Haven, Wishing Well, Feast, Throne Room, Vault,
Great Hall, Coppersmith, Navigator, Laboratory, Grand Market
(Click for enlarged link at dominiondeck.com)
This is a 2-player game played between me and Blooki (aka Triceratops). The log is available here (spoiler alert!). Blooki/Triceratops is one of the top players on the leaderboard, primarily due to his creative and innovative decks.
Analyzing the Set
The dominant combo in this game is Vault->Grand Market. Grand Market is basically the best Action in the game, but is a huge pain to buy, especially in Province games. Vault, however, makes it trivial to buy Grand Markets by allowing you to discard your Coppers and use them too. In fact, you’re guaranteed a Grand Market every time you play Vault.
So given Grand Market, Vault enters the high echelon of “must-have-$5’s”, enough to make me open Feast. (Most of the other must-have $5’s are attacks: e.g., Mountebank).
The better question is what the $3 should be. On most boards, Silver is the obvious choice. But there are a couple considerations that push for Wishing Well here instead:
1) The average card value in my deck will be well over $2 by endgame, making Wishing Well more valuable than Silver even if it misses;
2) Wishing Wells cycle my deck more quickly, particularly in the early game where I am more likely to guess correctly, and therefore allow me to play the critical card in my deck (Vault) more often;
3) My deck’s money is primarily going to come from Grand Markets and Vaults, and in that case the Silver’s extra $1 is not really all that important compared to being able to play my Grand Markets and Vaults.
Accordingly, I open Feast/Wishing Well. Blooki, seeing the same Vault->Grand Market interaction, opens Feast/Silver.
Game Analysis
It’s very difficult to capture in text the feeling of playing a Dominion game. In annotating this particular game, I have decided to include all Actions played, but abbreviate their titles.
If you have any suggestions for the notation system, please note them in the comments; this is far from perfect, and I greatly appreciate any ideas as to how to efficiently display relevant game data (including game state).
| theory | Blooki | |
|---|---|---|
| 1 | $4 -> Feast | $4 -> Feast |
| 2 | $3 -> Wishing Well (reshuffle) | $3 -> Silver (reshuffle) |
| 3 | $3 -> Wishing Well | $5 -> Vault |
| 4 | Feast(Vault) $3 -> Wishing Well (reshuffle) |
$3 -> Silver (reshuffle) |
| 5 | WW x2 $5 -> Laboratory |
Feast(Laboratory) $5 -> Laboratory |
| 6 | WW (reshuffle) Vault $7 -> Grand Market |
Vault $6 -> Grand Market |
The opening didn’t go particularly well for me, as Blooki’s superior buying power nabs him an extra Lab. But the Wishing Well is already starting to cycle my deck a little faster.
| 7 | Lab; WW x3 (reshuffle) Vault $7 -> Grand Market (reshuffle) |
Lab x2; GMarket $7 -> Throne Room, Great Hall |
| 8 | GMarket $5 -> Laboratory |
Vault (reshuffle) $7 -> Grand Market |
| 9 | Lab; GM x2; WW x3 (reshuffle) Vault $13 -> Grand Market x2 (reshuffle) |
GH; GMarket $6 -> Laboratory (reshuffle) |
My Wishing Wells are starting to pay off. Had Blooki’s Silver been a Wishing Well instead, he would have drawn several more Action cards instead of being stuck at $6.
| Key Cards | 4 Grand Markets, 2 Laboratories, 1 Vault, 3 Wishing Wells | 2 Grand Markets, 3 Laboratories, 1 Vault, 1 Throne Room, 2 Silvers |
| 10 | GMarket x3; Lab x2; WW x3 Vault $14 -> Grand Market, Throne Room x2 (reshuffle) |
GH; Lab x2; Throne(GMarket) $14 -> Throne Room, Laboratory x2 |
This highlights how much those Silvers really hurt. Had Blooki’s Silvers been Wishing Wells, he would have gotten to his Vault like I did. Unsurprisingly, GMarket/Throne/Throne turns out to be superior to Throne/Lab/Lab. By the time he does get to his Vault next turn, it is too late.
