Dominion Strategy Forum

Please login or register.

Login with username, password and session length

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - JThorne

Filter to certain boards:

Pages: 1 ... 3 4 [5] 6 7 ... 12
101
Dominion General Discussion / Trashing vs. Buying
« on: June 05, 2017, 10:58:19 am »
I've been mulling this one over, waiting, but what the heck, I'll do it. Someone said "start a new thread" so I'm starting a new thread.

I was shocked to see that anyone would even bother to argue this, but to review: You open Chapel/Silver.

Question: With a hand of Chapel-Silver-Copper-Copper-Copper, do you trash three coppers or buy a key power-$5?

Answer: You trash three coppers.

Trash down/build up is such a fundamental principle it's hard to imagine why there would even be an argument. You're going to have at least another $3 hand before your next shuffle, and if you want to be sure to hit $5 you can always buy another Silver. Your odds of colliding two Silvers increases dramatically with three fewer coppers in your deck.

I tried to imagine any situation in which it would be beneficial to buy the $5, and I had a really hard time doing so. If you're worried about losing the split, well, you'll probably lose it anyway to the player who builds correctly and can buy $5s reliably every turn. I've seen many people make this mistake, and inevitably, they build up their drawing capability while their deck is still half-full of junk and then start complaining about how they never draw their Chapel with their junk cards, or draw their Chapel dead.

It's even worse with Steward. The number of times I've seen Steward used for money early in a Steward, Copper, Copper, Copper, Estate hand because they wanted that power-$5 so badly is astounding. Or even Steward-4C being used to buy Gold. (Ick. Yellow cards.) The beauty of Steward is that you can trash Coppers even more aggressively than with Chapel because it can switch to giving economy for the build phase.

There's nothing like the audible "click" you hear when an engine snaps into place and starts humming because you've draw a tiny deck into your hand. The world is your oyster. You've made your own shuffle luck. Anyone who has bought a Donate on turn three or four and trashed down to five cards or less knows what I'm saying. Did you really need a power-$5 first, or are you just going to start getting them by the handful starting now?

It would be nice to have some clarity about this, if nothing else for the benefit of newer readers.

102
Dominion General Discussion / Re: Playing Vassal
« on: June 05, 2017, 10:18:36 am »
Quote
I mean, using Vassal as a ghetto +2 Coin to make a double Tac engine work.

Trouble is, Vassal wants actions in your deck, not in your hand. Seems like an anti-synergy unless there's a ton of +actions, and even then, a hand of 10 cards, half of which need to be +actions to make the other half work seems less effective than a hand of 5 cards where Vassals can just play more Vassals off the top of the deck without +actions.

My favorite Vassal engine so far has been Herald/Vassal: The Forrest Gump deck.

103
All this talk of piling out Villas makes me want to take a moment to offer a special message to all of the non-drawing Villages out there. Villa, Shanty Town, Festival, Squire, sometimes even Nobles. Sometimes you guys are great. Sometimes you're the only source of +Action and we have to live with you. Other times, you're just Necropolises (Necropoli?) with delusions of grandeur, offering us a +buy and maybe some coin in a desperate attempt to make us like you. (Fishing Village and Native Village, you guys get a pass.)

104
Quote
Because occasionally, you just want to empty a pile in one turn to prove you can.

Did that with Highway/Forum once. Highway will take its +buys where it can get them.


105
Rules Questions / Re: Gaining, with nothing to Gain
« on: June 01, 2017, 08:18:50 pm »
Ah, those three little words... "if you do"

They make all the difference. I can't tell you how many times people at the IRL table have read a card and said "why is this worded so strangely?" I end up explaining all of the lovely broken things you could do if the card/event was missing those three little words, or if the instructions were in a different order, or if some of the text was above the line or below the line.

Kudos to Donald X for continuing to get better at anticipating when there will be rules issues or exploits and heading them off in the wordings and/or behaviors of newer cards.


