"Should we ban it?" comes up whenever one card decides too many games. Most cards on community ban lists (King's Court, Goons, Rebuild, Cultist, Possession, Wharf) come from later expansions. We started with the two boxes most people own, the base game and Intrigue, and asked three questions about every card:
- Does it win by a mile in a simple deck? We add the card to Big Money (buy Silver, Gold and Provinces, nothing else) and play that against Big Money without it.
- Does it still win against other strong cards? Beating Big Money is easy. So every card's bot also played every other card's bot.
- Can you get it in time, and is it only good late? A $5 card you can't afford until turn 9 is a different card.
A card that passes all three is a problem: if it is on the table you have to buy it, and the game becomes about that one card. That is the real case for a ban.
Follow-up: we have since run the same tests on five ban-list cards from later expansions (Wharf, Goons, King's Court, Rebuild and Cultist). Witch beats all of them; Cultist comes closest. Read the follow-up.
The five strongest cards in a money deck
Share of games won by Big Money + this card against Big Money alone.
Test 1: every card against Big Money
Each card gets its own tuned buy rule: how many copies, and whether it comes before Silver or even before Gold. We tried up to 24 rules per card and kept the best one. Then every card played 20,000 fresh games against Big Money, going first in half of them.
All 52 cards, ranked by what they add to Big Money
Win rate of Big Money + the card against Big Money. 50% means the card adds nothing. Hover a card for its buy rule.
Show the numbers
| Card | Cost | Set | Wins vs Big Money | Best buy rule | Copies at end |
|---|---|---|---|---|---|
| Witch | $5 | Base | 96.2% | Buy up to 3, even over Gold | 3.0 |
| Courtyard | $2 | Intrigue | 81.6% | Buy up to 2, instead of Silver | 2.0 |
| Torturer | $5 | Intrigue | 81.2% | Buy up to 3, instead of Silver | 1.8 |
| Patrol | $5 | Intrigue | 80.4% | Buy up to 2, instead of Silver | 1.4 |
| Militia | $4 | Base | 77.1% | Buy one, plus one more per 10 cards in your deck, instead of Silver | 2.8 |
| Masquerade | $3 | Intrigue | 76.1% | Buy one copy, instead of Silver | 1.0 |
| Swindler | $3 | Intrigue | 75.8% | Buy up to 2, instead of Silver | 2.0 |
| Courtier | $5 | Intrigue | 73.7% | Buy up to 3, instead of Silver | 1.6 |
| Smithy | $4 | Base | 73.3% | Buy one copy, instead of Silver | 1.0 |
| Council Room | $5 | Base | 73.2% | Buy one copy, instead of Silver | 0.9 |
| Mill | $4 | Intrigue | 72.3% | Buy up to 4, instead of Silver | 3.6 |
| Bandit | $5 | Base | 70.7% | Buy one copy, even over Gold | 1.0 |
| Nobles | $6 | Intrigue | 70.7% | Buy one, plus one more per 10 cards in your deck, even over Gold | 2.7 |
| Library | $5 | Base | 69.7% | Buy one, plus one more per 10 cards in your deck, instead of Silver | 1.6 |
| Baron | $4 | Intrigue | 66.4% | Buy one, plus one more per 10 cards in your deck, instead of Silver | 2.6 |
| Replace | $5 | Intrigue | 65.9% | Buy up to 2, instead of Silver | 1.6 |
| Bridge | $4 | Intrigue | 62.8% | Buy one copy, even over Gold | 1.0 |
| Bureaucrat | $4 | Base | 62.2% | Buy one copy, even over Gold | 1.0 |
| Ironworks | $4 | Intrigue | 62.1% | Buy one copy, instead of Silver | 1.0 |
| Minion | $5 | Intrigue | 61.6% | Buy as many as you can, instead of Silver | 2.1 |