| 11 | Lab x2; GMarket x4; WWx2; Throne(GMarket); Vault $18 -> Grand Market x3 (reshuffle) |
Vault $5 -> Laboratory (reshuffle) |
| 12 | Throne(GMarket); GMarket x7; Lab x2; WWx2; Vault $25 -> Province, Laboratory x2, Great Hall x2 (reshuffle) |
GH; Lab x3; Throne(GMarket); GMarket $15 -> Province, Duchy, Estate (reshuffle) |
Whenever you come out the better after an intense competition for Actions, you should look to see if you can end it on piles rather than VPs (especially if you are first player). Sometimes you can’t, because of a lack of +Buy, but more often than not there’s a way to safely end the game on piles rather than going for a more risky VP exhaustion. You want to look for non-terminals that essentially guarantee your deck can’t stall out while ending the game. Here, I empty out the Labs (which had already been somewhat depleted) and begin to hammer the Great Halls. Blooki, recognizing this, scrambles to grab as much VP as he can, hoping to stop me from ending it on piles by amassing a small lead. But this hurts his deck much more than it hurts mine.
| 13 | WW; GMarket x3; GH $10 -> Province, Haven |
Throne(Lab); GH; Lab x2; Vault; GMarket $11 -> Province, Great Hall |
| 14 | Lab x4; WW x2; GH x2; GMarket x4; Throne(GMarket) x2 Vault $26 -> Great Hall x4, Province, Estate x3 |
|
| Final Score | 30 (3 Provinces, 6 Estates, 6 Great Halls) | 21 (2 Provinces, 4 Estates, 1 Duchy, 2 Great Halls) |
Concluding Thoughts
I end up playing my Vault 7 times, while Blooki plays it only 4 times. This allows me to establish a massive Grand Market lead, which in the presence of non-terminals is nearly impossible to overcome because it can force a premature pile ending at will. Several times, Blooki’s Silvers get in the way; instead of allowing him to draw his Vault, they force him to settle for Labs or Throne Rooms instead. And except for a few early game purchases, they never actually boost his purchasing power by a significant amount. (See, e.g., turn 6, where Blooki’s Silver is about as useful as any other card.)
Dominion Strategy 
Impressive game – thx!
I wonder if the players ever thought of including a (throned) coppersmith – maybe late as DG suggested here:
Or was there simply no time? Maybe it was an option for Blooki at the point he had more labs but significantly less Grand Markets to catch up in money? But it would have hindered the cycling more – hm …
On a sidenote: Blooki/Triceratops, Dekker/Zeitgeist Movement and others. What’s the point in changing nicks? That will complicate matters for councilroom.com.
Adding the coppersmith is a flourish really and you might not have the right spend and buys to get it. In this case Theory was depleting piles and Blooki had too much silver. The guys did well to put these decks together on their initial look at the set, certainly better than I did at first glance.
Theory doesn’t really mention how he manages his buys on turn 12. If he’d leapt in with 3 provinces then his deck could have choked as he still only had two labs. By this stage the wishing wells are unreliable. Instead he put more drawing power into his deck increasing the chances of more mammoth hands. Only by retaining the big drawing hands can he control the three pile depletion in his favour.
If our friend with the simulator wants to check out an alternate strategy (probably simplifying the decisions rules on thrones in Theory’s deck) , I’d suggest navigator+treasure+labs. This won’t perform so badly once you add in some acceleration from the opponent’s vault.
Thx, DG. Can you explain the word “flourish” please (I’m not a native speaker, obviously). Is this in the sence of “icing on the cake”?
Yes, your understanding of “flourish” is correct. DG is saying that Coppersmith would be unnecessary and showy.
