Print

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

[Geronimoo]

In this topic I made a rather bold statement (it was in Bold anyway):

If your deck will always improve with a Silver buy, you should never open Mining Village/Silver over Silver/Silver

Here's the challenge: prove I'm an idiot! (post your bot(s) as proof)

[DStu]

I don't have fully understood how you check for "if it always improves your deck to buy silver", but anyway:

It's not optimized, but the best candidate I think is BM-Goons. Don't know how you optimal play it, but

Code: [Select]

<player name="BM - Goons/MV">
   <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="Goons">
      <condition>
         <left type="countCardsInDeck" attribute="Goons"/>
         <operator type="smallerThan" />
         <right type="constant" attribute="2.0"/>
      </condition>
   </buy>
   <buy name="Gold"/>
   <buy name="Mining_Village">
      <condition>
         <left type="countCardsInSupply" attribute="Mining_Village"/>
         <operator type="equalTo" />
         <right type="constant" attribute="10.0"/>
      </condition>
   </buy>
   <buy name="Silver"/>
</player>

vs. "delete the MV"

The point is that a being first to Goons is really really strong, because BM has problems coming to $6 under attack of Goons. And so to Goons or Gold itself.

edit: 45:52 starting 4/3

[Geronimoo]

So apparently I am an idiot Hadn't really considered Goons.

So an early Goons is one reason you'd want to buy MV over Silver in a Big Money type deck. Are there others?

It's hard to find a good definition of a Big Money strategy, but the "more Silver=better" seemed fitting

[WanderingWinder]

More silver=better is actually true of all bot the most engine-y combo-y of decks, especially without colonies in play. Also, without knowing your play rules on mining village, this is going to be somewhat difficult. When do you have it trash?
Finally, any set where you really need a village is going to like mining village better. Torturer decks come to mind, and I doubt you really want to not buy silver when you could in those decks.
As for your overall point, I'm pretty sure you're wrong, though I don't know that you aren't right (you probably are, depending somewhat on the 4s, 5s, and 6s on board) for specifically opening silver/silver vs silver/MV

[Captain_Frisk]

Too lazy to simulate, but I suspect that any power 5+ card (mountebank, witch, goons, torturer, trading post, wharf) might warrant mining village / silver over silver / silver, as being stuck @ double 4's is a killer.  In the torturer case, you'll want the mining villages anyway.

[WanderingWinder]

Here's a bot with nothing other than money and MV beats any straight money deck I've seen:

Code: [Select]

<player name="Mining Village Simple">
   <buy name="Province">
      <condition>
         <left type="getTotalMoney"/>
         <operator type="greaterThan" />
         <right type="constant" attribute="18.0"/>
      </condition>
   </buy>
   <buy name="Duchy">
      <condition>
         <left type="countCardsInSupply" attribute="Province"/>
         <operator type="smallerOrEqualThan" />
         <right type="constant" attribute="4.0"/>
      </condition>
   </buy>
   <buy name="Estate">
      <condition>
         <left type="countCardsInSupply" attribute="Province"/>
         <operator type="smallerOrEqualThan" />
         <right type="constant" attribute="2.0"/>
      </condition>
   </buy>
   <buy name="Gold"/>
   <buy name="Duchy">
      <condition>
         <left type="countCardsInSupply" attribute="Province"/>
         <operator type="smallerOrEqualThan" />
         <right type="constant" attribute="6.0"/>
      </condition>
   </buy>
   <buy name="Mining_Village">
      <condition>
         <left type="countCardsInDeck" attribute="Silver"/>
         <operator type="greaterThan" />
         <right type="constant" attribute="3.0"/>
      </condition>
   </buy>
   <buy name="Silver"/>
</player>

This seems to defeat your first point entirely. The point is that you just need to wait a while before getting MVs

[Thisisnotasmile]

Too lazy to simulate, but I suspect that any power 5+ card (mountebank, witch, goons, torturer, trading post, wharf) might warrant mining village / silver over silver / silver, as being stuck @ double 4's is a killer.  In the torturer case, you'll want the mining villages anyway.