106
Landmarks and Events are not cards. They are, officially, "card-shaped things" according to designer Donald X.

I wonder what the largest possible kingdom would be in terms of square inches of table space required? (Mats! Mats everywhere!)


107
Opponent tried to make a Mill/Watchtower disappearing money deck in a kingdom with no extra actions. Kept getting angry when drawing six cards with Watchtower and having most of them be dead action cards. Got a couple of Provinces anyway...but only when their hand started with the Necropolis in it.

Also, same game, same opponent ignored Cultist in a Wolf Den game. So I guess it doesn't really matter what they played.


108
Quote
Roughly speaking, out of 100 million possible boards, less than 1 has been played on average. You'd need to put some serious work in to make a neural network work under these circumstances.

But a neural network doesn't need to see every possible combination. That's the whole point. How many images of faces does an NN need to be fed before it develops the patterns necessary for facial recognition, where it can pick a face or multiple faces out of an image? Likewise, How many games of Dominion do you think it would need to be fed in order to look at a kingdom and be able to simply say "engine" or "BM" or "slog" reliably? I'm betting it would be pretty accurate given the database that exists right now.

Using similar technology to tune playstyles could also be done with targeted data analysis, but it would be a fascinating experiment just to start with the above and see what you get.

109
I had another thought about this recently as I was helping someone with a development project.

I wonder how much help a neural network would be in devising optimal buying/playing strategies? Neural networks work best when they can be fed a vast amount of data, but I started wondering: How many full games are now stored in the servers? Does the server have all of the old Isotropic games, Making Fun games, and the new server?

A neural network works by being fed lots of data and eventually building up knowledge that's useful for pattern recognition. If you feed it the game state for every single turn of every single game ever played online, with the goal of avoiding things that lead to losses and doing things that lead to wins, I wonder if there's enough data for an NN to learn from that?

Part of the reason that I thought about that was that when developing a genetic algorithm, you're still pre-programming it with certain strategies and playing bot generations against each other. It will essentially only learn to beat itself, and it will only be as good as the strategies and play styles that you anticipate using.

That said, even a NN might need a little help; some sort of "starter" that looked at a kingdom and tried to recognize "engine" or "BM" or "slog" or "rush" just from the kingdom contents, then looked at what the players actually played and what the final score was. An NN should, at the very least, be able to learn to recognize that from existing data. It would also learn to recognize certain patterns; cards that synergize or anti-synergize, based purely on buying patterns of winners and losers, or even just buying patterns in general.

The one thing I wonder is whether there's enough data to defeat the pairing system; beginners play beginners, advanced players play advanced players, meaning that terrible play can still win the low-ranked games and great play will still lose the high-ranked games. With enough data, though, that gets overwhelmed.

Of course, it may be that only the developers have access to all that data. If so, I wonder if they've played around with that concept.

110
Quote
I feel like this is like Forum. How about a card which doesn't cost a buy?

I mean the other way around. A card so powerful that it also had -1 buy to nerf it. For example, Royal Blacksmith makes you discard coppers. Imagine a similar card which said +5 Cards, -1 Buy. If you played that as your only action you might have $10 coin, but...no buys! You would have to put it together with cards generating +1 Buy or you wouldn't be able to buy anything. Or if you played two Candlestick makers, then the Unicorn Blacksmith, you would have 2 buys, not 3.

At first I wondered if that would be too confusing, but basic subtraction is still pretty simple compared to the math mechanisms that already exist, such as doubling, or "half, rounded down."

111
Well, now I'm more confused than ever.

Opening Remake/Silver and thinning your deck dramatically while also buying Bakers so that you can ultimately play 4-5 bakers a turn and a maybe one or two treasures sure sounds a lot more like a single-Province engine than BM to me.

Or is the distinction that if you don't play your whole deck every turn, it's not an engine, period?

Or is the distinction that an engine builds exponentially, seeks to gain pile control and end the game when it's ahead, whereas a BM deck greens early and continuously, forgets about pile control and the game simply ends usually whenever the Provinces are gone?

