Dominion Strategy Forum

Please login or register.

Login with username, password and session length
Pages: [1] 2  All

Author Topic: Card Strategy: Wharf  (Read 20513 times)

0 Members and 1 Guest are viewing this topic.

Epoch

  • Tactician
  • *****
  • Offline Offline
  • Posts: 421
  • Respect: +38
    • View Profile
Card Strategy: Wharf
« on: September 12, 2011, 01:49:46 pm »
0

Wharf is, bar none, the best card-drawer in the game.

And on some level, that's what you need to know.  Card drawing is valuable.  Wharf is the best one.  It works well in all the cases that you want to draw cards.  It excels, paired with a village-type card, in a cards/actions engine.  It is also extremely good, bought in moderation, with a big money deck, or practically any other deck.  As a terminal, it should be bought somewhat sparingly if you do not have a village-type card.  And that's pretty much how card-drawers work in Dominion.

But it may be a bit surprising that Wharf is the best card drawer in the game.  It, after all, gives only +2 Cards, in the same category as such low-tier terminal drawers such as Witch, Steward, and Vault (all of which are good cards, mind you, but not principally due to their ability to draw cards).  Of course, Wharf gives you the +2 Cards effect twice, but the general rule in Dominion is that one big turn counts more than two little turns, so you might imagine that Smithy, with its +3 Cards now, would beat Wharf, with its +4 Cards split up in two.

As it turns out, however, the +2 now/+2 later dynamic really works, and is, especially in a cards/action engine, perhaps better than a single massive draw, like you'd get with Council Room or Envoy.  If you're building an engine, what you really want is consistency.  It's easy to build an engine that sometimes draws your entire deck with 6 cards to spare, and other times stalls out instantly.  Village/Wharf engines are substantially less likely to do that than any other two-card engine.  The action economy of Wharf, and its consistency across turns, allows you to MUCH better weather turns when Villages are sparse in the top of your deck, and leaving several Wharves out each turn effectively lowers the size of your deck, increasing the density of the Actions you want to play next turn.  Nothing cures the deck-stalling blues like using Wharf as your card-drawer, and no two-card engine will remain powerful longer during the greening phase than village/wharf.

And, speaking of buying greens, the one thing that you always want with a cards/actions engine is +buy to let you catch up with faster-to-accelerate but lower top-speed decks like Big Money.  Convenient, then that Wharf gives you that built right in to your engine, which enables you to both often buy multiple engine components in the early-mid game, and then catch up on greens in the late game, without costly diversions to buy a Market, Hamlet, Pawn, Festival or even (shudder) a Woodcutter or Herbalist.

Indeed, the draw of Wharf is so powerful that you can more-or-less make it work with even the +2 Actions/no cards village variants such as Festival, University, or Shanty Town.  University arguably makes up for its lack of +Cards by enabling Wharf draws, but these engines are certainly more prone to stalling and failing than a vanilla Village/Wharf engine.  Still, Wharf is powerful enough that, given a decent board, they can still be victorious in a way that, say, Shanty Town/Smithy never could be.

How good is a Village/Wharf engine?  It's so good that Village/Wharf, with no other enablers, can completely stomp Big Money Ultimate on a Province board, a distinction it shares only with (to my knowledge) Village/Torturer.  In fact, this simple Village/Wharf engine also beats BMU + 1 Wharf, which is quite a high bar.  It utterly destroys BMU, 77-20-3.  (Notably, in Province games without other enablers, other BMU variants may beat all but the best-made Village/Wharf engines, such as BMU + 2 Wharves.  But that's the nature of engines in Province games, and the built-in +buy of Wharf does mean that when anything even vaguely enabling is on the board, you can pick it up pretty easily).

Code: [Select]
<player name="Village/Wharf Simple">
   <buy name="Province"/>
   <buy name="Duchy">
      <condition>
         <left type="countCardsInSupply" attribute="Province"/>
         <operator type="smallerOrEqualThan" />
         <right type="constant" attribute="3.0"/>
      </condition>
   </buy>
   <buy name="Estate">
      <condition>
         <left type="countCardsInSupply" attribute="Province"/>
         <operator type="smallerOrEqualThan" />
         <right type="constant" attribute="2.0"/>
      </condition>
   </buy>
   <buy name="Wharf">
      <condition>
         <left type="countCardsInDeck" attribute="Wharf"/>
         <operator type="equalTo" />
         <right type="constant" attribute="0.0"/>
      </condition>
   </buy>
   <buy name="Gold">
      <condition>
         <left type="countCardsInDeck" attribute="Gold"/>
         <operator type="smallerThan" />
         <right type="constant" attribute="2.0"/>
      </condition>
   </buy>
   <buy name="Wharf">
      <condition>
         <left type="countCardsInDeck" attribute="Wharf"/>
         <operator type="smallerOrEqualThan" />
         <right type="countCardTypeInDeck" attribute="Village"/>
         <extra_operation type="plus" attribute="3.0" />
      </condition>
   </buy>
   <buy name="Village">
      <condition>
         <left type="countCardsInDeck" attribute="Village"/>
         <operator type="smallerThan" />
         <right type="countCardsInDeck" attribute="Wharf"/>
         <extra_operation type="minus" attribute="1.0" />
      </condition>
   </buy>
   <buy name="Gold"/>
   <buy name="Silver"/>
</player>

When not paired with a Village, well, Wharf is still a terminal.  It's a good terminal, and +cards is welcome in pretty much any kind of deck, but you can't flood your deck with it.  Its Duration effect means you can buy a few more Wharves than you otherwise would -- 2 Wharves + BMU beats 1 Wharf + BMU, but a third Wharf doesn't help any (neither does it hurt).  In a non-Village situation, it's a strong terminal that doesn't require a lot of strategy to play.

