Dominion Strategy Forum

Dominion => Dominion General Discussion => Topic started by: GendoIkari on September 26, 2012, 05:19:56 pm

Title: Attack definition
Post by: GendoIkari on September 26, 2012, 05:19:56 pm
This discussion (http://forum.dominionstrategy.com/index.php?topic=4724.msg110316#msg110316) got me thinking... yes, of course we know that in Dominion an attack card is anything that says "attack" at the bottom. But I agree with eHalcyon's definition that an attack is something that attempts to hurt others. But that's still subjective. I think we can quantify it, though...

If playing a certain card during a game causes, on average, an opponent's deck to be slower*, that card should be an attack.

*Slower is defined as taking more turns to accomplish the same thing (whether that thing is gaining 5 Provinces, or reaching some other sort of Engine that wins the game).

So with this, Masquerade and Tribute and Possession are not attacks because, on average, an opponent playing those cards will not slow your deck down. Of course, there are edge cases. There's always edge cases. Here at f.ds, we live and breath edge cases. Yes, Masquerade can slow you down if it forces you to pass a good card. Yes, Possession can slow you down if there's also Ambassador or Masquerade or Island in your deck. Tribute, I'd argue, is completely neutral, over 1000 games you should get just as many times that Tribute helps as the number of times that it hurts.

And of course there's edge cases in the other direction. Thief and Pirate Ship could quite possibly make your deck faster instead of slower. They are really dependent upon your strategy. But I'm willing to bet that if you took 100 different random simulator buy rules, checked how long it takes them to get 5 Provinces, and then played them against any official attack card, on average it will take them more turns to get 5 Provinces.
Title: Re: Attack definition
Post by: Drab Emordnilap on September 26, 2012, 05:53:00 pm
I'm pretty sure that every attack in Dominion both

A) Is a mandatory effect
B) Affects all opponents "equally"

I put equally in quotes because of cases where there's not enough curses to go around, or I militia when one person already has a 3 card hand and someone else doesn't. Edge cases.

The only exception I can think of to these is Masquerade. Maybe you have to say an attack doesn't affect yourself?
Title: Re: Attack definition
Post by: Young Nick on September 26, 2012, 05:56:34 pm
What do you classify this card as?

$3
Benefactor
+$1
Each other player gains a Copper.

Is this an attack? Maybe so. I've always wondered.
Title: Re: Attack definition
Post by: Drab Emordnilap on September 26, 2012, 06:02:43 pm
What do you classify this card as?

$3
Benefactor
+$1
Each other player gains a Copper.

Is this an attack? Maybe so. I've always wondered.

I think so.
Title: Re: Attack definition
Post by: Archetype on September 26, 2012, 06:04:16 pm
What do you classify this card as?

$3
Benefactor
+$1
Each other player gains a Copper.

Is this an attack? Maybe so. I've always wondered.

I would think so. A same argument can be made with an Estate giver.

It junks your deck, but not as badly as a Curse. This card just isn't doesn't feel like an attack in games where Copper isn't a bad thing, but it still.

Same thing with Militia on a Tunnel board. Sometimes your helping them, other times, it's an Attack.
Title: Re: Attack definition
Post by: clb on September 26, 2012, 06:15:09 pm
What do you classify this card as?

$3
Benefactor
+$1
Each other player gains a Copper.

Is this an attack? Maybe so. I've always wondered.

I would think so. A same argument can be made with an Estate giver.

It junks your deck, but not as badly as a Curse. This card just isn't doesn't feel like an attack in games where Copper isn't a bad thing, but it still.

Same thing with Militia on a Tunnel board. Sometimes your helping them, other times, it's an Attack.

Ambassador is a perfect example of all of that. Some games, you are happy to have the copper or estates thrown at you. Not usually. Sometimes you can ignore the junk, but in general, you would rather they have not "gifted" you with that card.
Title: Re: Attack definition
Post by: Cuzz on September 26, 2012, 06:19:44 pm
What do you classify this card as?

$3
Benefactor
+$1
Each other player gains a Copper.

Is this an attack? Maybe so. I've always wondered.

I would think so. A same argument can be made with an Estate giver.

It junks your deck, but not as badly as a Curse. This card just isn't doesn't feel like an attack in games where Copper isn't a bad thing, but it still.

Same thing with Militia on a Tunnel board. Sometimes your helping them, other times, it's an Attack.

Ruins aren't always so bad either.
Title: Re: Attack definition
Post by: theory on September 26, 2012, 06:19:53 pm
What do you classify this card as?

$3
Benefactor
+$1
Each other player gains a Copper.

Is this an attack? Maybe so. I've always wondered.

I would think so. A same argument can be made with an Estate giver.

It junks your deck, but not as badly as a Curse. This card just isn't doesn't feel like an attack in games where Copper isn't a bad thing, but it still.