Maybe I'm letting Awaclus get to me with his "there's nothing between" statement. Where's the line in the sand that separates BM and engine?

112
I was just teaching Dominion IRL to a few newbies. I've hooked a great many gamers on it.

I was explaining how some cards, such as Village and Crossroads, add two or more actions to the action pool, allowing you to play more action cards that turn. One of them asked an interesting question:

"Are there any action cards that consume more than one action from the pool?"

No, but...wow. How powerful would an action card have to be to require spending two actions on it? What would be worth crowding your terminal space that much?

Anyway, just thinking about things that are relatively simple and haven't been done yet. A card like that would say, what, -1 Action? Maybe it could be played with only 1 action available, but would consume two if relevant, meaning that your counter would have to be at 3 if you wanted to continue your turn.

It does raise the question of negative stuff on the cards. There's already -1 Card in the form of cards like Courtyard. -1 Buy? Ouch. -1 Coin? Yes, there's a token, and debt, but no action card.


113
Dominion General Discussion / Re: Help with Watchtower?
« on: May 10, 2017, 08:21:32 pm »
One more edge-case: topdeck tricks. Topdecking buys or gains under normal circumstances is so-so. Doing it when you still have actions and can play stuff like Mystic or Herald/Magpie if you can topteck two gains during your action phase can be interesting. Also, if you're good at deck-tracking, consider using the topdeck ability to line up key cards together, such as two Urchins, if you know what's coming up in your next hand.

But again, it comes down to BM vs. Engine. It's not a BM card, and if you're drawing your deck, topdecking a gain is identical to a regular gain if your discard and deck are empty.

114
Dominion General Discussion / Re: Help with Watchtower?
« on: May 10, 2017, 08:15:15 pm »
It may be more useful to look at when it's good.

It's good against junking, whether by the opponent or self-junking-for-benefit cards (watch out for the dreaded "if you do" wording, though...) and against handsize reduction attacks.

But it's draw effect as the only positive has to be used carefully. It only works with disappearing money engines. In BM, it's no better than Moat, and Moat/BM isn't a thing. Even in an engine, it's not enough that you just have "non-drawing Villages." Villa/Watchtower or Shanty Town/Watchtower may be terrible unless you can drop most of the cards that you've drawn on the table before you play your next Watchtower. Villa/Watchtower for 3 cards/Villa/Watchtower for 2 cards/Villa/Watchtower for 1 card? Not good. You've just played 6 cards for $3 and +1 card. You need non-terminal cards that reduce handsize for benefit; even cheap ones work (Pawn for +action/+coin, Oasis, Poacher, Warehouse)

I've skipped it on plenty of kingdoms where there was no junking, or if the game was a clear BM or deck-drawing engine. You feel pretty silly drawing a Watchtower into a 10-card hand.


115
I have to agree that Baker is a weak peddler variant. If you spend the coin this turn, it's a Peddler variant that does nothing, which the community has agreed is worth $4. If you don't spend the coin this turn, it's a do-nothing cantrip, like a Pawn with no choices or a Pearl Diver that doesn't peek; the community has proposed that an actual do-nothing cantrip would likely be priced at $1. So Baker is a $5 that can either act like a $4 or a $1.

The fact that it gives coin tokens instead of coins is more of an edge case. The thing about coin tokens is that they're only valuable when they smooth out your buys, allowing you to hit $5 and $5 instead of $4 and $6, or possibly even $8 and $16 instead of $12 and $12 if you really want to stretch.

Smoothing out buys over multiple turns is mostly useful in BM when you're trying to hit $8 or $11. With a deck-drawing engine, the difference between coin tokens and coin is almost irrelevant, other than a couple of engine-building buys early. But cantrip-money in general is weak in BM. So Baker's got a built-in anti-synergy.

Sure, there's plenty of "depends on the kingdom" cases, such as $5-gainers, Butcher, etc., and possibly slogs, though I'm not convinced that saving up tokens over multiple turns for VP cards is an efficient use of resources, even in a slog.