Getting back to the central claim of this article:  Wharf is the best card drawer in the game.  By that, I mean that it is the best card drawer qua card drawer.  Torturer will usually beat it as a compliment to Village if your opponent can't deal with the attack of the Torturer, certainly, but Wharf will allow you to more consistently draw your deck.  Council Room is the only card that challenges Wharf's supremacy in terms of allowing you to draw, but the advantage it gives to your opponents is potent.  Council Rooms paired with Militia or Goons will tend to beat out Wharf, however -- +4 Cards in one turn is better than +4 Cards over two turns, if Council Room's singular disadvantage can be ignored.

The final verdict: Wharf is one of the best all-round cards in the game.  There are hardly any boards on which it should be ignored, and if there is a Village-type card in the game, it is likely that Wharf will dominate the game and its supply pile will be bought out.
Logged

ackack

  • Explorer
  • *****
  • Offline Offline
  • Posts: 302
  • Respect: +19
    • View Profile
Re: Card Strategy: Wharf
« Reply #1 on: September 12, 2011, 02:05:53 pm »
+1

. . . and if there is a Village-type card in the game, it is likely that Wharf will dominate the game and its supply pile will be bought out.

Which leads to the more interesting strategy problems with Wharf - in big, robust engine games with tons of buy and increased risk of abrupt 3 piling, how does endgame timing work? Wharf is one of the most prominent cards to ask this about because one big Wharf turn suggests another one will be coming, which is an important part of the calculus. In general I feel heuristics for this are quite primitive, at this point.
Logged

WanderingWinder

  • Adventurer
  • ******
  • Offline Offline
  • Posts: 5275
  • ...doesn't really matter to me
  • Respect: +4381
    • View Profile
    • WanderingWinder YouTube Page
Re: Card Strategy: Wharf
« Reply #2 on: September 12, 2011, 02:07:08 pm »
+1

Your bot gets stomped by a village-less BM/Wharf Bot, ~54-40:
Code: [Select]
<player name="Wharf">
   <buy name="Gold">
      <condition>
         <left type="countCardsInPlay" attribute="Wharf"/>
         <operator type="greaterThan" />
         <right type="constant" attribute="0.0"/>
      </condition>
      <condition>
         <left type="countAvailableMoney"/>
         <operator type="equalTo" />
         <right type="constant" attribute="9.0"/>
      </condition>
      <condition>
         <left type="countCardsInSupply" attribute="Province"/>
         <operator type="greaterOrEqualThan" />
         <right type="constant" attribute="6.0"/>
      </condition>
   </buy>
   <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="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="Wharf">
      <condition>
         <left type="countCardsInDeck" attribute="Wharf"/>
         <operator type="smallerThan" />
         <right type="countCardTypeInDeck" attribute="Treasure"/>
         <extra_operation type="divideBy" attribute="4.0" />
      </condition>
   </buy>
   <buy name="Silver"/>
</player>

Epoch

  • Tactician
  • *****
  • Offline Offline
  • Posts: 421
  • Respect: +38
    • View Profile
Re: Card Strategy: Wharf
« Reply #3 on: September 12, 2011, 02:21:34 pm »
0

Your bot gets stomped by a village-less BM/Wharf Bot, ~54-40:

Yeah, I know.  As I said, "Notably, in Province games without other enablers, other BMU variants may beat all but the best-made Village/Wharf engines, such as BMU + 2 Wharves."

My feeling is that with a high-quality human player on the Village/Wharf engine, it can recover a lot of ground there by playing more sensitively than you (or, at least, I) can code into the simulator: when to go for broke with a big green buy-out, when to save up for a mega-turn, etc.

But my point is not that that rather simple bot is the be-all, end-all of Village/Wharf engines, just that the engine has such strong qualities that it's not a total non-starter, like Village/Smithy would be under the same conditions.  Obviously, there are the other 8 Kingdom cards to consider, too.
Logged

tko

  • Young Witch
  • ****
  • Offline Offline
  • Posts: 129
  • Respect: +1
    • View Profile
Re: Card Strategy: Wharf
« Reply #4 on: September 12, 2011, 02:25:17 pm »
0

. . . and if there is a Village-type card in the game, it is likely that Wharf will dominate the game and its supply pile will be bought out.

