reactions wut.

[DStu]

The central theme is cards that do something immediately when you buy them or gain them.

You mean immediately in the sense of at least faster compared to other cards that do something at first at the next turn?

@Thisisnotasmile: Maybe they don't react on attacks, but on buys? Including their own...

[Thisisnotasmile]

Possibly. I'm just surprised to see the jump from roughly 1 reaction per expansion to 3. That's quite significant. And I guess the "do something immediately" implies that we're going to be getting more cards similar to Mint.

Also, "26 new Kingdom cards to Dominion, including 20 Actions, 3 Treasures, 3 Victory cards, and 3 Reactions. The central theme is cards that do something" implies three more dual-type cards, or maybe even a triple-type card or two.

[kn1tt3r]

The game descriptions sounds like it was written during a party with some friends and some more wine. Very simpatico. *g*

Btw: Is there sth. missing or does this just sort of match the overall babble of the text?

But in a big world like this one - big when compared to smaller worlds anyway, if such things exist;

[Glooble]

It reminds me of The Hitchhiker's Guide to the Galaxy.

I was imagine cards that say things like "when you buy this card, draw two cards" or something. It would certainly be a new mechanic. And some.. unusual interactions with Ironworks/Bridge I imagine.

[DStu]

Also, "26 new Kingdom cards to Dominion, including 20 Actions, 3 Treasures, 3 Victory cards, and 3 Reactions. The central theme is cards that do something" implies three more dual-type cards, or maybe even a triple-type card or two.

I was thinking that too, but then I thought 20 Actions (including 3 Action-Reaction) + 3 Victory + 3 Treasure = 26.

[Silverback]

Drawing cards in the buy phase does not make any sense to me (or am I missing something apart from Black Market?).
In fact I could imaging cards that give you virtual money or +buys, when you buy them. But they would have to be really bad when you draw them.

[Thisisnotasmile]

Oh yeah. So 20 Actions, 3 Treasures and 3 Victory. 3 Of which are also Reactions, but I don't think we can assume that they will all be Action-Reaction type cards.

[kn1tt3r]

I could imagine some quite expensive Victory card (like $5) that gives you just 1 VP but hands out 1 curse to each opponent when you buy it. Or any other action that attacks other players in some way when bought.

[guided]

WOOOOOOOOOOOO

[Detonator]

I'd like to see a treasure-reaction or a victory-reaction.

[WanderingWinder]

Lots of ideas.
"When you gain this card, every other player gains a curse"
"When you gain this card, you may trash a card from your hand"
"When you gain this card, +1 VP token" (unlikely)
"WYGTC, you may put it on top of your deck"
"WYGTC, you may gain (a silver)(another copy of this card)(an estate)(a duchy)"

Reactions
"When you gain a card, you may reveal this from your hand. If you do, set this aside on the table. At the beginning of your next hand +2 cards and discard this card." (or many similar things) - incidentally, this could be your first just reaction that's not also an action. Conceivably you could also have victory-reactions or treasure-reactions.
"When an opponent plays an attack card,  you may reveal this from your hand. If you do, gain a card costing up to $4 and discard this card"


I'm thinking these cards may have pretty big impacts on the very first two turns. I mean, whatever you do on turn one might actually impact turn 2. That's pretty crazy.

I'm actually most intrigued by the 3 victory cards. That's quite a few, and victory and this theme don't obviously go together to me.

[guided]

A possible workaday effect: "When you gain this card, +2 Buys"

[rinkworks]

Three Piler
Action - $2
When you buy this, +1 Buy and all cards cost $2 less, but not less than $0.

[DStu]

Three Piler
Action - $2
When you buy this, +1 Buy and all cards cost $2 less, but not less than $0.

Recursive!
Buy Three Piler -> You have still 1 buy and everything costs $2 less, especially Three Piler Costs $0, so buy some more Three Piler for free and buy whatever you want.