Compare it with Artificer: Even playing a single Artificer, discard 3, play two coppers allows you to gain two $3s. Two Artificers and Copper+Silver allows a gain $3, buy $5 or gain $4 buy $4 (kinda Baker-ish smoothing, no? but two gains per play!) If an engine is feasible, Artificer is much more powerful, not to mention that it happily discards green cards for benefit.




116
Dominion General Discussion / Re: Transmogrify Rush
« on: May 09, 2017, 07:49:00 pm »
Actually, it's all over the place:

http://forum.dominionstrategy.com/index.php?topic=17114.msg685194#msg685194

..."it's hard to play against an opponent that could empty 3 Provinces in a single turn by calling 2 Transmogrifies."


http://forum.dominionstrategy.com/index.php?topic=16863.msg671968#msg671968

..."Transmogrify: kick-starting your engine, milling Provinces"


http://forum.dominionstrategy.com/index.php?topic=17138.msg690229#msg690229

..."Mat the T-Mogs. Start milling Province"

more:

..."In my experience, milling Provinces with Transmogrify is definitely a thing"

Maybe you didn't search for "mill" or "milling"? What, you've never played Magic? Weird how the term "milling" means "to put a card directly from the supply/deck into the trash/discard without ever drawing it" because of the Millstone magic card that has outlived its usefulness, but the term lives on.


117
I hate to be that guy, but I'm not even that guy. I'm not a ranked on-line player, just an avid student of the game who probably enjoys beating the snot out of people IRL more than he should, based on reading literally every page of the wiki followed by almost every thread in this forum, digging for deep analysis and wisdom about the game.

First, you need to take every thing on your list that ends with "per turn" list and change it to "per shuffle." (per Stef.) The Shuffle is the defining boundary in decision-making (partly because your deck doesn't get better or worse when you buy cards. It only changes when you shuffle it.)

Next, you need to take the "per shuffle" thinking process and even expand THAT to "per game."

The reality is that BM is usually the wrong play, and is primarily used as a yardstick against which other strategies are measured in simulators when testing out combos. An evolutionary algorithm for the modern game is simply going to have to know how to build and play an engine from start to finish.

Engines are almost always built with a combination of trashing and gaining; trashing down, then building up. I don't see the word "payload" anywhere in your list of considerations. I don't see any reference to cost reduction. I think with the "quick ending potential" you're referring to "pile control." I don't see any reference to tempo. Or opportunity cost. Or OVER-drawing your deck. There's a lot of already-established terminology you should probably incorporate somewhere along the line.

I'd say you should perhaps start by just trying experiments with the base set, but even that is far, far more complex than you're probably imagining.

Check out this thread:

http://forum.dominionstrategy.com/index.php?topic=17171.0

Watch the video for a clinic on base set engine-building and listen to the commentary (if you can get past the music) and try to imagine how you would build those sorts of considerations into an algorithm. How is your algorithm going to figure out that it's preferable to open two terminals, and which one to play if they collide? How is it going to figure out when it makes sense to buy the single Bandit, which is emphatically not an early-game card? Is it going to be sophisticated enough to do even simple tricks like gaining cards (like Bandit Golds) and drawing them on the same turn by ordering the play of actions correctly? Is it going to figure out when to stop playing Bandit altogether so that you're not gaining Golds that will gum up the engine?

Then, once you've pondered that and have some free time, read through the neat interactions thread. Will an evolutionary algorithm "see" any of those? Why or why not? AI has to focus on playing strategies as well as buying strategies. The cans of worms just keep opening.


118
Dominion General Discussion / Re: Peddler-type cards, part 2
« on: May 06, 2017, 11:54:36 am »
Is there any math-y treatment of Peddler variants talking about statistics and economy and engine-building?

Sometimes a kingdom has no draw other than peddler variants, but anyone that's ever played with something like Forager and Peddler knows that it's possible to end up with a deck with nothing but Peddlers in it pretty quickly.