Which leads to the more interesting strategy problems with Wharf - in big, robust engine games with tons of buy and increased risk of abrupt 3 piling, how does endgame timing work? Wharf is one of the most prominent cards to ask this about because one big Wharf turn suggests another one will be coming, which is an important part of the calculus. In general I feel heuristics for this are quite primitive, at this point.
It seems like Wharf speeds a Province game so quickly that you don't want get a lot of time to catch up if you fall behind.  Especially when it turns into a mirror match, it behooves you to take the initiative and buy the first province putting the pressure on your opponent (after getting a couple Gold and a couple Wharves in your deck, for example).

Here's a mirror match I lost where Sea Hag was hanging out by the Wharf:
http://councilroom.com/game?game_id=game-20110716-063016-5c13bd1c.html

It also speeds up a Colony game.  Here I got destroyed by 7 Wharves:
http://councilroom.com/game?game_id=game-20110509-043826-01394efe.html
And by a rule I try to limit my Wharf purchases... show's you where rules get you since the game is so dynamic from one kingdom to the next.
Logged

Geronimoo

  • Saboteur
  • *****
  • Offline Offline
  • Posts: 1058
  • Respect: +868
    • View Profile
    • Geronimoo's Dominion Simulator
Re: Card Strategy: Wharf
« Reply #5 on: September 12, 2011, 02:48:14 pm »
0

I doubt there's much +EV in adding vanilla Village to a Wharf deck. Fishing Village.... now that's another story!

Code: [Select]
<player name="FV Wharf">
   <buy name="Province">
      <condition>
         <left type="countAvailableMoney"/>
         <operator type="greaterOrEqualThan" />
         <right type="constant" attribute="16.0"/>
      </condition>
   </buy>
   <buy name="Province">
      <condition>
         <left type="countCardsInDeck" attribute="Province"/>
         <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="3.0"/>
      </condition>
   </buy>
   <buy name="Estate">
      <condition>
         <left type="countCardsInSupply" attribute="Province"/>
         <operator type="smallerOrEqualThan" />
         <right type="constant" attribute="2.0"/>
      </condition>
   </buy>
   <buy name="Wharf">
      <condition>
         <left type="countCardsInDeck" attribute="Wharf"/>
         <operator type="equalTo" />
         <right type="constant" attribute="0.0"/>
      </condition>
   </buy>
   <buy name="Gold">
      <condition>
         <left type="getTotalMoney"/>
         <operator type="smallerThan" />
         <right type="constant" attribute="16.0"/>
      </condition>
   </buy>
   <buy name="Wharf"/>
   <buy name="Fishing_Village"/>
</player>
Logged

WanderingWinder

  • Adventurer
  • ******
  • Offline Offline
  • Posts: 5275
  • ...doesn't really matter to me
  • Respect: +4381
    • View Profile
    • WanderingWinder YouTube Page
Re: Card Strategy: Wharf
« Reply #6 on: September 12, 2011, 04:18:11 pm »
0

Your bot gets stomped by a village-less BM/Wharf Bot, ~54-40:

Yeah, I know.  As I said, "Notably, in Province games without other enablers, other BMU variants may beat all but the best-made Village/Wharf engines, such as BMU + 2 Wharves."

My feeling is that with a high-quality human player on the Village/Wharf engine, it can recover a lot of ground there by playing more sensitively than you (or, at least, I) can code into the simulator: when to go for broke with a big green buy-out, when to save up for a mega-turn, etc.

But my point is not that that rather simple bot is the be-all, end-all of Village/Wharf engines, just that the engine has such strong qualities that it's not a total non-starter, like Village/Smithy would be under the same conditions.  Obviously, there are the other 8 Kingdom cards to consider, too.

I seriously doubt this for the case of village/wharf with nothing else - just wharves are so darn fast, your mega turn won't come before you can't possibly win. Actually, I don't see how wharf itself suggests going for a mega turn. I mean, you're going to be drawing the same money later that you are now, and it seems like it's likely to just be slower.
And yes, here, as in most places, FV is gobs better than village.

Epoch

  • Tactician
  • *****
  • Offline Offline
  • Posts: 421
  • Respect: +38
    • View Profile
Re: Card Strategy: Wharf
« Reply #7 on: September 12, 2011, 04:43:48 pm »
0

I seriously doubt this for the case of village/wharf with nothing else - just wharves are so darn fast, your mega turn won't come before you can't possibly win.

I certainly admit that this particular situation isn't one in which I have a lot of practical experience.

Actually, I don't see how wharf itself suggests going for a mega turn. I mean, you're going to be drawing the same money later that you are now, and it seems like it's likely to just be slower.

Uh, the point would be to draw more money.  Figure that LIKELY, you're wasting buys with the ordinary play of Village/Wharves -- skip a turn of $12 -> 1 Province, instead go $12 -> 2 Golds, and then next turn go $18 -> 2 Provinces, is my basic concept, here.