Same thing with Militia on a Tunnel board. Sometimes your helping them, other times, it's an Attack.

Ambassador is a perfect example of all of that. Some games, you are happy to have the copper or estates thrown at you. Not usually. Sometimes you can ignore the junk, but in general, you would rather they have not "gifted" you with that card.

Ambassador is different.  It's not an attack because it sends over coppers, it's an attack because the player can choose to send over coppers.
Title: Re: Attack definition
Post by: Insomniac on September 26, 2012, 06:22:57 pm
What do you classify this card as?

$3
Benefactor
+$1
Each other player gains a Copper.

Is this an attack? Maybe so. I've always wondered.

I would think so. A same argument can be made with an Estate giver.

It junks your deck, but not as badly as a Curse. This card just isn't doesn't feel like an attack in games where Copper isn't a bad thing, but it still.

Same thing with Militia on a Tunnel board. Sometimes your helping them, other times, it's an Attack.

Ruins aren't always so bad either.

I did once have a hand of KC, Ruined village, Goons, Goons, Goons...

Man was I ever happy KC+Ruined Village = a REAL village.
Title: Re: Attack definition
Post by: werothegreat on September 26, 2012, 06:52:52 pm
What do you classify this card as?

$3
Benefactor
+$1
Each other player gains a Copper.

Is this an attack? Maybe so. I've always wondered.

I think so.

It's an Attack if it's labeled as an Attack.  The only real reason to have the label "Attack" is so that you know what Moat is used for.
Title: Re: Attack definition
Post by: GendoIkari on September 26, 2012, 07:10:35 pm
What do you classify this card as?

$3
Benefactor
+$1
Each other player gains a Copper.

Is this an attack? Maybe so. I've always wondered.

I think so.

It's an Attack if it's labeled as an Attack.  The only real reason to have the label "Attack" is so that you know what Moat is used for.

Well yes, and I said this in the OP. But the question at hand was when should the label "Attack" be used on a card?
Title: Re: Attack definition
Post by: Titandrake on September 26, 2012, 07:18:42 pm
Originally, I would have said that Attacks are cards that both force your opponents to do something and apply that attack symmetrically. Except Governor and its gain Silver effect messes with that definition.

I think the real answer is that an Attack is a card that causes a symmetric effect you want to be preventable. You want to prevent Militia and Torturer effects (ignoring edge cases), but you don't want to prevent Governor (once again, ignoring decisions about whether you want Silver in a Governor mega-turn deck). Masquerade isn't preventable because if it was, Moat would ruin it. One player reveals Moat and doesn't have to pass a card, but gets a free card from someone else? That would be terrible. Possession is a combination of lack of symmetry and iffy reaction interactions. Moat stopping Possession would be somewhat annoying, and revealing a reaction like Horse Traders or Secret Chamber just makes the Possessed turn better.

Edit: I think the word to use here is "affects all players", not "symmetric effect". Consider the following card that should never be printed:

Silly Action
$2
The player to your left gains an Estate, the player 2 seats left gains a Curse, the player 3 seats left gains a Copper. This cannot cause you to gain any cards.

This should be an attack: you gain nothing, and each other player gains something bad. Sure, the things given are different, but it should be preventable. In comparison, an action that only affects 1 opponent should not be an attack. If it was, then unaffected players could get the benefits of Horse Traders and Secret Chamber at no cost.
Title: Re: Attack definition
Post by: SirPeebles on September 26, 2012, 09:20:34 pm
Originally, I would have said that Attacks are cards that both force your opponents to do something and apply that attack symmetrically. Except Governor and its gain Silver effect messes with that definition.

I think the real answer is that an Attack is a card that causes a symmetric effect you want to be preventable. You want to prevent Militia and Torturer effects (ignoring edge cases), but you don't want to prevent Governor (once again, ignoring decisions about whether you want Silver in a Governor mega-turn deck). Masquerade isn't preventable because if it was, Moat would ruin it. One player reveals Moat and doesn't have to pass a card, but gets a free card from someone else? That would be terrible. Possession is a combination of lack of symmetry and iffy reaction interactions. Moat stopping Possession would be somewhat annoying, and revealing a reaction like Horse Traders or Secret Chamber just makes the Possessed turn better.

Edit: I think the word to use here is "affects all players", not "symmetric effect". Consider the following card that should never be printed:

Silly Action
$2
The player to your left gains an Estate, the player 2 seats left gains a Curse, the player 3 seats left gains a Copper. This cannot cause you to gain any cards.

This should be an attack: you gain nothing, and each other player gains something bad. Sure, the things given are different, but it should be preventable. In comparison, an action that only affects 1 opponent should not be an attack. If it was, then unaffected players could get the benefits of Horse Traders and Secret Chamber at no cost.