| Market | $5 | Base | 61.5% | Buy up to 4, instead of Silver | 2.0 |
| Sentry | $5 | Base | 61.1% | Buy up to 2, instead of Silver; trash Coppers too | 1.5 |
| Laboratory | $5 | Base | 59.8% | Buy as many as you can, instead of Silver | 2.0 |
| Poacher | $4 | Base | 59.6% | Buy as many as you can, instead of Silver | 4.8 |
| Mine | $5 | Base | 59.0% | Buy one copy, instead of Silver | 0.9 |
| Steward | $3 | Intrigue | 57.6% | Buy one copy, instead of Silver | 1.0 |
| Moat | $2 | Base | 57.5% | Buy one copy, even over Gold | 1.0 |
| Duke | $5 | Intrigue | 57.3% | Big Money, plus Duke and Duchy on $5+ once 6 or fewer gains end the game | 0.8 |
| Remodel | $4 | Base | 56.8% | Buy one copy, instead of Silver; trash Coppers too | 1.0 |
| Gardens | $4 | Base | 56.0% | Big Money, plus Gardens on $4+ once 5 or fewer gains end the game | 0.9 |
| Harem | $6 | Intrigue | 56.0% | Buy as many as you can, even over Gold | 5.5 |
| Diplomat | $4 | Intrigue | 55.8% | Buy one copy, even over Gold | 1.0 |
| Shanty Town | $3 | Intrigue | 55.6% | Buy one copy, even over Gold | 1.0 |
| Pawn | $2 | Intrigue | 54.9% | Buy up to 2, only with $2 hands | 0.4 |
| Festival | $5 | Base | 54.4% | Buy up to 3, instead of Silver | 1.9 |
| Moneylender | $4 | Base | 54.1% | Buy one copy, instead of Silver | 1.0 |
| Trading Post | $5 | Intrigue | 53.2% | Buy one copy, instead of Silver; trash Coppers too | 0.9 |
| Upgrade | $5 | Intrigue | 51.9% | Buy one copy, instead of Silver; trash Coppers too | 0.9 |
| Secret Passage | $4 | Intrigue | 51.2% | Buy one copy, instead of Silver | 1.0 |
| Conspirator | $4 | Intrigue | 50.5% | Buy one copy, even over Gold | 1.0 |
| Vassal | $3 | Base | 50.0% | Buy one copy, instead of Silver | 1.0 |
| Chapel | $2 | Base | 49.7% | Buy one copy, only with $2 hands; trash Coppers too | 0.3 |
| Cellar | $2 | Base | 49.2% | Buy one copy, only with $2 hands | 0.3 |
| Merchant | $3 | Base | 48.6% | Buy one copy, even over Gold | 1.0 |
| Lurker | $2 | Intrigue | 44.7% | Buy up to 3, only with $2 hands | 0.5 |
| Wishing Well | $3 | Intrigue | 44.2% | Buy one copy, instead of Silver | 1.0 |
| Workshop | $3 | Base | 43.9% | Buy one copy, even over Gold | 1.0 |
| Harbinger | $3 | Base | 42.0% | Buy one copy, even over Gold | 1.0 |
| Artisan | $6 | Base | 41.5% | Buy one copy, even over Gold | 1.0 |
| Mining Village | $4 | Intrigue | 38.7% | Buy one copy, instead of Silver | 0.3 |
| Village | $3 | Base | 32.5% | Buy one copy, even over Gold | 1.0 |
| Throne Room | $4 | Base | 13.7% | Buy one copy, instead of Silver | 1.0 |
Witch is alone at the top with 96.2%. It draws two cards and hands the opponent a Curse every time: a minus point that clogs their deck for the rest of the game. By the end, the Big Money player holds 9.3 of the 10 Curses on average.
The cards at the bottom are not bad cards. Throne Room, Village and Mining Village only work with other Actions, and a money deck has none, so they lose (14%, 33%, 39%). This test measures one thing: what a card does for a simple deck.
Test 2: against every other card
Big Money is a low bar. So we played all 52 tuned bots against each other, and against plain Big Money: 1,378 pairings of 2,000 games each, with both cards in the supply.
The top 12, head to head
Each cell: how often the row card's bot beats the column card's bot. Blue = row wins, red = row loses.