In the torturer case you'll want the Mining Villages anyway, but you won't want to be trashing them.

To be honest, I've never really got the whole "Open MV and use it as a 1-shot Grand Market" thing. I understand how it works. I just don't get it. If you don't trash it the first time you draw it, it was a wasted buy (unless you want it for the Village effect in which case you don't open with it). If you draw an Estate with it (and there's a not negligible chance of that happening), it was a wasted buy. If you DO draw a Copper with it and trash it for the $2, you have to spend every $ you have available to you that turn otherwise it was a wasted buy (Silver would have worked fine). And if you do trash it, was losing the chance to have an extra Silver in your deck for the rest of the game REALLY worth it for that card you just bought that cost $1 more than the card you would have been able to buy otherwise?

In my mind it's just not worth it in most situations. If you don't want the Silver in your deck (see: Minion), then fine. Otherwise I just don't get it.

[Davio]

I don't think Geronimoo's an idiot.

But I may have to cool my own fiery obsession with MV.
I always thought that if I drew a Copper with it and trashed the MV, it was as good as a played Gold and it was worth passing up on the Silver to get an early Gold. I guess I am mistaken and the list of "crucial" cards is small indeed.

I also thought that the absence of that extra card was beneficial for attacks that you want to play as much as possible (like the Cursers), but I guess the advantage of 1 or 2 extra plays is not enough to come from behind in the money race.

@WanderingWilder, Geronimoo is looking for an excuse to open MV/Silver over Silver/Silver, your bot doesn't open MV.

[rod-]

Here's a bot with nothing other than money and MV beats any straight money deck I've seen:

This seems to defeat your first point entirely. The point is that you just need to wait a while before getting MVs

The point could also be that even in a money deck, you don't want to buy more than 3 silvers if you can possibly avoid it.

[WanderingWinder]

Here's a bot with nothing other than money and MV beats any straight money deck I've seen:

This seems to defeat your first point entirely. The point is that you just need to wait a while before getting MVs

The point could also be that even in a money deck, you don't want to buy more than 3 silvers if you can possibly avoid it.

Except that this is extremely obviously false.

@Davio, seems your last point is right. I was thinking that he was making two claims, the one he does here, and one where he was saying that you basically should never prefer it to silver in a deck where you want more silver.

[DStu]

Putting the numerics here:
I simulated the following bot:

Code: [Select]

<player name="BM - Big Money Ultimate/MV">
   <buy name="Gold">
      <condition>
         <left type="countCardsInPlay" attribute="Mining_Village"/>
         <operator type="greaterThan" />
         <right type="constant" attribute="0.0"/>
      </condition>
   </buy>
   <buy name="Province"/>
   <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="Gold"/>
   <buy name="Mining_Village">
      <condition>
         <left type="countCardsInSupply" attribute="Mining_Village"/>
         <operator type="equalTo" />
         <right type="constant" attribute="10.0"/>
      </condition>
   </buy>
   <buy name="Silver"/>
</player>
and parsed for "how many $ would you have had the turn you trashed your MV if you started Silver/Silver" with the following little script (as you might see, I'm not an expert and just hacked straight forward with the 2 commands I googled)

Code: [Select]

rm output.dat
for i in /tmp/SimulatorSampleGame*.html; do
sed -e 's/<BR>/\n/g' -e 's/&nbsp;//g' $i | grep -v -e '..[*]~[trashes]' | grep -B6 trashes | sed -e 's/<FONT style=\"BACKGROUND-COLOR: #F3F584\">//g' -e 's/<FONT style=\"BACKGROUND-COLOR: #8EBF75\">//g' -e 's/<FONT style=\"BACKGROUND-COLOR: #D9D9D9 \">//g' -e 's$</FONT>$$g' | grep -e 'Hand' | grep -v -e '\.\.\.' | sed -e 's/Plr [12]//g' -e 's/Silver/\n1\n1\n/g' -e 's/Copper/\n1\n/g' -e 's/Mining Village/\n1\n1\n/g' -e 's/Gold/\n1\n1\n1\n/g' | grep -c -e '1' >> output.dat
done
grep -c -e 0 output.dat
grep -c -e 1 output.dat
grep -c -e 2 output.dat
grep -c -e 3 output.dat
grep -c -e 4 output.dat
grep -c -e 5 output.dat
grep -c -e 6 output.dat
grep -c -e 7 output.dat
grep -c -e 8 output.dat
Hammering a little bit on the "SampleGame" Button yields then the output