Now, like I said, maybe that's not a common enough situation to actually work, maybe the whole thing fails.  I haven't ever tried "pure Village/Wharf versus BM + several Wharves."  But, again, the point is not to spend a ton of time on the artificial case of "pure Village/Wharf," it's to point out how much superior a Village/Wharf engine is to the (also artificial; default; baseline) Village/Smithy engine.  I did try to disclaim that in the first post.
Logged

tko

  • Young Witch
  • ****
  • Offline Offline
  • Posts: 129
  • Respect: +1
    • View Profile
Re: Card Strategy: Wharf
« Reply #8 on: September 12, 2011, 04:46:17 pm »
0

If you're playing BMU + 2 Wharves, and you hit 11 coin and 2 buys, I see the benefit in buying Province + Silver.  What I wonder is this... is Province + Village also acceptable in that case?  On one hand, opportunity cost is not much an issue.  On the other hand, it introduces a potentially dead card (which Silver would not be) and therefore maybe still a bad idea.
Logged

WanderingWinder

  • Adventurer
  • ******
  • Offline Offline
  • Posts: 5275
  • ...doesn't really matter to me
  • Respect: +4381
    • View Profile
    • WanderingWinder YouTube Page
Re: Card Strategy: Wharf
« Reply #9 on: September 12, 2011, 05:02:07 pm »
0

Uh, the point would be to draw more money.  Figure that LIKELY, you're wasting buys with the ordinary play of Village/Wharves -- skip a turn of $12 -> 1 Province, instead go $12 -> 2 Golds, and then next turn go $18 -> 2 Provinces, is my basic concept, here.
Yeah, but if you do that (instead of province-silver or province-village or even province-estate at some point), then you need more than one mega turn to make that worth it; just one double-province and you're not at all ahead, it's not a mega enough turn. Of course this is a very common idea, but the problem is that with this kind of thing, you end up building super-strong engines which can still sometimes gum up but more importantly are falling way behind in points.
As to your other point about wharf being better than smithy for engines.... maybe. The problem is, you still need a reason to go for such a thing, and wharf costs 5. And by the time you're drawing your whole deck, as these engines often do, you start running into other issues. So probably Wharf is normally better than smithy there, but it's not entirely cut-and-dried.

Epoch

  • Tactician
  • *****
  • Offline Offline
  • Posts: 421
  • Respect: +38
    • View Profile
Re: Card Strategy: Wharf
« Reply #10 on: September 12, 2011, 05:30:49 pm »
0

As to your other point about wharf being better than smithy for engines.... maybe.

Well, actually, not "maybe."  That was the point.  Village/Smithy chokes to Big Money Ultimate, and utterly dies to BMU + 1 Smithy.  Village/Wharf crushes BMU and solidly beats BMU + 1 Wharf (and BMU + 1 Wharf is better than BMU + 1 Smithy).

And, again, because it really seems like you're missing this: the point is not to get too caught up in what pure Village/Wharf or pure Village/Smithy does, because we don't actually play with two-card tables.  The point is, okay, try to quantify how much better the baseline engine is, so that we can then go on and say, "Okay, given this board, and given that Village/Wharf is a pretty strong base for a deck, now what could I do with the other cards on the table with it?"

This seems really obvious to me, but maybe that's idiosyncratic, because other people I know to be really smart don't seem to get it.  But you need to understand the strength of your basic engine in order to say, "the deck I build off the base of Village/Wharf with this particular table is... good/bad/whatever."

Maybe another way of saying this is, "Suppose you have Village, 8 cards, and then 1 of Smithy or Wharf.  Depending on those 8 other cards, you'll be contemplating a cards/actions engine.  The other 8 cards can be less ideal support for a cards/actions engine if your 10th card is Wharf than if your 10th card is Smithy."

That's not universally, completely true, I know.  But we have to have some way to address talking about a card without talking specifically about one particular table.  I think in this case, it's true a large enough percentage of the time that we can usefully make that kind of comparison, and I think that there is value in the observation that there base Village/Wharf engine is a strong one, as +2 Action/Card-Drawer engines go. 


EDIT:  To be clear, WanderingWinder is one of the people I categorize as "I know to be really smart."
« Last Edit: September 12, 2011, 05:45:42 pm by Epoch »
Logged

WanderingWinder

  • Adventurer
  • ******
  • Offline Offline
  • Posts: 5275
  • ...doesn't really matter to me
  • Respect: +4381
    • View Profile
    • WanderingWinder YouTube Page
Re: Card Strategy: Wharf
« Reply #11 on: September 12, 2011, 06:02:09 pm »
0

As to your other point about wharf being better than smithy for engines.... maybe.

Well, actually, not "maybe."  That was the point.  Village/Smithy chokes to Big Money Ultimate, and utterly dies to BMU + 1 Smithy.  Village/Wharf crushes BMU and solidly beats BMU + 1 Wharf (and BMU + 1 Wharf is better than BMU + 1 Smithy).
Well, actually, yes maybe  ::). This argument is non-persuasive because...

Quote
And, again, because it really seems like you're missing this: the point is not to get too caught up in what pure Village/Wharf or pure Village/Smithy does, because we don't actually play with two-card tables.  The point is, okay, try to quantify how much better the baseline engine is, so that we can then go on and say, "Okay, given this board, and given that Village/Wharf is a pretty strong base for a deck, now what could I do with the other cards on the table with it?"
...of this, which you just seem to ignore above. Yes, a deck with villages and wharves is better than one with villages and smithies. But that's not why you're building the engine, so it's really pointless. So for all the reasons above (particularly the cost difference), it's not so easy to just say wharf>smithy for engines.

