Dominion Strategy Forum
Dominion => Dominion General Discussion => Topic started by: Lhurgoyf on October 27, 2011, 07:14:32 pm
-
Ok, this question seems a little bit weird, because most of the time, you don't want any additional Copper in your deck. I am also not thinking about Counting House and Coppersmith and such, my concern are generally 2 $5 Cards:
Ghostship and Witch.
Because if you buy one of these / nothing, your Deck will consist of only 11 Cards, which means, that if you draw your Attack on the 4th round there is only 1 Card left in your pile and playing the Attack will make you lose a whole shuffle through your deck. I don't know, how bad that is and if it remotely justifies buying that Copper.
Maybe the fact that the additional Copper increases the risk of the Witch being drawn just in turn 5, will make all the considerations obsolete.
I started this thread because sometimes being the only one with 5/2 start and a witch feels less powerful than I think it should, or it could be because they could upgrade their decks with Silver faster.
-
I've thought about this for the specific case of witch before and I would say it's almost definitely never worth it.
Think of it this way - opening witch/nothing gives you a 6/11 chance of missing the reshuffle. If you do miss it, that's one less curse you can give out. Opening witch/copper gives you a 1/6 chance of missing the reshuffle. That's much better. But is that extra probability of giving out one curse to clog your opponents deck worth the one copper you bought which clogs your own? I'd say no.
-
Witch/nothing even has the better best-case scenario. You can draw your witch on turn 3 AND 4 with probability 5/11*1/7 ~= 6.5% of the time. And missing a mid-turn reshuffle in the early game (during turn 4) is not that bad. Not all shuffles are created equal. You will end up shuffling again right after the next turn anyway.
-
Simulator says no. Better off not buying the copper. Still holds if you use a vault.
-
Witch/nothing even has the better best-case scenario. You can draw your witch on turn 3 AND 4 with probability 5/11*1/7 ~= 6.5% of the time. And missing a mid-turn reshuffle in the early game (during turn 4) is not that bad. Not all shuffles are created equal. You will end up shuffling again right after the next turn anyway.
This. The reshuffle it misses is not a full reshuffle; it is a half-reshuffle, because there would usually be only 6 cards in your discard pile (the 5 cards you drew on Turn 3, plus the card you bought) at the time you play Witch/Ghost Ship. Also, you pretty much guarantee that the card you bought on Turn 3 appears on Turn 4 or 5, and then everything gets reshuffled afresh for Turn 6.
-
Simulator says no. Better off not buying the copper. Still holds if you use a vault.
Did you simulate with trashing involved? I don't think you'd even consider this if there's no trashing.
If the setup is Witch, Moneylender, and no other trashing, then in that case it would definitely be worth it I'd expect.
-
If the setup is Witch, Moneylender, and no other trashing, you really don't want a Moneylender. The other player is going to be cursing you to death with curses you can't trash while you screw around with a terminal silver that trashes one copper at a time.
I'm talking 6% win rate here.
-
I don't see why not.
I wish I could solitaire that to see if you're right.