Code: [Select]

0: 8
1: 0
2: 0
3: 1
4: 2
5: 15
6: 19
7: 5
8: 0
I'm not completely sure how the sim works, I assumed that it will not trash for Provinces, but I'm nt sure. Perhaps I should parse for this too, we will see. I 8 cases it does not trash at all, and in more of half of the remaining it would have had $6 anyway. Pushing it from 5 to 6 in 15/42 ~ 1/3 of the cases.
 

[Geronimoo]

Play rules for Mining Village:

http://dominionsimulator.wordpress.com/play-rules/intrigue/mining-village/

[DStu]

Play rules for Mining Village:
http://dominionsimulator.wordpress.com/play-rules/intrigue/mining-village/

That's what I thought. The question is if the Gold conditioned on "inPlay(MV)" prevends the MV from being trashed. I think that should be the case. But than I hadn't understand how you could have $7 when you trash the MV in 10% of the cases. I think I do now, that should be you have in hand $5+MV, draw Estate.

The 8 times where it didn't trash anything also strongly indicades that the Gold works. Any idea how i can prevent it from getting trashed for silver? I would need a "if inPlay(MV) buy nothing" clause in front of the "buy silver", but probably its not so important anyway. Could of course buy Copper, is probably better than trashing MV for Silver.

[DStu]

Did the same for Witches:

Code: [Select]

<player name="BM - Witch/MV">
   <buy name="Gold">
      <condition>
         <left type="countCardsInPlay" attribute="Mining_Village"/>
         <operator type="greaterThan" />
         <right type="constant" attribute="0.0"/>
      </condition>
   </buy>
   <buy name="Province"/>
   <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="smallerThan" />
         <right type="constant" attribute="2.0"/>
      </condition>
   </buy>
   <buy name="Gold"/>
   <buy name="Mining_Village">
      <condition>
         <left type="countCardsInSupply" attribute="Mining_Village"/>
         <operator type="equalTo" />
         <right type="constant" attribute="10.0"/>
      </condition>
   </buy>
   <buy name="Silver"/>
</player>
vs

Code: [Select]

<player name="BM - Witch">
   <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="smallerThan" />
         <right type="constant" attribute="2.0"/>
      </condition>
   </buy>
   <buy name="Gold"/>
   <buy name="Silver"/>
</player>
(BM-Witch with Witch>Gold) and get

Code: [Select]

0: 1
1: 0
2: 1
3: 6
4: 22
5: 36
6: 28
7: 3
8: 0
So about 22-28 "good cases" where you escaped the $4 in 97 total.

[Fangz]

I think in the "strong $5 terminal" case, the advantage is also that if you don't trash the MV and still get to $5, you aren't worse off than if you had a silver, and are actually more or less a bit better off because (a) your deck is thinner so you can play that $5 more often, and (b) you have a village so that's one village you don't have to waste a buy on down the line if you are playing an engine strategy.

The key part I guess does come down to the 'if your deck is always improved by a silver', which I dunno, I don't really hold is true *that* often, except for very simple BM+* strategies. As a (terrible, I know) example, compare how often people pick the +4 silvers option from lucky steed, instead of +2 actions or +2 cards whatever. That seems to say that an extra action is worth a lot more than even 4 silvers in one's deck.

[WanderingWinder]

The key part I guess does come down to the 'if your deck is always improved by a silver', which I dunno, I don't really hold is true *that* often, except for very simple BM+* strategies. As a (terrible, I know) example, compare how often people pick the +4 silvers option from lucky steed, instead of +2 actions or +2 cards whatever. That seems to say that an extra action is worth a lot more than even 4 silvers in one's deck.