An all-Peddler deck is super-consistent...until the deck has 5 stop cards in it, at which point, it's essentially guaranteed to stall, but only after playing a few cards (how many, on average?) and finally drawing all 5 stops.

So: What's the formula for average cash-per-hand based on a five-card hand, the total number of stop cards in the deck and the total number of Peddlers? That might offer some guidance about when to start greening, how many Provinces/Duchies you'll likely get, especially if one of the stop card is a terminal +buy.

119
Dominion General Discussion / Re: Most brutal possible game?
« on: May 03, 2017, 11:03:24 am »
Wait for it... Awaclus is going to post a FTFY that crosses out your entire guess other than the 6P. And he may add 3-5 for good measure.

Anyway, here's the thing: most people characterize games with attacks as "brutal" because the attacks can be so effective, particularly in multi-player and particularly junkers since the curse and ruins piles are +10 per extra player. But some junkers aren't so bad. Torturer gives the curse to hand so players can take the curse if they have a trasher in hand, and once curses run out, they do nothing, and they also guarantee the kingdom has a big draw card. Mountebank is worse. Also Cultist.

But even a game with both curses and ruins, if the junkers are good ones, isn't going to last long even in multi-player, because that's two piles down. It may even end up being an Estate rush and end pretty quickly. Players in my IRL playgroup often insist on trying to build to Provinces even in a slog kingdom, and that can be an easy win on piles.

There aren't many great trashing attacks, though again in multiplayer they can be bad, but most of them have cost ranges and can be bought around if necessary. Discard attacks work nicely, but do nothing in multiples.

The Magpie/Scrying Pool suggestion is pretty darn good. Cards with fiddly mechanics or lots of decision-making can be worse than attacks in terms of slowing down games. I'll add a few:

Plenty of actions, Oracle as the only draw.
Wolf Den + Swindler + Ruins + lots of $3/$4 cards and a Trash-one. Good luck remembering what you have only one copy of. Or what you've given opponents only one copy of.
Mountebank + Wall + no trashing.

And what's the worst attack to be on the receiving end of?

Other than a pin like one of the Masquerade pins or the rare Bureaucrat pin, how about one of the various Possession pins? Possession + Haunted Woods allows your opponent to just give you a junk hand every turn.

KC/Village/Torturer is starting to sound pretty good, isn't it? Heck, that's probably even a nice engine board that's fun to play.

Looks like you're new. What you'll discover with further reading and experience is that Dominion is a far more complex and interesting game than you ever imagined, and that games can get both a lot worse and a lot better than you've even seen yet. You haven't seen bad until you've won a game with a negative score, or seen good until you've piledrived (piledriven?) all the Provinces or Colonies in one turn.

120
One thing I can recommend is to play a few hundred games against Provincial yourself just to get a feel for how well that program works.

What's interesting is that there are some BM kingdoms and slogs where it's hard to beat the AI consistently because that really is the best strategy, and the AI counts and deck tracks perfectly, so the result is almost a coin flip. In fact, some of them the AI wins more often if there's no really good complex strategy.

However, there a LOT of kingdoms where proper play will repeatedly destroy the AI.

It can't play a Gardens rush at all, and there are a number of Ironworks/Gardens kingdoms.

It can't play a megaturn. It'll settle for double Provinces, or even triples, not "getting" that you can easily build to a turn where you buy six or seven Provinces, or even three Provinces and five or more Duchies to overwhelm the rest of the Provinces.

It can't play Goons or any other VP-chip strategy, and you can beat it by hundreds of points, because it never tries building an engine that buys and trashes copper or curses for handfuls of chips at a time. Heck, you can even beat it with a KC/Monument starvation deck (my nickname for a deck which, if both players play it, results in an infinite loop of players gaining points and no cards. The official ruling from Donald X is that the loser is whoever starves to death first.)

And that program was written well before the game got REALLY complex with Dark Ages/Adventures/Empires. Not to mention all of the incredible subtle issues that players have enumerated since then that have appeared in the forum, such as thinking of progress in terms of shuffles instead of turns, tracking issues like cards missing the shuffle, or triggering a reshuffle at the wrong time. So many cards now reward careful play as well as buying strategy.