-
Why solitaire that? The simulator (http://users.telenet.be/jeroenaga/dominion.jnlp)can handle it perfectly. Here's the Witch/Moneylender bot which will get crushed by the built in Witch bot:
<player name="Witch/Moneylender" author="Geronimoo" description="Witch is probably the strongest attack card in the game, but combining it with Moneylender is just awful :)">
<type name="BigMoney"/>
<type name="Competitive"/>
<type name="UserCreated"/>
<type name="TwoPlayer"/>
<type name="Province"/>
<type name="SingleCard"/>
<type name="Attacking"/>
<type name="Bot"/>
<buy name="Province">
<condition>
<left type="countCardsInDeck" attribute="Gold"/>
<operator type="greaterThan" />
<right type="constant" attribute="0.0"/>
</condition>
</buy>
<buy name="Duchy">
<condition>
<left type="countCardsInSupply" attribute="Province"/>
<operator type="smallerOrEqualThan" />
<right type="constant" attribute="5.0"/>
</condition>
</buy>
<buy name="Estate">
<condition>
<left type="countCardsInSupply" attribute="Province"/>
<operator type="smallerOrEqualThan" />
<right type="constant" attribute="2.0"/>
</condition>
</buy>
<buy name="Witch">
<condition>
<left type="countCardsInDeck" attribute="Witch"/>
<operator type="equalTo" />
<right type="constant" attribute="0.0"/>
</condition>
</buy>
<buy name="Gold"/>
<buy name="Witch">
<condition>
<left type="countCardsInDeck" attribute="Witch"/>
<operator type="smallerThan" />
<right type="constant" attribute="2.0"/>
</condition>
</buy>
<buy name="Moneylender">
<condition>
<left type="countCardsInDeck" attribute="Moneylender"/>
<operator type="equalTo" />
<right type="constant" attribute="0.0"/>
</condition>
</buy>
<buy name="Silver"/>
</player>
-
Either I was doing something wrong before (maybe I was leaving chapels in) or I'm doing something wrong now. It wins about 40% of the time versus the version without the moneylender, not 6%. You're still better off without the moneylender.
Anyway, don't just take my word for it, here's a link (http://rspeer.github.com/dominiate/play.html#DoubleWitch/MoneylenderWitch) to the simulation.
(Yes, I should make it so people without a GitHub account can do that. Working on it.)
-
The bot is buying Moneylender ASAP. I think you'd want to buy it a little later than that, once the odds of drawing double terminal drops more.
But, you're probably right.
How do you give it logic for playing the cards?
-
Gardens or some theoretical future victory card worth 1 per X treasures in deck. That's all I can think of for getting the copper.
Most of the time I don't even want a 2 if it gives me worse odds of drawing the great 5 I just bought.
-
You also often really don't want a cost-2 if your cost-5 is terminal and draws cards. Drawing your cost-2 at that point will make you very sad, as it's way worse than drawing a Copper.
-
There are a couple situations where you want copper; actually copper is somewhat my specialty - check out my effect with and effect without on it, it's sick. Actually it's really sick. Anyway, here's basically when you'd want to open copper:
When you're going to buy a lot of green cards very fast - this includes some (though actually not THAT many) gardens decks, and, quite importantly, most duchy/duke decks, where copper is a very good card. It can be good for some decks where you're opening potion and trying to make sure you get up to say 3p - say you get 5/2 and familiar's on board without any other particularly good alternative (like another curser). Finally, there are some copper-driven strategies. Possibly coppersmith, moreover if you want a counting house or apothecary deck you may well want the copper even from the beginning. But honestly, all these situations together come up so infrequently... I mean, it's maybe 1% of games, and then you have to get 5/2 start, AND not have a good 2 to buy.
-
There are a couple situations where you want copper; actually copper is somewhat my specialty - check out my effect with and effect without on it, it's sick. Actually it's really sick.
I seem to recall you saying once that you often buy coppers at the end of games you win just to mess with the CR stats. ;)
-
There are a couple situations where you want copper; actually copper is somewhat my specialty - check out my effect with and effect without on it, it's sick. Actually it's really sick.
I seem to recall you saying once that you often buy coppers at the end of games you win just to mess with the CR stats. ;)
Yup. If I can, I do. But on the other hand, that probably accounts for less than half the games I buy copper. So it shouldn't be so sick of an effect as it looks like, but I still think copper is a card I have a good handle on.
There's a reason I do it with copper instead of say curse, you know, and it's because I already like copper and think I've got a good handle on it.
-
I have opened ironworks/copper with a 5/2 in a gardens game. I doubt there's a strong difference between it and ironworks/nothing, though.
-
http://dominion.isotropic.org/gamelog/201111/03/game-20111103-214612-489a907d.html
5/2 coppersmith is legit?
-
http://dominion.isotropic.org/gamelog/201111/03/game-20111103-214612-489a907d.html
5/2 coppersmith is legit?
You spelt Counting House wrong.
-
Lol.
Yes I did.
Sorry.