Disclaimer: Dominion does a really great job of balancing its Kingdom cards. Pretty much every card has some situations where it shines, and some situations where it doesn’t. Nevertheless, some cards just end up being flat-out better than others, either because they are more useful more often, or just ridiculously good when they are useful. Don’t expect this list to be very scientific.
The biggest problem I had with this list was that any honest list of the best $5’s would basically just be attacks. As great as Laboratory is, it’s simply not a viable defense against Mountebank or Torturer or Minion. So I’ve decided to split this list into two parts: the best $5 attacks and the best $5 non-attacks. If you want, consider this list to be the “real” best $5 cards, since any of the top five are going to be more important than any other option at $5.
This image could have taken the place of this entire post.
Dominion: Prosperity
Honorable Mention: Rabble
It’s kind of silly to have an Honorable Mention, since there’s only seven $5 attacks, and one was already the worst $5 card, and in any case (by process of elimination), it’s obvious what belongs here. But since I’ve gone to the effort of writing up all the other cards, I might as well give Rabble its moment in the sun. It’s only really effective when your opponent has Victory cards, but it can be unbelievably devastating once chained together in the late game. Games with competing Rabble engines are a lot of fun, since you’re both locked into a game of chicken, daring the other into buying the Victory cards. It can shut down a draw chain without Cellar or Warehouse, or it can absolutely ruin your opponent’s next turn.
That having been said, the other $5 attacks are in a class by themselves. Rabble isn’t quite as bad as Saboteur, but it’s got a long way to go before it can join the elite pantheon of attacks. No other attack is as weak early on: in fact, this is one I’d almost certainly pass over for a peaceful $5 early on.
Dominion: Intrigue
5. Minion
It’s amazing to me that Minion can be #5 on this list, since passing it up in most 2-player games is suicide. But it’s less flexible than the other attacks: it’s not that effective for Colony, and it’s not great in multiplayer since it’s harder to get that critical mass of Minions. Plus, they’re quite vulnerable to getting bogged down with Curses, such that most of the Minions are used for drawing instead of the money.
But if you can do better than a 5-5 split in 2-player, you are usually well on your way to victory. Its non-attack function is very strong: quite a few cards combo well with a Minion deck, and because it self-synergizes, it’s usually easy (even without trashing) to get Minions going. But its attack really becomes killer once you get into Province-buying territory, since you are forcing your opponent to cycle faster, meaning his green cards are being mixed faster into his deck and he has fewer turns to run his fully-optimized engine.
Dominion: Intrigue
4. Torturer
This is probably the best-named card in the game. I’ve seen people quit Dominion over Torturer. And not in the completely unnecessary sense like with Saboteur, I mean in the completely justified sense of “because you got your Torturer chain off first, I am now doomed to be Torturer-pinned into oblivion for the rest of the game”. There’s an art to building an effective Torturer train as quickly as possible, but the first person to land the multiple-Torturer blow is usually the one that ends up prevailing. As much as I hate reactions in 2-player games, Torturer is one of those few situations where I give in and grab Lighthouses or Watchtowers to defend.
Unfortunately, Torturer is a total whiff if there aren’t any +Actions, since a single Torturer by itself is not very scary at all. And naturally, it works absolutely terribly with Curse-givers, since it only has bite if there are still Curses left to take. This also means the attack has a built-in expiration date: make it past the 10 (or however many) Curses and the attack becomes a total non-factor. Plus, gaining Curses into hand is usually not nearly as big a deal as gaining them into deck if you can just trash it immediately.
Dominion: Seaside
3. Ghost Ship
Ghost Ship draws only 2 cards, compared to Torturer and Rabble. Plus it has the unfortunate distinction of being the only $5 attack that doesn’t really work well Throned or Kinged. So you won’t be building a good engine with Ghost Ship any time soon. The good news is you won’t have to, because Ghost Ship is about the meanest single attack there is in the game.
Oh, yes, Mountebank and Witch (you know they’re coming up) are stronger, and there’s nothing meaner than a Torturer or Minion chain, but as single attacks go, it’s hard to fathom one crueler than Ghost Ship. With relatively few exceptions, you’re basically ruining either your current hand or your next. Worse, Ghost Ship is one of the two attacks in the game that does not anti-synergize with most other attacks; if anything, it gets even better.
Once in a while, Ghost Ship attacks are helpful as mini-Havens. If you’re lucky enough to draw 5 Platinums in one hand, you’d be pretty thankful to Ghost Ship for letting you save two for next turn. But most of the time you have no better option than to put off your good cards until next turn, hoping he doesn’t draw another Ghost Ship. Getting pinned by Torturer chains is painful and frustrating, but getting trapped in Ghost Ship lockdown is as depressing as Dominion can get.