Throned coppersmith could be an interesting deck, but it seems directly at odds with vault/GM. I can’t really see it working at all here until very late in the game, at which point I think I’d rather throne a grand market anyway.
throning a Grand Market gives you 2$. Throning a Coppersmith gives you 7$. You need the extra buys so obviously you want to throne Grand Markets anyway.
Normally I never would change my nick (which was indeed Dekker) but I lost a bet and had to change it to The Zeitgeist Movement (which btw I do support).
The key decision is buying WW over Silver. (I would have done it wrong, but if you think of it: you won’t buy GMs without Vault anyway, so there is no point in buying silver)
Feast into Vault into GM was imho obvious.
I also agree that 5$ should buy Lab after the first Vault
(some people suggested buying mora than 1 vault)
The tricky part comes when you get 4$ early in the game (e.g. T3/T4)
It’s too early for Throne Rooms, so the question is: WW or Feast?
What action density do you need to justify buying Throne Room?
That question never occurred to you, since you always had 5$ or more.
Another question is: What to do with 2$ Do you really want a Haven here? In some cases its handsize reduction can really hurt you.
Well if you start with $5/$2 and you don’t buy a Haven then Vault (and the bought GM) will miss the second reshuffle if you get it in turn 4. If you do buy Haven then Vault (and GM) won’t miss the 2nd reshuffle if it appears in turn 3 or 4 and you don’t play Haven. But other than that you’re right; Haven doesn’t do much in this deck.
If the Vault is the card 1-10 in shuffle 2, you’d pull it in T3 or T4, no prob. So you don’t need the Haven.
Does Haven actually help?
If you draw Vault before Haven, playing Haven just gives you one card worth of deck cycling – which is exactly where you’d be if you didn’t buy the Haven!
If the Vault is the 11th card, and you Haven in T3, then Vault comes T4 – good. But with Vault as card 11, if the Haven is T4, then bad – you draw the Vault with the Haven, and have to Haven a card for -1 hand size, so no GM.
And that’s the crux of it – Vault with 6 cards is perfect. One extra card doesn’t help, and one less card kills. Just don’t buy it.
If you draw vault in turn 4 (card 6-10) playing vault will cause you to draw the 11th card, reshuffle, then draw card 1 of the new shuffle. This means at the end of the turn your Vault and GM won’t go back into the draw pile.
If you bought the haven but didn’t play it, your Vault will simply draw cards 11 and 12, and you’ll discard before the reshuffle.
I guess you could discard the haven with vault.
Nice to see the often untapped power of the Wishing Well.
I don’t really understand the power of WW. I rarely get WW, pearl diver, or great hall early because I feel like they don’t add much benefit. Sure they cycle through your deck, but you could cycle through your deck equally as fast if you just removed the card from the deck. (unless you guess right with WW)
I guess I understand that when there are no terminal actions in your deck, then it really can’t hurt you. I realize that those cards are useful, I just feel like I always talk myself out of seeing a use for them.
I had the same thought at first, but then just realized the WW does more than just replace itself, it always draws two cards, the automatic +1 card, and the wish card. Whether you guess right and put it in your hand or guess wrong and discard it, that second card is out of the way and does speed recycling.
“or guess wrong and discard it, that second card is out of the way and does speed recycling.”
No, you don’t discard it. You put it back to the draw pile. Also note the synergy between WW and Lookout mentioned elsewhere in this thread. It bases upon this.
“Also note the synergy between WW and Lookout mentioned elsewhere in this thread.”
FYI, that was in the preview thread.
Oops, thanks. I guess that inclines me again to share Dominicus’s doubts… 😉
It’s often possible to guess right with Wishing Well. Early on, more than half of your deck is Copper. Later, you’re often just buying a few cards over and over again (Grand Market, for example). It’s not rocket science to keep general track of what cards you have left in your deck (let’s see, I have four GM’s in play, probably I shouldn’t wish for Grand Market).