Quote
This seems really obvious to me, but maybe that's idiosyncratic, because other people I know to be really smart don't seem to get it.  But you need to understand the strength of your basic engine in order to say, "the deck I build off the base of Village/Wharf with this particular table is... good/bad/whatever."

Maybe another way of saying this is, "Suppose you have Village, 8 cards, and then 1 of Smithy or Wharf.  Depending on those 8 other cards, you'll be contemplating a cards/actions engine.  The other 8 cards can be less ideal support for a cards/actions engine if your 10th card is Wharf than if your 10th card is Smithy."

That's not universally, completely true, I know.  But we have to have some way to address talking about a card without talking specifically about one particular table.  I think in this case, it's true a large enough percentage of the time that we can usefully make that kind of comparison, and I think that there is value in the observation that there base Village/Wharf engine is a strong one, as +2 Action/Card-Drawer engines go. 
Yes, more or less this is how you should look at it. But you really *can't* just look at the "base engine", because the "base engine" doesn't do anything by itself (of course this isn't, strictly speaking, true, but given that it's worse than playing straight wharf or straight smithy, we'll just make this assumption without too much harm). You *must* look at the other supporting cards, on a case-by-case basis, and where I disagree with you is that it's a large enough percentage of the time to just say one is a *better* engine than the other.


I get the feeling that you *probably* understand all this and just have different standards on how dominant something must be to be "better", and/or you've done a different (not necessarily worse, but I would obviously suspect worse, considering that the opposing position is my own) set of analyses than I have, which is fine.
In fact, I suspect that Village/Wharf engines are going to be the best strategy *less often* than Village/Smithy ones, but this is more due to wharf working better with money than smithy does than it actually being a worse engine; indeed I think it's usually a better engine, just not nearly often enough to flatly say "it's better".

guided

  • Jester
  • *****
  • Offline Offline
  • Posts: 940
  • Respect: +94
    • View Profile
Re: Card Strategy: Wharf
« Reply #12 on: September 12, 2011, 06:08:53 pm »
0

And, again, because it really seems like you're missing this: the point is not to get too caught up in what pure Village/Wharf or pure Village/Smithy does, because we don't actually play with two-card tables.
Amen. Simulation is an interesting exercise that can be one input to informing us about strategy, but focusing on simulation to the exclusion of what actually happens on real 10-card boards is missing the forest for the trees. Micro-optimizing 2-card strategies against BMU is a neat math problem, and I've spent some time doing it on occasion, but it does not help you all that much in becoming a better Dominion player.

Another musing: One thing that is all too easily forgotten is that real opponents simply do not play BMU, and they rarely play BMU plus 1 card, even when it might be possible through careful analysis and simulation to gain confidence that BMU plus 1 card is the best strategy on a particular board. I do feel a sense of accomplishment when I win a game against a high-ranked opponent with single-Smithy or single-Envoy (and I still do on occasion), but even then, it generally wasn't a terribly fun game for me to play!
Logged

Epoch

  • Tactician
  • *****
  • Offline Offline
  • Posts: 421
  • Respect: +38
    • View Profile
Re: Card Strategy: Wharf
« Reply #13 on: September 12, 2011, 06:41:32 pm »
0

Yes, more or less this is how you should look at it. But you really *can't* just look at the "base engine", because the "base engine" doesn't do anything by itself (of course this isn't, strictly speaking, true, but given that it's worse than playing straight wharf or straight smithy, we'll just make this assumption without too much harm). You *must* look at the other supporting cards, on a case-by-case basis, and where I disagree with you is that it's a large enough percentage of the time to just say one is a *better* engine than the other.

I think perhaps the source of our differences is this:  I think that "performance against BMU or BMU + 1 card that's part of the engine" is a good metric for overall utility of the engine; you don't.

Here's why I think that:

1.  Understand an engine vs. BMU rating to be essentially a race.  All engines will eventually become stronger than BMU, but usually won't consistently generate high-value hands until late in the game.  So if one engine does better than another vs. BMU, that basically means that that engine develops faster (that is to say: it consistently generates high-value hands earlier in the game).

2.  So that means that whatever other cards that you want to buy besides your basic engine components, a "better" engine like Village/Wharf will have more ability to buy them, without significantly impairing its development as an engine, notwithstanding the higher cost of Wharf compared to Smithy.  This is really very intrinsic to the concept.  If the lower cost of Smithy were enough to make the Village/Smithy engine good at buying whatever, including other enabling cards, then it would also make it good at buying green cards, and thus it would perform better against BMU.

3.  To a lesser extent, engine vs. BMU ratings measure the "completeness" of an engine.  Village/Wharf has plentiful +buy, which Village/Smithy lacks, of course.  Now, we know that qualitatively from the start, but to what extent does it affect things?  The rating vs. BMU is a very lossy way of determining that, but it's better than nothing.