Dominion
2. Witch
Witch is a defining card in Dominion: along with Chapel, Thief, and Gardens, Donald X. called it one of the four pillars of the base game. Even beginners quickly realize why: Curses are the strongest attack in the game, and Witch is as straightforward a Cursing attack as it gets. There’s no condition for gaining, like Mountebank; no option for the player to choose, like Torturer; no top-decking to let them trash it immediately, like Sea Hag: just “gain a Curse”. This unconditional Cursing makes it hands-down superior to Mountebank in games when both are present, since the Witch’s Curses will never be stopped without a Reaction. Indeed, if all five of these attacks were on the same board, Witch would be my first choice: it is the only one that will actually shut down all of the other attacks. (Except maybe Ghost Ship.)
Dominion: Prosperity
1. Mountebank
So why does Mountebank get the nod over Witch? Three reasons, in ascending order:
- It continues attacking even after the Curses are exhausted. Realistically, this is not a huge deal, since after the Curses run out there are not that many turns left before the game ends. But it’s nice to be able to do some damage to your opponent’s deck.
- It gives +$2 instead of +2 Cards. Although there are certain benefits to Witch’s +2 Cards (for instance, faster deck cycling), it is much easier to draw two Witches dead than two Mountebanks. And as you will often want at least two of these Cursing attacks, drawing one of them dead can spell the difference between splitting the Curses 5-5 and 6-4.
- It’s just flat-out meaner. Mountebank is the most devastating Turn 1 opening; it has the highest “win rate with” and the fourth-lowest “win rate without” out of any card in the game; it; an early Throned or Kinged Mountebank is basically game over. And although it’s theoretically counterable by Moneylender/Counting House/other Copper-dependent cards, the extra Coppers slow your opponents down from getting back around to their own Mountebank.
Dominion Strategy 
Hmm…if the average value of a card is better than 1, then I’d expect +2 cards to be better than +2 money. Perhaps with curses flying about it’s harder to get that protasis?
I would say +3cards >> +$2 > +2cards, as a general rule. I’ll get to Witch vs. Mountebank specifics a bit later, since there are extra wrinkles there beyond the relative value of the effects.
1. +cards effects can draw dead action cards when you have no actions left. In and of itself this pushes down the average card value if you have other action cards, but even worse, you’ve now lost the use of that action card the next hand.
2. You can ameliorate the dead action card issue by having extra actions, but that means you have to buy +2actions cards and then get them together with +cards effects. And indeed, draw engines built this way are often good strategy. But simply put, +2cards effects are almost never enough to build a a reliable draw engine. +3cards is twice as many net cards (2 net instead of 1 net), and +3cards effects are basically the minimum standard for building a draw engine. Ghost Ship is an excellent “kicker” for your draw engine (particularly if it involves Council Room), but you usually can’t build a draw engine solely around Ghost Ship, Moat, Witch, Courtyard, etc.
3. Similarly, +2cards is helpful in a big money deck, but only half as helpful as +3cards. Would you rather have a single +2cards action in a big money deck than not have it? Sure. If you could trade your +$2 action for a +2cards action late in the game, it’d probably be a good move. But the +$2 is generally going to be much more useful early in the game before your average card value gets high enough where 2 cards is better than $2.
4. Assuming you don’t have spare actions (see above), +2cards (6 other cards when starting from a 5-card hand) is ~50% more likely to give you a dead action card than +$2 (4 other cards), and there’s no guarantee of getting at least $2 even if you don’t draw a dead action. +$2 is money in the bank.
5. One slight mitigating factor: +cards effects cycle your deck faster, which is good in the early game assuming you don’t draw dead action cards.
#2/3 is really the key weakness of +2cards effects: they aren’t good enough to operate a reliable draw engine or to quickly build a strong treasure deck.
With Witch vs. Mountebank specifically, the critical period for measuring their effectiveness is early in the game. How fast can I give out the curses and buy other good cards? In cursing games, your average card value does take a bit longer to get to $1, especially if you buy multiple terminal actions (very often a good move to buy 2 Witches or 2 Mountebanks). Mountebank is just that much more likely than Witch to quickly get you a second copy of the attack or an early Gold that can power you to victory.
But! in the specific case of Witch vs. Mountebank the more important consideration is the strength of the attack. On many boards where both cards are available, my personal preference is to get a single Mountebank followed by a single Witch. Mountebank is much stronger than Witch when it hits, and it’s guaranteed to hit at least once right away. I haven’t done any extensive testing or simulation to see whether this strategy is actually better than going straight for Witch.