I think you have a much, much longer road ahead of you than you imagine. But it's certainly an interesting one!

121
Game Reports / Slog engine
« on: April 26, 2017, 09:15:08 am »
I'm mostly creating this post to annoy Awaclus with the subject line.

Kidding. Actually, it's partly because I need to get better at slogs. In my IRL playgroup, I feel like everyone almost always plays an engine, even when it's probably wrong. I can usually look at a kingdom and know immediately there's just no way anyone's going to draw deck before piles run out, or get to doubles before singles drain the pile. I also roll my eyes at players settling for singles when the kingdom clearly supports doubles/triples or even a megaturn.

So...Sea Hag. 3P, 20 curses. I opened SH and Silver. Festival/Oracle sure didn't sound enginey to me, but Herald? And City Quarter? Overlooked cards in my group.

As soon as I started hitting $5, I repeatedly overpayed for Herald, topdecking my one copy of Sea Hag. The only trashing was Raze. Picked up a Festival at some point for the +buy to get Pearl Divers in handfuls to make Heralds/City Quarter better.

The other two players struggled to do anything. One tried many Razes. One bought two copies of Sea Hag, but still lost the curse split badly. So many curses!

I picked up a couple of Pearl Divers on dud turns, and one City Quarter, and a late gold.

It seems ridiculous to call that an engine, because the whole build was predicated on playing key actions reliably without necessarily drawing deck and without sifting. I played SH far more often than opponents, and got to my razes far more often, killing the few curses I had (and the dead Sea Hag) but ended the game with a final turn in which I did, in fact, draw deck, and with $10 and two buys bought the only Province in the game and the last Pearl Diver, for something like a 6/-3/-2 victory. Silly. (Curse/Raze/Pearl Diver.)

And probably wrong. This is why I say I'm bad at slogs. I feel like I would have been killed by experienced players who know to grab lots of cheap VP (even Duchies and Estates) knowing that there's no way anyone's going to hit $8 before piles run out.

On the other hand, my build worked exactly as I planned, and feels like it may have out-performed a points slog in a 2P game because those are some nice components that could easily have bought a few Provinces before ending the game.

122
Dominion Articles / Re: Refreshing the Dominion Paradigms
« on: April 23, 2017, 12:22:49 pm »
Quote
If your game is as messy as JThorne described, you are misplaying it.

This may be a shocker, but I actually agree here, with a small caveat.

If I'm interpreting correctly, this statement is basically saying this: If you're buying village/smithy pairs, and a terminal coin/+buy card, then you're playing an engine. If you're only buying one Province a turn, then you are unequivocally doing it wrong. Either you should keep building until you can reliably draw deck and get to $16, OR you should NOT have been trying to draw deck in the first place and should just play a BM game. In BM, Villages are almost always do-nothing cards that might as well be Pearl Divers without the topdeck.

If you're playing a BM opponent and they start buying Provinces first, if you get nervous and stop building and start singling Provinces, you're also doing it wrong and you have lost. If you stick it out and build for doubles and you still lose, then you chose the wrong strategy for the kingdom in the beginning. I totally get it. I've been both on the winning and losing side of this situation. Mostly winning.

That's not to say I haven't played a single-province engine. In a kingdom with no buys and no gains, and especially with decent trashing, there are times when money and sifting can create an incredibly reliable engine that's more reliable than BM even if it's just singling, but that should be the plan from the beginning, not something you settle for when you realize you made a mistake.

Anyway, the caveat, and it's going to make Awaclus roll his eyes.

Multi-player.

The best laid plans can go radically awry depending on what other players do, or if just one of a few opponents gets a lucky draw and grabs too many of a key card early. You may have to bail halfway through engine-building. In my experience, the most effective way to victory in multi-player is one of two things: 1. Figure out if there's an engine that can run on no more than 5 copies of any one card. 2. Pile control.