Even if you don’t want to try keep track of what’s in your draw deck, just wish for the card that’s somewhere in your deck and which makes your hand work. Like, if you need one more money to get a useful buy, wish for a Copper. If it turns out it’s an Estate instead, well, c’est la vie, you weren’t going to make your hand work even if you brought the Estate into your hand.
Even if you guess purely based on just “what have I bought,” it’s potentially valuable. You’ll be right some of the time.
And if you’re comparing it to “not buying anything at all,” which is what you’re implicitly comparing it to when you say, “your deck equally as fast if you just removed the card from the deck,” then even just “hitting” with WW once per game makes it a better buy than nothing at all (unless you’re in a situation where you have terminal card drawers, so you can potentially get WW and not be able to use it. The deck under discussion includes only one terminal card drawer, Vault, and then obviously you can discard the drawn WW for money).
If you keep track of your cards you can guess right pretty often.
Nobody seems to have mentioned yet that the wishing well can work in combination with the navigator in this deck. A Navigator/wishing well opening is competitive and as soon as your deck fills with drawing cards you can use the throne for an extra action, play the navigator, and sort the cards to suit your wishing wells.
Another hypothetical question: I just realized that there aren’t so may +1 card/+1 action cards for 3$. If you would add Menagerie to the setup – say instead of Navigator – would you, theory or anyone in his place, have ever bought it over Wishing well? I guess there are many situations where Menagerie is superior for cycling, but without trashing it might not be the case.
Theory bought WW on turn 2, 3 and 4. I would have considered the Menagerie no sooner than turn 4 or maybe even later instead of Great Hall.
As a variatian to this question: Replace in the setup WW with Menagerie. Would Menagerie be as critical to Silver as WW was? Theory’s WW guessed right 7 times (always for Copper).
Here on the other hand I think Menangerie/GreatHalls is still better than silver. The beginning of the game is over before it really starts by Feast->Vault, from this time on you want self-replacing cards over silver. Even if they don’t help you in the beginning, because you will leave the beginning-phase without silver quickly anyway.
Don’t think that Menangerie over WW is a good idea. In this setting I only see one possibility to get more than +1action/+1card out of it, and this is if you discard Coppers/estates before playing it (execpt of having good luck like copper/estate/men/lab/gm-hands which I wouldn’t rely my strategy on). But this would a) require getting some TR in play before for +actions and b) that you can’t discard what the Menangerie draws, so you better don’t draw copper/estates. Or Labs/GM/Menangeries that theirselves redraw copper/estates.
So we’re back to being lucky, which is also impossible in the beginning where being lucky with the WW is quite common. And more important. Like the Lab bought on turn 5 thanks to “luck” on WW, where the Menangerie would not have helped.
After some quick simulations with this setup I found these results:
Menagerie/Feast (45%) – Wishing Well(expert deck tracker)/Feast (52%)
Menagerie/Feast (51%) – Wishing Well(amateur deck tracker)/Feast (46%)
The reason Menagerie comes so close here is probably due to the opponent’s Vault helping to activate the Menagerie.
I did some simulations with this setup and let 3 players play against each other who all use the exact same strategy as theory (open Feast/Wishing Well, get Vault, get GM, etc…), but player 1 (theory) is a master deck tracker and will very often wish for the right card, player 2 (Geronimoo) can’t be bothered tracking his deck and will wish for a random card in his deck (which will still be the right card quite often) and finally there’s player 3 (Junkpit Von Lederhosen -> check him out on the leaderboard) who really doesn’t have a clue and will always wish for a Province. Here are the results of their clash:
theory wins 46% of the games
Geronimoo wins 34% of the games
Junkpit Von Lederhosen wins 11% of the games
Great game! Your deck really took off in the midgame.