Revealing a Secret Chamber usually wouldn't make the Possession better.  If I were to reveal a Secret Chamber, I would use it to push forward my best cards out of your reach, or at least break up any combos.
Title: Re: Attack definition
Post by: Drab Emordnilap on September 26, 2012, 10:19:55 pm
Originally, I would have said that Attacks are cards that both force your opponents to do something and apply that attack symmetrically. Except Governor and its gain Silver effect messes with that definition.

Also Council Room, et al.
Title: Re: Attack definition
Post by: Jimmmmm on September 27, 2012, 02:22:00 am
I think a card should be an attack if you usually buy it at least in part for the direct effect it has on your opponent/s when you play it.
Title: Re: Attack definition
Post by: jotheonah on September 27, 2012, 02:36:53 am
What do you classify this card as?

$3
Benefactor
+$1
Each other player gains a Copper.

Is this an attack? Maybe so. I've always wondered.

What if you replace Copper with Silver? With Duchy? Potion? What is a "bad card"?

I think an Attack should be any card that's named after a mean person. The only official exceptions I can think of are Ghost Ship, Pirate Ship, and Scrying Pool.
Title: Re: Attack definition
Post by: Young Nick on September 27, 2012, 03:14:26 am
And of course all of the Knights. We know they are mean because they get to play-test cards and don't leak spoilers to us, obviously.

tl;dr: Knights are obviscum.
Title: Re: Attack definition
Post by: Qvist on September 27, 2012, 04:12:34 am
Don't forget that IGG isn't an Attack either because you don't attack on-play, you attack on-gain. You can therefore not block it with Moat.
Govenor isn't an Attack either although giving out Silvers might be hurtful in edge cases.

I think a card should be an Attack if playing the card hurts your opponents in the majority of the games. Giving out Coppers does that, giving out Silvers not.
Giving out Curses and Estates is an Attack, although it would be interesting to see a card which gives out Duchies. I think this wouldn't be an Attack.
Title: Re: Attack definition
Post by: aaron0013 on September 27, 2012, 08:05:37 am
Governor should totally  be an attack card.....if your opponent is a newbie and forgets that the card he trashed must be replaced by one costing exactly one more!

Oh, ha. Sorry about that gold... :D
Title: Re: Attack definition
Post by: GendoIkari on September 27, 2012, 08:17:38 am
Don't forget that IGG isn't an Attack either because you don't attack on-play, you attack on-gain. You can therefore not block it with Moat.
Govenor isn't an Attack either although giving out Silvers might be hurtful in edge cases.

I think a card should be an Attack if playing the card hurts your opponents in the majority of the games. Giving out Coppers does that, giving out Silvers not.
Giving out Curses and Estates is an Attack, although it would be interesting to see a card which gives out Duchies. I think this wouldn't be an Attack.

Right, I ment to mention IGG but forgot. It's definitely an attack in terms just about any definition you can think of, it's just that to avoid Moat confusion reasons, Donald chose to not put that label there.
Title: Re: Attack definition
Post by: Octo on September 27, 2012, 08:39:07 am
Well, no, the Attack wording is not there because it's not an Attack. It's an aggressive card, but it's not an Attack - Attacks (as far as the game is concerned) must be played it seems. To lump everything into 'Attack' then nullifies the actual terminology. It would be clearer to use a different word, but even that is flawed. I think trying to label cards as one or other is inaccurate - it's the move within the context that's aggressive/attacking and that can come in lots of forms and with lots of cards. Sometimes Attacks are not used to attack the opponents, you're just running a pile down to end the game perhaps. As stated in the beginning, sometimes a cards helps, sometimes it hinders, it all depends, it and doesn't depend on the card, it depends on the context and the intent, so labelling things outside of that sphere will always be inaccurate to some extent.
Title: Re: Attack definition
Post by: werothegreat on September 27, 2012, 08:58:41 am
What do you classify this card as?

$3
Benefactor
+$1
Each other player gains a Copper.

Is this an attack? Maybe so. I've always wondered.

I think so.

It's an Attack if it's labeled as an Attack.  The only real reason to have the label "Attack" is so that you know what Moat is used for.

Well yes, and I said this in the OP. But the question at hand was when should the label "Attack" be used on a card?

Whenever you feel like it.  But it does have to fit certain criteria.

1) It makes your opponent(s) do something.
2) That something could be construed as undesirable to said opponent(s) in a majority of cases.
3) It makes sense for Moat to block it.
4) Unless it makes you trash or gain Curses, it should give some benefit to the player if it is blocked by Moat.
Title: Re: Attack definition
Post by: Brando Commando on September 27, 2012, 09:18:26 am
I'm not sure that coming up with a hardcord definition for Attack is possible or even super useful. For one, I'm having a hard time thinking of situations where an attack is straight-up bad for you in every circumstance. Witch seems like the most likely candidate (off the top of my head) since all it does is give you a curse.