4.  The above does not follow in the case of a board which has specific synergy with Smith, but does not have that synergy with Wharf, but...  what exactly would that board look like?  King's Court is the only thing that jumps to mind as probably better with Smithy than Wharf.
Logged

Epoch

  • Tactician
  • *****
  • Offline Offline
  • Posts: 421
  • Respect: +38
    • View Profile
Re: Card Strategy: Wharf
« Reply #14 on: September 12, 2011, 06:42:42 pm »
0

Another musing: One thing that is all too easily forgotten is that real opponents simply do not play BMU, and they rarely play BMU plus 1 card, even when it might be possible through careful analysis and simulation to gain confidence that BMU plus 1 card is the best strategy on a particular board. I do feel a sense of accomplishment when I win a game against a high-ranked opponent with single-Smithy or single-Envoy (and I still do on occasion), but even then, it generally wasn't a terribly fun game for me to play!

Yeah, I've always kind of wondered how highly ranked a bot would be that plays Isotropic routinely and just always, always plays BMU + one card.  I suspect that the answer would be "depressingly highly rated."  Quite possibly higher-rated than I am.
Logged

ftl

  • Mountebank
  • *****
  • Offline Offline
  • Posts: 2056
  • Shuffle iT Username: ftl
  • Respect: +1345
    • View Profile
Re: Card Strategy: Wharf
« Reply #15 on: September 12, 2011, 07:09:06 pm »
0

Would Isotropic allow such a bot? I'm tempted to put one together.
Logged

tlloyd

  • Tactician
  • *****
  • Offline Offline
  • Posts: 404
  • Respect: +84
    • View Profile
Re: Card Strategy: Wharf
« Reply #16 on: September 12, 2011, 08:03:20 pm »
0

You both seem like smart people to me, but you also seem to be talking past each other. Without claiming to know who is right, let me see if I can clarify the disagreement:

Imagine a set of ten kingdom cards for which you both agree that a draw engine with Smithy at its core is the best strategy. Now I add Wharf as an 11th kingdom card. Epoch thinks that his draw engine is now even stronger, because Wharf gives comparable card draw and extra buys to boot. WW is having second thoughts about a draw engine altogether, because Wharf + BM is so strong.

So Epoch is in essence saying that a Wharf engine will beat a Smithy engine head-to-head. WW is saying that a Smithy draw engine is more likely to be a winning strategy than a Wharf draw engine, because Wharf helps BM more than it helps a draw engine. You may BOTH be right!

And forgive me if I have inaccurately translated - I may not be one of those smart people.  :D
Logged

WanderingWinder

  • Adventurer
  • ******
  • Offline Offline
  • Posts: 5275
  • ...doesn't really matter to me
  • Respect: +4381
    • View Profile
    • WanderingWinder YouTube Page
Re: Card Strategy: Wharf
« Reply #17 on: September 12, 2011, 09:15:06 pm »
0

You both seem like smart people to me, but you also seem to be talking past each other. Without claiming to know who is right, let me see if I can clarify the disagreement:

Imagine a set of ten kingdom cards for which you both agree that a draw engine with Smithy at its core is the best strategy. Now I add Wharf as an 11th kingdom card. Epoch thinks that his draw engine is now even stronger, because Wharf gives comparable card draw and extra buys to boot. WW is having second thoughts about a draw engine altogether, because Wharf + BM is so strong.

So Epoch is in essence saying that a Wharf engine will beat a Smithy engine head-to-head. WW is saying that a Smithy draw engine is more likely to be a winning strategy than a Wharf draw engine, because Wharf helps BM more than it helps a draw engine. You may BOTH be right!

And forgive me if I have inaccurately translated - I may not be one of those smart people.  :D
I think mostly right, and, as I said earlier, I think the two of us actually largely agree, just disagree on the terminology we're using and/or the implications of what we're saying.

DG

  • Governor
  • *****
  • Offline Offline
  • Posts: 4074
  • Respect: +2624
    • View Profile
Re: Card Strategy: Wharf
« Reply #18 on: September 12, 2011, 09:42:30 pm »
+1

Nothing after the original post really helps anyone understand the wharf any better either.

The big advantage from the wharf is the enlarged hand size on the following turn. This makes it ideal in setting up draw chains that aim to draw the majority of a deck each turn since a hand of 7 or more cards is far less likely to fail than a hand with 5 cards. This consistent drawing can be so strong that you can often delay treasure purchases, knowing that the wharves will draw enough of your existing treasures each turn. As the deck improves the wharves can buy more and more engine cards using the extra buys until you can buy many big treasures then buy many green cards.

Thrones and king's courts work especially well with wharves since you get more benefits without needing more actions, the big wharf drawn hands are less likely to draw thrones without actions, and the extra drawing from a throne on wharf is well worth leaving the throne as a duration card in clean-up. Using wharves to strengthen another drawing engine, such as hunting parties or torturers, can also work since playing just one wharf can considerably strengthen the following turn.