One point you didn’t really seem to highlight is how you took advantage of being the first player. You mentioned that you wanted to end on piles, rather than risky VP exhaustion, but that plan really takes advantage of the fact that you had an extra turn than he did. Now, I’m not saying that you would’ve lost if you were second, but had Blooki had an extra turn, he probably would’ve picked up an extra Province. At that point he would have been at 27 points instead of 21, but the decision to end the game might not have been so straightforward. If you counted properly, (or used a point counter, but that’s a different discussion) and known at this point you had 17 points, then the outcome would have been the same. But had you not, it might have caused you to play differently, and have to go for risky VP exhaustion, as you said.
Now, had Blooki had a lucky draw, and was able to buy a Province and a Duchy, giving him 30, then ending the game would’ve resulted in your loss. So your best play would either be Province x3 + Estate, or Province x2 + GH x2 + Estate, or something similar. Whatever your choice is, there’s a (small) chance that Blooki might have been able to sneak up a win from behind, if he got a second lucky draw.
While I’m not trying to disparage your play, your article or your analysis, I just wanted to point out here that you used your first player advantage to turn what was a probable win, into a definite win. On a side note, I wouldn’t mind some articles on endgame play — when to take gambles, when to blow out your deck for VP, etc.
Oh, I totally agree that first player advantage is key in pile endings. I mentioned it in a parenthetical in the article.
I’ll look into writing endgame articles soon. It’s so difficult to synthesize general strategy articles! 🙂
Yeah, you did mention that didn’t you. I guess I need to brush up on my reading skills 🙂
In the early game, theory’s buying power is a vault and only a vault….
This is ideal!
The easiest way to explain WW is that drawing silvers dilutes vaults raw power…
In some powerhouse decks we quickly catapult beyond silver and would like to trash it… (we see this immediately on turn 4 when vault for GM race begins)
That aside, if blookie had throne’d his lab on turn 10, vault was in the next hand just before a reshuffle! Which, if it were drawn, would of resulted in a 5-4 split of grand markets with an extra lab for the silver deck.
This result also meant Vault wouldn’t miss the entire next draw (!) which was unfortunate worst case scenario for blooki.
How does 4cards compared to 2cards+4$? when would you think of doing such a gambit, throning Lab over GM?
Well if you’re average card is = 2$ then theres no difference other than cycling, and when you need the vault every turn, u prefer to cycle!
Obviously such a move is situational and not true in theory’s deck as he virtually always needs to throne GMs to make up the spread in buying power with zero silver to boast in the buy-phase.
On a side note, if card-by-card analysis is ever done on cornucopia cards, feast-vault is a perfect example of where hunting-party crushes lab.
Feast also enables the skipping of silver. You couldn’t pull this off without feast… the likelihood of $4/$4 and not getting a vault on turns 3,4 is basically a death knell.
Out of curiosity, is there some automated way you have of getting your games into this format, or do you manually type in the turns yourself? I’m looking at possibly writing some stuff up on the forums.
I think he always does it manually, but it could be reasonable feature for councilroom.com. export to theory style game log. I am sure it wouldn’t be that hard.
Yep, I do it manually. A little tedious, which is why you don’t see so many of them 😉
If you like, I can email you a template in HTML. But in general, I’d say, content trumps formatting. A good annotated game is going to be good to read no matter how it’s formatted, so long as it’s reasonable.
I need some annotated cornucopia! Please!
What about opening navigator/silver?
That leaves you a 10% chance of not hitting the 5 on any of the two turns, and leaves you with a useless silver later on, but the navigator cycling potentially 5 cards every now and then might give you another few extra vault turns?
Also, I was wondering whether an article on how the second player strategy might differ from the first one might be intereting. Beyond countering, I was wondering if it was in the second player’s interest to go for a higher variance strategy, given that the first player seems to statistically have a significant edge over him, so will always win when strategies are deterministic. Bootstrapping that analysis seems to indicate that, for a same mean, the second player would go for higher variance. Is that really true?