(The only situation I can think of where this is a net advantage for the attacked player is where they use Watchtower to trash the curse in a situation where that ends the game on 3 piles to their advantage...which only makes sense if attacking player was relying on the that curse for the win.)

So I guess my question is, what do you gain by (re)defining "attack"?

Title: Re: Attack definition
Post by: rinkworks on September 27, 2012, 09:24:43 am
Whenever you feel like it.  But it does have to fit certain criteria.

So, not whenever you feel like it.

Quote
1) It makes your opponent(s) do something.
2) That something could be construed as undesirable to said opponent(s) in a majority of cases.
3) It makes sense for Moat to block it.
4) Unless it makes you trash or gain Curses, it should give some benefit to the player if it is blocked by Moat.

You lose me on #4.  Saboteur.  Often Rogue.  But even if there weren't any official counterexamples, why would you say this anyway?  Consider this:

Poor Militia
$3 - Action
Each other player discards down to 3 cards in hand.

Are you saying that if this card existed, it shouldn't have the Attack type?
Title: Re: Attack definition
Post by: rinkworks on September 27, 2012, 09:27:20 am
So I guess my question is, what do you gain by (re)defining "attack"?

This comes up all the time in the variants subforum.  People post fan cards that straddle the line between what should and shouldn't warrant the Attack type, and there is debate about it.

Even outside the context of fan cards, though, it's not unfruitful to think about the logic and order of the game.
Title: Re: Attack definition
Post by: werothegreat on September 27, 2012, 09:55:57 am
Whenever you feel like it.  But it does have to fit certain criteria.

So, not whenever you feel like it.

Quote
1) It makes your opponent(s) do something.
2) That something could be construed as undesirable to said opponent(s) in a majority of cases.
3) It makes sense for Moat to block it.
4) Unless it makes you trash or gain Curses, it should give some benefit to the player if it is blocked by Moat.

You lose me on #4.  Saboteur.  Often Rogue.  But even if there weren't any official counterexamples, why would you say this anyway?  Consider this:

Poor Militia
$3 - Action
Each other player discards down to 3 cards in hand.

Are you saying that if this card existed, it shouldn't have the Attack type?

Notice how it said "unless it makes you trash or gain a Curses".  I meant trash other cards, not trash Curses.
Title: Re: Attack definition
Post by: Brando Commando on September 27, 2012, 09:56:06 am
So I guess my question is, what do you gain by (re)defining "attack"?

This comes up all the time in the variants subforum.  People post fan cards that straddle the line between what should and shouldn't warrant the Attack type, and there is debate about it.

Even outside the context of fan cards, though, it's not unfruitful to think about the logic and order of the game.

I see your point, but what I'm saying is, If it plays well, does it matter if it's correctly labeled as an Attack or not?

My (theoretical) counterexamples are cards which perhaps should be labeled attacks but might be too weak against Reaction cards if they were:

Famine -- Cost 3
+1$
All players discard down to three cards.

or maybe

Pestilence -- Cost 4
+3$ (Compare to Militia.)
All players discard down to three cards.

I feel like either of these might be labeled "Attacks" -- probably justifiably -- but I also think they'd be too weak if other players could use Reactions and/or Lighthouse against them.
Title: Re: Attack definition
Post by: Drab Emordnilap on September 27, 2012, 11:08:10 am
I feel like those cards couldn't be printed without the Attack typing, for player sentiment reasons if nothing else. Some players are already offended that things like Possession and Tribute can't be blocked by Moat; this would take that to a new level of ire.
Title: Re: Attack definition
Post by: GendoIkari on September 27, 2012, 11:12:53 am
Whenever you feel like it.  But it does have to fit certain criteria.

So, not whenever you feel like it.

Quote
1) It makes your opponent(s) do something.
2) That something could be construed as undesirable to said opponent(s) in a majority of cases.
3) It makes sense for Moat to block it.
4) Unless it makes you trash or gain Curses, it should give some benefit to the player if it is blocked by Moat.

You lose me on #4.  Saboteur.  Often Rogue.  But even if there weren't any official counterexamples, why would you say this anyway?  Consider this:

Poor Militia
$3 - Action
Each other player discards down to 3 cards in hand.

Are you saying that if this card existed, it shouldn't have the Attack type?

Notice how it said "unless it makes you trash or gain a Curses".  I meant trash other cards, not trash Curses.

True, your criteria of excluding trashing attacks does allow for Saboteur, but I still don't see any reason to treat those as special, just because the 4 attacks in the game that give you no benefit if Moated happen to be 3 trashers and 1 curser (Thief, Sir Michael, Saboteur, Sea Hag). Yes that's how the official cards are, but there's no reason to think that those 2 categories are that special. Those cards don't give the player a benefit because they are considered too powerful of an attack to do so. But if "Poor Militia" were a card, it would certainly be an attack as well.