I don't think there are really THAT many boards where you're chaining so much that you don't really want to add any number of silvers. And just because people don't take the 4 silvers off steed doesn't mean they shouldn't. Finally, 2 cards and 2 actions is huge, would have to cost at least 6, so that's not a great thing to compare it to. Also, by the time you're playing steed, the extra silver won't generally mean nearly as much as it does in the opening. Basically, there are almost no decks I can think of that don't want at least 2 silvers.

[chwhite]

The key part I guess does come down to the 'if your deck is always improved by a silver', which I dunno, I don't really hold is true *that* often, except for very simple BM+* strategies. As a (terrible, I know) example, compare how often people pick the +4 silvers option from lucky steed, instead of +2 actions or +2 cards whatever. That seems to say that an extra action is worth a lot more than even 4 silvers in one's deck.

I don't think there are really THAT many boards where you're chaining so much that you don't really want to add any number of silvers. And just because people don't take the 4 silvers off steed doesn't mean they shouldn't. Finally, 2 cards and 2 actions is huge, would have to cost at least 6, so that's not a great thing to compare it to. Also, by the time you're playing steed, the extra silver won't generally mean nearly as much as it does in the opening. Basically, there are almost no decks I can think of that don't want at least 2 silvers.

Really?  What about:

*Minion
*Lighthouse + any strong attack
*FV + good card-draw
*Conspirator chains
*double-Tactician
*Colony, Gardens, or Vineyard (well maybe not necessarily Gardens)
*etc. etc. etc. that's just some of the more obvious choices

Even if BM-esque strategies are stronger than people think they are, there are going to be a large fraction of games in which excess Silver just weighs you down. 

[DG]

Basically, there are almost no decks I can think of that don't want at least 2 silvers.

I suspect you mean silver or silver-equivalent cards. The council room says that players buy/gain silver 88.4% of the time, the win rate with silver is 0.98, the win rate without silver is 1.15!

[WanderingWinder]

Basically, there are almost no decks I can think of that don't want at least 2 silvers.

I suspect you mean silver or silver-equivalent cards. The council room says that players buy/gain silver 88.4% of the time, the win rate with silver is 0.98, the win rate without silver is 1.15!

Indeed I did! My mistake!

[Buggz]

If you buy silvers and end up trashing them all by the time you win a game, does that count as a win with or without?

[Fangz]

That counts as a win with. Otherwise stats for things like treasure map would be silly.

[Buggz]

Then I'll take a guess and say that rate (for silver) is a bit biased, because any player who deliberately skips buying silvers alltogether are more likely to have a plan and follow it through while beginners will clutter their deck with silvers even in late game. And lose.

[WanderingWinder]

Basically, there are almost no decks I can think of that don't want at least 2 silvers.

I suspect you mean silver or silver-equivalent cards. The council room says that players buy/gain silver 88.4% of the time, the win rate with silver is 0.98, the win rate without silver is 1.15!

Indeed I did! My mistake!

Actually, thinking more about what I was thinking, what I meant was that there are almost no decks (which is different from 'no decks', the almost is important) that don't want at least two silvers (not even silver equivs), but that's presuming there's no opportunity cost to buying silver, i.e. you can put two free silvers in your discard pile at some point, there's few decks that don't want that.

[WanderingWinder]

Then I'll take a guess and say that rate (for silver) is a bit biased, because any player who deliberately skips buying silvers alltogether are more likely to have a plan and follow it through while beginners will clutter their deck with silvers even in late game. And lose.

I can't imagine that there are almost any players who buy a bunch of silvers and then would trash them later with something like chapel, except the good players that know that on an extremely rare occasion you need the silver to ramp you up early, then you have some mammoth combo that needs absolutely no fat later (can we say KC-KC-Goons-Masq?) Using the argument for trash-for-benefit would then imply that like province is a bad card, 'cause you'll salvage it to end the game, but that's clearly not true.

[Buggz]

No, I meant that if you're not buying any silvers you must have quite a good reason for it. That's my interpretation of the win without silver being higher than with.