I don’t understand why people like Mountebank that much. And if you look at the councilroom stat you posted, it’s still not clear (of course one or two Mountebanks is better than two Minions, but nobody buys only two minions). The main problem I have with Mountebank is that in every (two player) game I ever played (where both my opponent and me bought at least one Mountebank), the Curse stack was never emptied by the end, and will typically have 4 curses left. I looked at one game you featured here, and the same was true.
This means that in general the Mountebank will hit at most 3 turns and in all other turns the player will discard a curse. So that means the Mountebank will give a curse and a copper three times in the entire game and make the player discard a card the other times. This doesn’t look *that* strong.
Stats seems to agree with you, so I am wrong, but I don’t see why. Your first argument doesn’t work as, like I said, the curse stack will usually not be emptied (if there is only Mountebank). So there must be reasons for why the Mountebank is best, but I don’t think your arguments explain it well.
To be clear, I am talking only about two player games.
Adding a copper and a curse to your opponents deck is far worse than adding just a curse. Although a deck can saturate with curses and negate some later attacks, the damage has already been done with the first couple of hits. Those four extra cards could be considered as an extra hand of rubbish each time you cycle your deck, four more cards to trash out, or four cards dragging down your coin totals. However you look at it the damage is there.
I suspect it’s because people play more than one per turn, so the discard isn’t such a good defense.
While sometimes mountebank will miss, one of the other effects is that by discarding a curse you’re not trashing it. Nothing is more frustrating than having my salvager or upgrade with a curse in hand only to get hit by a mountebank.
When a card self-synergizes with itself – meaning the more you have, the more effective they are, you keep buying more of them. This applies to minion and laboratory for example. When a card is a terminal action, having 10 of them and nothing else in your deck lets you do… basically nothing, so you have to balance them with cards that let you play them, as well as the fact that there are usually only 10 curses in the game, so 3-4 witches/mountebanks is overkill.
1. Buy another Mountebank. Having one more than your opponent is VERY powerful. All the best cards are of the “buy or die” variety. You need to at least match the opponent’s striking power.
2. Extra coppers require extra trashing. If trashing is easy, with cheap early trashers and efficient late game trashers, neither Witch nor Mountebank will be that useful. But if trashing demands scarce actions, or is simply not available, then Mountebank clogs decks much faster than Witch.
3. In the opening, $2 is better than +2 cards. If you open with $5 (and that’s what the opening stats reflect), you can’t buy silver. Mountebank guarantees +$2, Witch doesn’t.
$2 is also better than 2 cards later on in decks which have been clogged with Curses and Coppers, such as what happens when people are playing Mountebank.
While +Cards are better than +Coin in the general case, +Coin is usually better when Curses are in abundance, and +2 Cards without any +Actions is in many cases worse than nothing. (How often do people buy Moat for the card draw?)
…
“And although it’s theoretically counterable by Moneylender/Counting House/other Copper-dependent cards, the extra Coppers slow your opponents down from getting back around to their own Mountebank.”
I’ve seen Apothecary work in this capacity, not surprising since it’s the one Copper-strategy card (I don’t consider a Turn 1 or 2 Moneylender to be “Copper strategy”) which is actually any good.
http://councilroom.com/win_rate_diff_accum.html?cards=apothecary%2C%20moneylender%2C%20counting%20house%2C%20coppersmith
Of course, even if you’re packing Apothecary to cycle through your Coppers and increase buying power, you’re going to want Mountebanks of your own.
Just got throned mountbanked yesterday on turn 5 or 6, was indeed game over 🙂
Hey, that was me 😀
My good luck against you was the only thing stopping me from losing leaderboard standing yesterday due to much, much worse luck against very much weaker players.
As much as the Saboteur is hated I like to see variations in viable strategies. So how do you think the Saboteur could be enhanced to be useful? Discard a Treasure card to put Saboteur back on your deck?
With a Tactitcian, I once got an engine running where I consistently played a Kinged Saboteur, and it ended with Danny having no cards with cost above 3 (he Islanded some away, but it didn’t help much by this point…) Can’t find the game for some reason.
Can’t find it on councilroom? Do you have an idea about the player names/game day? I think I have a bug somewhere that is dropping some games :(.
You also seem to have a bug in the game-search for King’s Court and Worker’s Village, when I query for one of these I get no results. Seems to be the ‘ .
Or do I use the wrong letter? C&P it from the results I think…
Might be the problem in this case when you try to find a KCed Saboteur. At least doesn’t make it easier…
Yeah, I should fix that one. I think the fix is simple, I am just not url unqouting the query parameter, and it works most of the time because only a few cards have annoying symbols in them.