Also, given your criteria, a card that reveals your opponent's hand would still be possibly an attack, but maybe not. It forces your opponent to do something. Something that your opponent would rather not do. But unless he's about to play an Envoy or Contraband, it doesn't actually slow down or hurt his playing.
Title: Re: Attack definition
Post by: Qvist on September 27, 2012, 11:18:53 am
How about: "Every other player with at least 5 cards in hand discards his hand and draws 5 cards."? Would that be an Attack? I would say No because in the majority of the cases it wouldn't hurt. But in combination with Governor or Council Room, this would feel like an Attack.
And if you think this is an Attack, what about: "Every other player discards his hand and draws the same amount of cards he had before."? This would still hurt against Scheme shenigans or some mind play in combination with Ghost Ship but are edge scenarios. Would that be an Attack?
Title: Re: Attack definition
Post by: Brando Commando on September 27, 2012, 11:26:44 am
I feel like those cards couldn't be printed without the Attack typing, for player sentiment reasons if nothing else. Some players are already offended that things like Possession and Tribute can't be blocked by Moat; this would take that to a new level of ire.

Quite possibly true, but I don't know that player ire is a good reason to pursue a robust definition of "Attack."

And, actually, I'm not so sure about the ire for my "self-Attack" cards; they would probably only come out in games where they combo well with something (like Famine + Tunnel or Famine + Fishing Village + Watchtower, etc.) in which case, often the other players' Attacks might be better for you -- e.g., I drew a FV and a Watchtower, but no Famine, so I don't get to discard and redraw. But now my opponent has played a Famine, so I'm all set.

In sum, I'm still not convinced that we need to define Attack so much as we need to talk about which cards get labeled as Attack and which don't.
Title: Re: Attack definition
Post by: jotheonah on September 27, 2012, 01:25:35 pm
I'm not sure that coming up with a hardcord definition for Attack is possible or even super useful. For one, I'm having a hard time thinking of situations where an attack is straight-up bad for you in every circumstance. Witch seems like the most likely candidate (off the top of my head) since all it does is give you a curse.

(The only situation I can think of where this is a net advantage for the attacked player is where they use Watchtower to trash the curse in a situation where that ends the game on 3 piles to their advantage...which only makes sense if attacking player was relying on the that curse for the win.)

So I guess my question is, what do you gain by (re)defining "attack"?

I have all 12 Fairgrounds, 19 unique cards, and no curses. Your Witch just gave me 23 VP.
Title: Re: Attack definition
Post by: eHalcyon on September 27, 2012, 01:39:57 pm
I feel like those cards couldn't be printed without the Attack typing, for player sentiment reasons if nothing else. Some players are already offended that things like Possession and Tribute can't be blocked by Moat; this would take that to a new level of ire.

Quite possibly true, but I don't know that player ire is a good reason to pursue a robust definition of "Attack."

And, actually, I'm not so sure about the ire for my "self-Attack" cards; they would probably only come out in games where they combo well with something (like Famine + Tunnel or Famine + Fishing Village + Watchtower, etc.) in which case, often the other players' Attacks might be better for you -- e.g., I drew a FV and a Watchtower, but no Famine, so I don't get to discard and redraw. But now my opponent has played a Famine, so I'm all set.

In sum, I'm still not convinced that we need to define Attack so much as we need to talk about which cards get labeled as Attack and which don't.

I don't think "self Attack" is a thing.  Even if an action is generally negative for you, it is still your choice to take it, therefore it is not a self attack.  Consider Minion -- the "attack" portion forces everyone with more than 4 cards (including you) to discard their hand and draw 4 new cards.  Since you had the choice and your opponents did not, it will generally be more harmful to them than it is to you.  If playing the card forces something bad on an opponent, it is an attack regardless of what penalty there is for playing it.

Sometimes attacks can help opponents, but these are edge cases.

Defining "Attack" is a good way of discussing which cards get labelled as Attacks and which ones don't.



IMO, Attacks force an effect on opponents that is bad for them on average.  Hand-size reduction, forced trashing, gaining junk, discarding good cards/leaving junk on deck (Rabble, Fortune Teller).  And given the precedence of IGG, the effect must be caused on play, not on buy/gain.

There are some cards that walk the line a bit.  Governor is probably the best example, as it can force opponents to take Silver.  The question is whether Silver is junk.  It usually isn't, though it may be in certain decks.

Masquerade is hard to talk about simply because the times when you are forced to pass a good card will stick in your mind.  But it really doesn't do anything in general -- often everyone just passes an Estate or Copper.

Possession definitely doesn't hurt the target player, on average.
Title: Re: Attack definition
Post by: Captain_Frisk on September 27, 2012, 01:48:56 pm
So, i think that the reason why governor is not an attack is because most of the time, on most of the choices, the other player would not want it auto blocked by lighthouse.  The silver gaining is one of those things were I'd prefer a "may", but that has been covered elsewhere, and I'm never going to win that battle.