Once you have your wharves drawing most of your deck each turn you don't want it to stop. As soon as you go back to small starting hands you're going to draw less cards each turn and the wharf engine may never restart before the end of the game, especially if you're filling your deck with green cards. The ways to stop this stalling are to trash out rubbish, balance the number of wharves played each turn, and overbuild your drawing engine before you jump for treasures and green cards.
Logged

WanderingWinder

  • Adventurer
  • ******
  • Offline Offline
  • Posts: 5275
  • ...doesn't really matter to me
  • Respect: +4381
    • View Profile
    • WanderingWinder YouTube Page
Re: Card Strategy: Wharf
« Reply #19 on: September 12, 2011, 10:37:44 pm »
0

I agree with DG that the next-turn effect is the big boon (it's a smithy with a buy next turn, as opposed to a moat with a buy this turn), but disagree with most of the rest - wharf's sheer power makes it a very good engine card, and if you can throne or king it, great, but I think it's most disproportionately powerful spot is in money decks.

Mean Mr Mustard

  • Witch
  • *****
  • Offline Offline
  • Posts: 452
  • First to 5000 Isotropic wins
  • Respect: +118
    • View Profile
Re: Card Strategy: Wharf
« Reply #20 on: September 13, 2011, 12:21:53 am »
0

But the KC/KC effect with a great linchpin card can wreak some serious havoc.  Often.


My feeling is that with a high-quality human player on the Village/Wharf engine, it can recover a lot of ground there by playing more sensitively than you (or, at least, I) can code into the simulator: when to go for broke with a big green buy-out, when to save up for a mega-turn, etc.

This deserves a thread of it's own.
Logged
Jake <a href=http://dominion.isotropic.org/gamelog/201203/17/game-20120317-030206-6456f97c.html>opening: opening: Silver / Jack of All Trades</a>
<b>IsoDom1 Winner:  shark_bait
IsoDom2 Winner: Rabid
Isodom3 Winner: Fabian</b>
Utúlie'n aurë! Aiya Eldalie ar Atanatári, Utúlie'n auré!

Epoch

  • Tactician
  • *****
  • Offline Offline
  • Posts: 421
  • Respect: +38
    • View Profile
Re: Card Strategy: Wharf
« Reply #21 on: September 13, 2011, 02:01:31 am »
0

You both seem like smart people to me, but you also seem to be talking past each other. Without claiming to know who is right, let me see if I can clarify the disagreement:

...

WW is saying that a Smithy draw engine is more likely to be a winning strategy than a Wharf draw engine, because Wharf helps BM more than it helps a draw engine. You may BOTH be right!

Ah, I didn't get that at all, thanks for the translation.

I think that WW is just wrong, though.  There are, of course, situations where a BM + Wharf strategy is the strongest on the board, but to the extent that ANY Village/Smithy village is viable (stronger than BM + Smithy on that board), I think that Village/Wharf will be viable (stronger than BM + Wharf on that board), at least 99% of the time.

But... it's interesting to try to prove that.

I mean, I can modify my basic Village/Wharf engine to throw in an enabling card (I did, using Bank, the most enabling of all engine enabling cards), and show that it beats WW's BM + No Village + Wharf bot, also modified to add Bank (I did that, too).  But that doesn't really show anything.  There could be a BM bot out there, structured differently from mine, that beats my engine bot.  Conversely, there could be a cleverer engine bot out there that beats the BM bot.  I can't exhaust the result space of either BM bots or engine bots, and if there's one thing that I've learned from playing with the simulator, it's that surprisingly subtle buy rules can have a big effect on the game.

EDIT:  Actually, a stronger statement: it's not just that there "could" be better BM bots and engine bots.  There almost certainly ARE.

Which I suppose just indicates that Dominion is a good game that's susceptible to more complex strategies than "just buy good cards."

Did people look at WW's bot?  It's got a pretty interesting top buy rule, which, when I figured it out, translates to, "If you have $9 and 2 buys, and there are 6 or more Provinces in supply, instead of buying a Province, buy a Gold and a Silver."  When I made my Bank bots, I used a variation of that, and it really helped my engine's success.  But that's why I'm not even going to bother to post my new engine, because it's just down the rabbit hole -- more and more complicated bots that ultimately prove narrower and narrower things.

So: I think that WW is wrong that Wharf helps BM strategies more than it helps engine strategies.  But I don't think I can offer convincing proof that he's wrong.
« Last Edit: September 13, 2011, 02:03:46 am by Epoch »
Logged

WanderingWinder

  • Adventurer
  • ******
  • Offline Offline
  • Posts: 5275
  • ...doesn't really matter to me
  • Respect: +4381
    • View Profile
    • WanderingWinder YouTube Page
Re: Card Strategy: Wharf
« Reply #22 on: September 13, 2011, 10:07:51 am »
0

You both seem like smart people to me, but you also seem to be talking past each other. Without claiming to know who is right, let me see if I can clarify the disagreement:

...

WW is saying that a Smithy draw engine is more likely to be a winning strategy than a Wharf draw engine, because Wharf helps BM more than it helps a draw engine. You may BOTH be right!

Ah, I didn't get that at all, thanks for the translation.

I think that WW is just wrong, though.  There are, of course, situations where a BM + Wharf strategy is the strongest on the board, but to the extent that ANY Village/Smithy village is viable (stronger than BM + Smithy on that board), I think that Village/Wharf will be viable (stronger than BM + Wharf on that board), at least 99% of the time.