Show the numbers
| Witch | Torturer | Patrol | Courtyard | Masquerade | Swindler | Militia | Courtier | Council Room | Smithy | Mill | Bandit | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Witch | - | 83% | 75% | 84% | 52% | 70% | 90% | 83% | 87% | 86% | 87% | 86% |
| Torturer | 17% | - | 50% | 53% | 54% | 49% | 69% | 58% | 57% | 57% | 71% | 58% |
| Patrol | 25% | 50% | - | 49% | 52% | 53% | 61% | 63% | 58% | 59% | 66% | 63% |
| Courtyard | 16% | 47% | 51% | - | 54% | 35% | 56% | 62% | 60% | 59% | 62% | 61% |
| Masquerade | 48% | 46% | 48% | 46% | - | 69% | 43% | 55% | 54% | 54% | 62% | 56% |
| Swindler | 30% | 51% | 48% | 65% | 31% | - | 52% | 53% | 61% | 60% | 50% | 55% |
| Militia | 10% | 31% | 40% | 44% | 57% | 48% | - | 46% | 48% | 48% | 69% | 56% |
| Courtier | 17% | 42% | 38% | 38% | 46% | 47% | 54% | - | 46% | 48% | 51% | 54% |
| Council Room | 14% | 43% | 42% | 40% | 47% | 39% | 52% | 54% | - | 52% | 49% | 51% |
| Smithy | 14% | 43% | 42% | 41% | 46% | 41% | 52% | 52% | 48% | - | 51% | 51% |
| Mill | 13% | 29% | 34% | 38% | 38% | 50% | 31% | 50% | 51% | 49% | - | 51% |
| Bandit | 14% | 42% | 37% | 39% | 44% | 45% | 44% | 46% | 49% | 49% | 49% | - |
Witch beats all 52 opponents, with an average win rate of 91%. The best answer our bots found was Masquerade, which wins 48.1%: it makes every player pass a card to the left, and the Witch victim passes a Curse straight back. Everything else loses badly. The practical answer to Witch is buying Witch yourself.
Below Witch, the strong cards beat each other in circles, which is what you want from a card game: Swindler beats Courtyard (65%), Masquerade beats Swindler (69%) and Militia beats Masquerade (57%).
The best answers to Witch
Win rate of each card's bot against Big Money + Witch. A Witch mirror would be 50%.
Show the numbers
| Card | Wins vs Witch |
|---|---|
| Masquerade | 48.1% |
| Swindler | 30.4% |
| Patrol | 24.9% |
| Sentry | 18.2% |
| Courtier | 16.9% |
| Torturer | 16.8% |
| Moat | 16.0% |
| Courtyard | 15.7% |
| Replace | 14.5% |
| Bandit | 14.4% |
| Smithy | 13.7% |
| Council Room | 13.5% |
| Trading Post | 13.2% |
| Mill | 13.0% |
| Library | 11.3% |
| Remodel | 10.6% |
| Militia | 10.4% |
| Upgrade | 9.8% |
| Baron | 9.7% |
| Bureaucrat | 8.4% |
| Gardens | 7.6% |
| Steward | 7.6% |
| Ironworks | 7.4% |
| Chapel | 7.1% |
| Minion | 6.7% |
| Mine | 6.6% |
| Bridge | 6.6% |
| Nobles | 6.6% |
| Duke | 6.0% |
| Lurker | 5.7% |
| Poacher | 5.4% |
| Laboratory | 5.2% |
| Pawn | 5.2% |
| Market | 5.0% |
| Vassal | 4.8% |
| Diplomat | 4.8% |
| Festival | 4.7% |
| Moneylender | 4.3% |
| Shanty Town | 4.3% |
| Secret Passage | 4.3% |
| Merchant | 4.1% |
| Cellar | 4.0% |
| Conspirator | 3.8% |
| Workshop | 3.4% |
| Artisan | 2.9% |
| Harbinger | 2.6% |
| Wishing Well | 2.4% |
| Mining Village | 2.2% |
| Harem | 2.0% |
| Village | 1.5% |
| Throne Room | 0.3% |
Test 3: can you get it in time?
Cost comes before power. Your first two turns split your 7 Coppers into $4/$3 or $5/$2, and the 5/2 split happens in only 16.7% of games (that one is exact math). So a $5 card like Witch usually arrives a little later: our Witch bot got its first copy on turn 3.2 on average, the $4 Smithy on turn 1.5.
To see what a slow start costs, we banned each bot from gaining its card before turn 5, 9 or 13. A game against Big Money lasts about 23 turns each with Witch, and 19 with Smithy.