I think it’s the same bug. When I search for King’s Court and Saboteur, it doesn’t find it.
Btw., the councilroom site is absolutely awesome, thank you so much for making it!
Glad you like it 😉
Saboteur is correctly balanced and costed. Make it cheaper or better and you utterly ruin multiplayer games, and not just because of groupthink.
Being able to get multiple mountebanks off a turn by more than person means that the curses are only going to be around for so long. So there’s no guarantee that curses out means that it’s endgame. And mountebank really shines when it’s used mroe than once a turn. Once KC a mountebank enough times that he couldn’t do anything.
Also, there’s a typo with minion where you say 5-5 split
I believe there’s mroe than one typo in your own post, bud.
I believe he’s talking about a 5-5 split in that you are all set if you have better than half of the Minions in a 2 player game. Not a 5/2 opening.
Tough to argue with these picks. I’d maybe switch Torturer and Ghost Ship but that’s nitpicking. Just curious. When the new expansions come out are you going to revise these lists in case new cards take there place??
My guesses for top 5 non-attacks
Laboratory
Wharf
Tactician
Apprentice
Venture
Hon. Mention
Library or Council Room?
In the seaside secret history Donald said that Wharf was given a +buy to make it as good as Merchant Ship. I think that Merchant Ship needs the +buy more than Wharf. It would even work better theme wise. But that’s just me. Sorry to derail this thread.
Duke for best $5 non-attack!
> [Minion] But it’s less flexible than the other attacks …
Yesterday I won a 3p game with Pirate Ships, and it was thanks to my sole Minion in play (along with Grand Halls and Pearl Divers) that I was able to let one of my 2 $8 Pirate Ships sail in incessant turns towards the end of the game, even though the deck became severely clogged in the process. I rarely used this card for $ in this game.
I’m generally in agreement with this list. I’ve had games where the Curse in hand from Torturer helped me, and not because of trashing it! (For instance, Hamlet/Cellar). Torturer is a beast of a card and not one to ignore on (almost) any board that has it, but it has its limitations. But on a board with +actions it’s definitely the best of the group. I think my favorite games I’ve ever played have been ones with a lot of +actions and either Torturers or Rabbles. My absolute favorite start is Torturer/Native Village.
The problems with torturer/+2 action sets is that 1) it can be swingy (first person to get off the multiple-torturer chain gets a huge advantage, much like the first person to get a province in a tournament game) and 2) it makes people hate Dominion.
The latter is less of a problem on isotropic, but very real if you’re playing with less than hardcore gamers. So then your choices are either to ignore the combo, potentially having it played on you, or use it anyway, and risk your friends not wanting to play Dominion with you anymore.
In multiplayer, Torturer really shines because you don’t need +Actions to truly punish your opponents. You just need another player who wants to buy Torturers. >: )
I was intrigued by having both Witch and Mountebank on the board at the same time and played some games with different card types. Overall it appeared that Mountebank was superior… especially in Colony games due to the extended turns necessary to buy that high. The only exception was when a village was on the board–which in my simulations was about 35% of the time. In nearly every game I played Witch won when it was accompanied by a Village type card. My basic premise for this was pretty simple. Both players buy Feast–I made sure that card was in every game to get the attacks out as soon as possible–and silver to start (the only exception was if Fishing Village was available then I made both buy that when $ was below 5.) In different scenarios I had both sides buy equal amounts of the different attack card (so 3 witches to 3 mountebanks or 2 to 2 or 1 to 1.) After that players bought gold when at 6. Provinces when at 8 and having at least 2 golds. Silver, after buying the allotted amount of attack cards, when 5 or less (except for when there was a village present. In these cases fishing village was bough constantly over silver, and other villages were bought at an interval of 1:3 compared to silver when the cost was 4.) The villages I used were regular, fishing, farming, walled and worker’s. Duchies instead of gold when there were 4 or less provinces, estates when there were 2 or less provinces.
The villages gave the witch a much stronger advantage, I think partly because of the ability to have multiple witches attack in the same turn–this was especially true with fishing village where extra actions were moved over to the next turn. In these situations the side with Mountebanks as the attack card usually had 6 or 7 of the curses at the end of the game!
In non village games, Mountebanks were again superior and better able to overcome curses and hamper with coppers.
The Native Village counter to Rabble is pretty strong, if you didn’t intend on using the NV to set up combos. It lets you green earlier, turning into a reusable Island that takes the Victory Points out of your deck without sacrificing them. The only drawback is that it becomes difficult to use your Native Village blind for fear of accidentally sniping one of your good cards, but then again, if you have multiple copies you can replenish your deck and save them for that final turn when having 1000 green cards in your discard pile doesn’t matter anymore.