But... it's interesting to try to prove that.

I mean, I can modify my basic Village/Wharf engine to throw in an enabling card (I did, using Bank, the most enabling of all engine enabling cards), and show that it beats WW's BM + No Village + Wharf bot, also modified to add Bank (I did that, too).  But that doesn't really show anything.  There could be a BM bot out there, structured differently from mine, that beats my engine bot.  Conversely, there could be a cleverer engine bot out there that beats the BM bot.  I can't exhaust the result space of either BM bots or engine bots, and if there's one thing that I've learned from playing with the simulator, it's that surprisingly subtle buy rules can have a big effect on the game.

EDIT:  Actually, a stronger statement: it's not just that there "could" be better BM bots and engine bots.  There almost certainly ARE.

Which I suppose just indicates that Dominion is a good game that's susceptible to more complex strategies than "just buy good cards."

Did people look at WW's bot?  It's got a pretty interesting top buy rule, which, when I figured it out, translates to, "If you have $9 and 2 buys, and there are 6 or more Provinces in supply, instead of buying a Province, buy a Gold and a Silver."  When I made my Bank bots, I used a variation of that, and it really helped my engine's success.  But that's why I'm not even going to bother to post my new engine, because it's just down the rabbit hole -- more and more complicated bots that ultimately prove narrower and narrower things.

So: I think that WW is wrong that Wharf helps BM strategies more than it helps engine strategies.  But I don't think I can offer convincing proof that he's wrong.

It's a little more subtle than that. I think Wharf helps money strategies compared to engine strategies more than smithy helps money strategies compared to its engine strategies. So the position you're attributing to me is only right if you ignore the fact that engine strategies are rather often more potent than the simple money strategies.
As for the top condition on my wharf bot - it's not just that I have this one improvement randomly. I spent quite a lot of time tuning the bot, and that was actually the only thing that really helped. Going for double wharf on 10 didn't, going for wharf-gold on 11 didn't, double gold on 12 didn't, etc. Now of course there are other ways of improving my bot - changing the buy rules on the first province to a total-money-in-the-deck thing, figuring out when you should prefer wharf to gold more precisely, changing the buy condition on wharf to a total-cards-in-deck condition (now that we can), I think this is really about all that can be optimized. Possibly there's some PPR-like stuff and I'm-not-going-to-make-it-to-the-next-reshuffle-so-I-should-buy-estates-a-little-early kind of stuff too, but I think it's about right. As far as a big money+1 strategy, I don't see any reason why we can't, for wharf, program the bot to basically do as well as any human player, given we give it enough conditions. For other cards, this won't be right at all.
Village/Wharf being stronger than Wharf/money on 99+% of boards is ridiculously too high. You have to have some reason to go for an engine, and not nearly 99% of boards actually have that. You also have to have time to set your fancy engine up. There simply isn't much of that in these games. I would guess that the engines are going to be significantly better roughly 40% of the time, and pretty close around 30% (clearly worse around 30%). Switch to fishing or some other village and you've got a different story.

WanderingWinder

  • Adventurer
  • ******
  • Offline Offline
  • Posts: 5275
  • ...doesn't really matter to me
  • Respect: +4381
    • View Profile
    • WanderingWinder YouTube Page
Re: Card Strategy: Wharf
« Reply #23 on: September 13, 2011, 10:12:36 am »
0

I should note that there's also a lot of room for improvement in reacting to your opponent. The bots blindly play their own strategy, but if your opponent is playing some fancy strategy, you should adjust yours somewhat, either to make it faster or to make it have better longevity - whatever you need to overcome your opponent's strategy, much as guided suggests above. Now, I think guided might be overrating this a little bit in his comments (my personal feel), but this does have an absolutely HUGE impact.

DG

  • Governor
  • *****
  • Offline Offline
  • Posts: 4074
  • Respect: +2624
    • View Profile
Re: Card Strategy: Wharf
« Reply #24 on: September 13, 2011, 11:05:41 am »
0

Quote
I think Wharf helps money strategies compared to engine strategies more than smithy helps money strategies compared to its engine strategies.

That's quite easy to disagree with. The extra buy help engine strategies where you need many cheap cards for engine building. The extra cards drawn at the start of a new turn helps bring together engine card combinations. This is particularly true of king's courts and thrones such as when a king's court x wharf will draw you an 11 card hand next turn. From those 11 cards you can often construct the kings's court x king's court combination that will win the game (if king's court x wharf hasn't done that already).


Enlarged hand size improves engine based strategies, trashing strategies, and decision strategies far more than it improves treasure strategies. The difficulty is enlarging the hand size without using all your actions and the wharf can (partially) do that through the duration effect. The smithy will always consume actions when it improves the hand size which makes it ideal for a money strategy. There's a separate argument to say that the baseline wharf + treasure deck is so good that it's rarely improved, however even that's not true.

If you're not convinced, can you do this this smithies? http://councilroom.com/game?game_id=game-20110217-185017-8fb9815b.html
« Last Edit: September 13, 2011, 11:12:30 am by DG »
Logged
Pages: [1] 2  All
 

Page created in 0.062 seconds with 21 queries.