Masquerade probably should be an attack, but what would happen if it was  blocked?  You'd have to put a bunch of special language in there, and we know how DXV feels about excess language.

Every "proper" attack has the "messes with the opponent" function different than the "does good things for me" function.  Sea Hag and Sab are special cases that have no "do good things for me" function.
Title: Re: Attack definition
Post by: GendoIkari on September 27, 2012, 02:13:30 pm
I'm not sure that coming up with a hardcord definition for Attack is possible or even super useful. For one, I'm having a hard time thinking of situations where an attack is straight-up bad for you in every circumstance. Witch seems like the most likely candidate (off the top of my head) since all it does is give you a curse.

(The only situation I can think of where this is a net advantage for the attacked player is where they use Watchtower to trash the curse in a situation where that ends the game on 3 piles to their advantage...which only makes sense if attacking player was relying on the that curse for the win.)

So I guess my question is, what do you gain by (re)defining "attack"?

I have all 12 Fairgrounds, 19 unique cards, and no curses. Your Witch just gave me 23 VP.

You forgot the 12 points from your 12 Gardens going from 39 to 40 cards! Oh, and it's 36 points total, not 35, because I revealed a Trader to the Witch, and hadn't had any Silver in my deck before that.
Title: Re: Attack definition
Post by: jotheonah on September 27, 2012, 02:21:23 pm
I think if Dominion ever needs a new advertising slogan it should be "Dominion: There's an edge-case for that."
Title: Re: Attack definition
Post by: ftl on September 27, 2012, 02:23:36 pm
I'm not sure that coming up with a hardcord definition for Attack is possible or even super useful. For one, I'm having a hard time thinking of situations where an attack is straight-up bad for you in every circumstance. Witch seems like the most likely candidate (off the top of my head) since all it does is give you a curse.

(The only situation I can think of where this is a net advantage for the attacked player is where they use Watchtower to trash the curse in a situation where that ends the game on 3 piles to their advantage...which only makes sense if attacking player was relying on the that curse for the win.)

So I guess my question is, what do you gain by (re)defining "attack"?

I have all 12 Fairgrounds, 19 unique cards, and no curses. Your Witch just gave me 23 VP.

You forgot the 12 points from your 12 Gardens going from 39 to 40 cards! Oh, and it's 36 points total, not 35, because I revealed a Trader to the Witch, and hadn't had any Silver in my deck before that.

If you're gaining a silver, you might as well have it power up your 12 feodums by one level, for 12 more points.
Title: Re: Attack definition
Post by: GendoIkari on September 27, 2012, 02:26:24 pm
I'm not sure that coming up with a hardcord definition for Attack is possible or even super useful. For one, I'm having a hard time thinking of situations where an attack is straight-up bad for you in every circumstance. Witch seems like the most likely candidate (off the top of my head) since all it does is give you a curse.

(The only situation I can think of where this is a net advantage for the attacked player is where they use Watchtower to trash the curse in a situation where that ends the game on 3 piles to their advantage...which only makes sense if attacking player was relying on the that curse for the win.)

So I guess my question is, what do you gain by (re)defining "attack"?

I have all 12 Fairgrounds, 19 unique cards, and no curses. Your Witch just gave me 23 VP.

You forgot the 12 points from your 12 Gardens going from 39 to 40 cards! Oh, and it's 36 points total, not 35, because I revealed a Trader to the Witch, and hadn't had any Silver in my deck before that.

If you're gaining a silver, you might as well have it power up your 12 feodums by one level, for 12 more points.

Actually doesn't quite work... Feodums need 3 Silvers per point, so to power them up you would have had to already have 2 Silvers in your deck, meaning that it doesn't help Fairgrounds. Of course, KC-Witch could do it.
Title: Re: Attack definition
Post by: ftl on September 27, 2012, 02:28:20 pm
Aww, you're right. Damn.
Title: Re: Attack definition
Post by: Octo on September 27, 2012, 02:38:00 pm
So, what happens when I play Minion for the +$2? Am I attacking? Sure, you don't want to Moat it (there's no need), but I still converted my urchin into a Mercenary so its status as an Attack is still relevant, even though it doesn't fall into any of the above criteria.
Title: Re: Attack definition
Post by: GendoIkari on September 27, 2012, 02:39:58 pm
So, what happens when I play Minion for the +$2? Am I attacking? Sure, you don't want to Moat it (there's no need), but I still converted my urchin into a Mercenary so its status as an Attack is still relevant, even though it doesn't fall into any of the above criteria.

Same with Pirate ship. And Rogue, though with Rogue it's not a choice. Those cards say attack on them because they have the ability to be used in such a way that meets whatever criteria we are choosing for "attack" cards.
Title: Re: Attack definition
Post by: Donald X. on September 27, 2012, 03:27:21 pm
I feel like I've covered this one previously, but here goes.