4-player games almost always pile out, but I've certainly played deck-drawing engines that started doubling before piles emptied, forcing other players to play catch up (Thanks Donate! You're the best!) 3-player games give you more ingredients, and if one player ignores a key card, you can play that strategy like a 2-player game.

But you have to keep your mind open to many possibilities. I just won a game where I had one opponent playing BM, one playing a Hireling engine (no, no, no a thousand times, no) and I won what basically amounted to an extremely weird rush. Procession/Lurker, with both Museum and Obelisk in play, and Lurker was the Obelisk card. You can guess how that went. One opponent was Processing Lurkers in order gain Hirelings. I was Processing Lurkers...in order to gain their lurkers from the trash. (And processing into $3 Chariot races, Processing them into more Processions, then into one of each $5...)

In a kingdom with no decent draw, no trashing, and no +buy, I emptied piles with a final massive Lurker party, bought the last lurker, and landmarked my way to a victory over opponents with several provinces each (I bought one for the 8VP, because that's hard to pass up. That and the 5VP Duchy.)

Also, another weird game: No +buy, no +actions, lots of terminals. I don't remember the rest of the Kingdom, but here's what I did: Banquet Catacombs, Transmogrify. Silver, Banquet two more Catacombs. On the first Catacombs play, buy a Gold. On the second, Donate all starting cards and take only one debt. Next, Transmogrify the Catacombs into a Gold and another T-Mog, buy a Province. Mat the T-Mogs. Start milling Province/buying Province. At some point I think I did a T-Mog a Catacombs into Gold/Remodel late for a Remodel Gold to Province ending. It was ridiculous fast. Game was over in close to ten turns, and I ended up with six Provinces (out of 12.)

So, what's that? I only bought one Province at a time. It's definitely not an engine, because I never actually drew deck, but I could sure line up $8 with Catacombs pretty easily in a deck with that few cards. From my perspective, that seems like a BM strategy, but it certainly underscores the fact that BM isn't simply "buy money and a couple of terminal draw cards and then points."

123
Transmogrify's big trick is the "to hand" part. When it's good, it's because it allows you to collide elements exactly when necessary instead of buying the parts and praying. Sure, drawing deck is better, but still.

Open Transmogrify/Urchin. Mat the Transmogrify. When you draw Urchin/Estate, Transmogrify the Estate to get your second urchin in hand and enjoy your early Mercenary.

Playing BM and stalling late? Transmogrify that Silver to a Remodel in hand and remodel that Gold to a Province. Beats the heck out of buying a Remodel early and drawing it dead several times, or settling for a Duchy.

Also, the "to hand" trick makes milling insanely fast since one Province in hand allows you to mill as many Provinces as you have Transmogrifies on the mat.

124
4P game. Started building a modest duration engine: Two lookouts, two Walled Villages, two Haunted Woods, two Squires, two Golds. Other players started following a similar path, trashing their Estates and emptying the WV/HW piles. Piled out Squires for a 3/2/0/-9 win.

Wolf Den FTW!

125
Let's Discuss ... / Re: let's discuss guilds cards: baker
« on: April 14, 2017, 01:18:46 pm »
Gold is often bad. The difference between hitting $6 this turn and $4 the next turn vs. being able to buy two power-$5 cards can be significant in the building phase. That's where I find coin tokens to be the most useful, but it also underscores the fact that several Bakers may be overkill.

Baker also serves an important role when I'm explaining Dominion strategy in that you should not underestimate the importance of +card. Baker is $5, Candlestick maker is $2. That tiny +1 card is worth $3 more, and even then, CM needed a +buy boost to make it relevant. Then I point to the treasure piles and ask: And what are those poisonous yellow cards missing? I swear I'm going to start keeping a flyswatter under the table for when people keep reaching for the Gold pile just because they've hit $6.

Pages: 1 ... 3 4 [5] 6 7 ... 12

Page created in 0.094 seconds with 18 queries.