The point to "attack" was to let Moat refer to it. This is an old trick. If the original game didn't have "attack" and at some point I wanted to add Moat, it would be messy. You can cobble something together but it's not great. Probably I would just do it over multiple cards instead - Watchtower for Cursing although it also covers discard although not by letting you not discard, Market Square for trashing, after-the-fact stuff like Mystic for Spy. But if I stick on this label "attack," suddenly I can just say, bam, stop attacks. And I wanted Moat. So I did that.

It has to be timed for when they play the card, before it does anything, because man, who knows what complex things that card will do. This means that cards that have attacking as an option, e.g. Pirate Ship, end up letting you Moat them before they decide, and they may not even end up doing anything to you. Which is different from if I say Watchtower a Curse you're giving me. But that seemed like something I could live with.

It has to be clear what happens when you Moat. This ruled out putting "attack" on Masquerade.

And finally some things can hurt another player but don't have much spirit of attacking to them, like Council Room. I always say, Moat doesn't have a responsibility to stop every bad thing possible from happening to you. It doesn't stop them from buying Province, that's a big one. So Council Room is not an attack. It would look weird, and would slow down the game, as you thought, hmm is there some reason not to draw the card this time. Tribute meanwhile refers to the top cards for information; it has to get that information to be playable, we can't have someone Moating it. It might feel like Moating a Spy to you, but to the guy with Tribute it's like you Moated his Gold.
Title: Re: Attack definition
Post by: shMerker on September 27, 2012, 07:18:15 pm
This sort of reminds me of discussions about why we can have a victory card worth 0 points (Overgrown Estate) or a treasure card that gives 0 coins (Horn of Plenty). It's because these are shorthands for behavior. Making Horn of Plenty a treasure instead of an action means you can play as many as you want even if you have no actions left and that you can play them after you've played your treasure cards by default. Would you want all of that information on an action card?

Everyone keeps talking about attacks that may benefit opponents but I don't see why an attack has to do anything at all. How about this?

Declaration of War - $2 Action Attack

+1 action
+$2

Each other player does nothing

You might buy that on a $2 hand because aside from the Attack label and some other random gotchas like Mine it's pretty much interchangeable with silver. Except shoot now you're creating opportunities for other players to activate their Secret Chambers and Horse Traders even though nothing will happen to them even if they don't. On some boards it's like a whole different card just because it says "attack" on it.
Title: Re: Attack definition
Post by: Davio on September 28, 2012, 04:04:09 am
So anything that messes with your opponent's deck isn't necessarily an Attack.

To me, an attack is something that, when it succeeds without being reacted to (Horse Traders on Minion etc.), hurts your opponent in some way most of the time. "Hurts" is relative though, Silver presents can hurt. And even attacks that succeed can be helpful for your opponent with cards like Menagerie. There is no 100% in Dominion.

One crucial part about attacks is that because they hurt the opponent a lot of the time, their other bonus is diminished. Also, even if they have a similar bonus, they are more expensive. Sea Hag and Saboteur even have no other bonus. Margrave's bonus is less than Council Room's. Witch draws 2 cards to Smithy's 3 and is even more expensive!
Title: Re: Attack definition
Post by: GendoIkari on September 28, 2012, 07:48:08 am
So anything that messes with your opponent's deck isn't necessarily an Attack.

To me, an attack is something that, when it succeeds without being reacted to (Horse Traders on Minion etc.), hurts your opponent in some way most of the time. "Hurts" is relative though, Silver presents can hurt. And even attacks that succeed can be helpful for your opponent with cards like Menagerie. There is no 100% in Dominion.

This is basically what I was saying, and I think we can quantify "hurt" instead of keeping it subjective/relative, by talking about whether or not it causes the opponent's deck to take longer on average to get points.
Title: Re: Attack definition
Post by: Captain_Frisk on September 28, 2012, 09:47:07 am
There is no 100% in Dominion.

That depends on the kingdom.
Title: Re: Attack definition
Post by: One Armed Man on September 28, 2012, 10:35:08 am
What about an Attack card that waits until a certain phase to deal its damage. Like an attack card that waits until the end of the action phase.

Assassin
5
Action - Attack
+4 Cards. Discard 2 cards or 1 copper.         
At the end of your Action phase, each other player with 5 or more cards in hand reveals a number of cards equal to the number of Assassin cards you have in play. Each player discards one revealed card of your choice.

How would a Moat react to this? What if the player draws a Moat between the first and second time this is played in a turn?
Title: Re: Attack definition
Post by: DStu on September 28, 2012, 10:55:41 am
I would say Moat has to be revealed when the Attack is played. So if you draw it between two plays of Assassin, you can only reveal it for the second, and so you are only saved from the second.
Title: Re: Attack definition
Post by: One Armed Man on September 28, 2012, 11:18:18 am
I would say Moat has to be revealed when the Attack is played. So if you draw it between two plays of Assassin, you can only reveal it for the second, and so you are only saved from the second.
There isn't a second "attack". The "attack" happens as a result of any or all of the Assassins. If that happens, do the other players reveal 1 card or 2? I don't care which, I just want rules that work. Is there a wording of this card or rules clarification that makes it work?
Title: Re: Attack definition
Post by: DStu on September 28, 2012, 11:27:17 am
I would say Moat has to be revealed when the Attack is played. So if you draw it between two plays of Assassin, you can only reveal it for the second, and so you are only saved from the second.
There isn't a second "attack". The "attack" happens as a result of any or all of the Assassins. If that happens, do the other players reveal 1 card or 2? I don't care which, I just want rules that work. Is there a wording of this card or rules clarification that makes it work?

It doesn't matter when the "attack" 'happens', Moat reacts and has to be revealed on when an "Attack card" is 'played', and these are the Assassins, and they are played at two different times, where at the first you haven't revealed, and for the second you have. So there's nothing unclear.

Anyway, I should read the text better to give more clear answers. Because, of course this kind of does not really matter. Because what does "so you are only saved from the second" mean?

Assasin says "...reveals a number of cards equal to the number of Assassin cards you have in play."  It doesn't matter how many Assassins effect you (as long as it's at least one). The first Assassin effects you. So you reveal 2 (because that's the number of Assassins you have in play) cards and resolve the rest.
The second Assassin doesn't effect you, so you don't care what it would do.

For comparision, if you don't reveal any Moats at all:
The first Assassin effects you, so you reveal 2 cards and resolve the rest.
The second Assassin effects you, but you probably only have 4 cards in hand, so you can ignore the rest.

Would you have more than 5 cards in hand, the second one would also have been resolved.
Title: Re: Attack definition
Post by: dondon151 on September 28, 2012, 12:28:59 pm
I don't think that Moat will do anything. There is nothing to defend against when the card is played.
Title: Re: Attack definition
Post by: Octo on September 28, 2012, 12:50:33 pm
I think this kind of discussion is really...odd. DXV frequently talks about how he changes/tweaks rules that are causing confusion - he just said the term Attack itself was created to stop precisely such messiness. If the current state of dominion is anything to go by, a card causing this much confusion right out of the gate (as opposed to the when combined in some exotic, unusual scenario) would not make the cut so having a hypothetical discussion about it is just totally pointless.

How would this hypothetical situation pan out? It wouldn't. That's the answer. (sorry if I'm being a curmudgeon about it)
Title: Re: Attack definition
Post by: AJD on September 28, 2012, 12:54:22 pm
What about an Attack card that waits until a certain phase to deal its damage. Like an attack card that waits until the end of the action phase.

Assassin
5
Action - Attack
+4 Cards. Discard 2 cards or 1 copper.         
At the end of your Action phase, each other player with 5 or more cards in hand reveals a number of cards equal to the number of Assassin cards you have in play. Each player discards one revealed card of your choice.

How would a Moat react to this? What if the player draws a Moat between the first and second time this is played in a turn?

Follow instructions in order, do all you can! So here's a potential Action phase:

I play Village and get +1 card, +2 actions.
I play Assassin and get +4 cards and discard a copper.
I play Governor and get +1 action, +3 cards. You get +1 card.
I play Assassin. You reveal Moat; I get +4 cards and discard a copper.
End of Action phase. The two Assassins activate simultaneously, so I choose the order.
First Assassin activates. You have 6 cards in hand and two Assassins are in play, so you reveal 2 cards and discard one.
Second Assassin activates. You have 5 cards in hand, but you Moated this Assassin, so it doesn't affect you.
I proceed to the Buy phase.
Title: Re: Attack definition
Post by: shMerker on September 28, 2012, 01:16:18 pm
I don't think that Moat will do anything. There is nothing to defend against when the card is played.

So? Moat doesn't specify anything about the timing of what it stops. It says "you are unaffected by the attack". If the effect takes a while to resolve, so what? You still revealed the moat for that attack . If the same attack could hit you after that then you aren't following the instructions on Moat.

But I agree that a slow attack would be confusing. It can be easy to forget something like revealing a moat. Maybe after playing an assassin you also played a vault and the player who revealed moat decided to discard it because he also has two smithies and no village. The moat is still going to block the effect of the attack but now there's no easy way to remember that it was revealed.
Title: Re: Attack definition
Post by: PSGarak on October 02, 2012, 11:35:42 am
... It doesn't stop them from buying Province, that's a big one. ...
Now that would be an interesting reaction card.

Pre-emptive Saboteur - $5
Action/Reaction
[something action-y]
You may reveal this card when your opponent gains a card. If you do, your opponent trashes that card, and may gain one costing up to $2 less than the trashed card.