What a late start costs
Win rate against Big Money when the card can't be gained before a given turn.
- Witch
- Courtyard
- Torturer
- Nobles
- Duke
Show the numbers
| Card | From turn 1 | Turn 5 | Turn 9 | Turn 13 |
|---|---|---|---|---|
| Witch | 96% | 87% | 65% | 51% |
| Courtyard | 82% | 74% | 58% | 51% |
| Torturer | 81% | 70% | 54% | 51% |
| Patrol | 80% | 72% | 55% | 51% |
| Militia | 77% | 66% | 53% | 51% |
| Masquerade | 76% | 61% | 53% | 50% |
| Swindler | 76% | 63% | 54% | 50% |
| Courtier | 74% | 63% | 51% | 50% |
| Smithy | 73% | 67% | 56% | 51% |
| Council Room | 73% | 67% | 54% | 50% |
| Mill | 72% | 70% | 60% | 54% |
| Bandit | 71% | 56% | 44% | 48% |
| Nobles | 71% | 71% | 69% | 57% |
| Library | 70% | 64% | 53% | 50% |
| Baron | 66% | 57% | 54% | 51% |
| Replace | 66% | 61% | 53% | 50% |
| Bridge | 63% | 51% | 49% | 49% |
| Bureaucrat | 62% | 46% | 45% | 48% |
| Ironworks | 62% | 53% | 50% | 51% |
| Minion | 62% | 57% | 51% | 50% |
| Market | 62% | 56% | 51% | 51% |
| Sentry | 61% | 54% | 50% | 51% |
| Laboratory | 60% | 56% | 51% | 50% |
| Poacher | 60% | 54% | 51% | 50% |
| Mine | 59% | 51% | 50% | 49% |
| Steward | 58% | 52% | 50% | 51% |
| Moat | 57% | 50% | 49% | 50% |
| Duke | 57% | 56% | 56% | 57% |
| Remodel | 57% | 48% | 51% | 50% |
| Gardens | 56% | 57% | 57% | 56% |
| Harem | 56% | 57% | 60% | 56% |
| Diplomat | 56% | 50% | 48% | 49% |
| Shanty Town | 56% | 50% | 48% | 49% |
| Pawn | 55% | 51% | 49% | 51% |
| Festival | 54% | 53% | 50% | 51% |
| Moneylender | 54% | 49% | 49% | 51% |
| Trading Post | 53% | 44% | 47% | 50% |
| Upgrade | 52% | 45% | 47% | 49% |
| Secret Passage | 51% | 52% | 51% | 50% |
| Conspirator | 51% | 43% | 44% | 50% |
| Vassal | 50% | 49% | 50% | 51% |
| Chapel | 50% | 48% | 50% | 50% |
| Cellar | 49% | 51% | 50% | 50% |
| Merchant | 49% | 46% | 46% | 49% |
| Lurker | 45% | 49% | 50% | 50% |
| Wishing Well | 44% | 48% | 48% | 50% |
| Workshop | 44% | 35% | 35% | 46% |
| Harbinger | 42% | 44% | 45% | 49% |
| Artisan | 42% | 40% | 43% | 50% |
| Mining Village | 39% | 50% | 52% | 50% |
| Village | 33% | 34% | 38% | 47% |
| Throne Room | 14% | 28% | 41% | 48% |
Witch is an opening card. Push it back to turn 9 and it drops from 96% to 65%; from turn 13 it is a coin flip (51%). A Curse hurts most when it sits in the opponent's deck for the whole game. The other early-game cards fade the same way: Torturer goes from 81% to 54%, Bandit goes from 71% to 44% and Patrol goes from 80% to 55%.
The cards that don't care are the green ones: Duke (57% from turn 1, 57% from turn 13), Gardens (56% from turn 1, 56% from turn 13) and Harem (56% from turn 1, 56% from turn 13). They are worth points whenever you get them, and Nobles (two points and an Action) still wins 69% from turn 9. So "can you get it in time" does not save us from Witch: it costs $5, and you will have $5 by turn 3 or 4 in almost every game.
Is it only strong late?
Some cards win early, some score late. Here is the average score difference against Big Money after each round. Once a game is over, it counts with its final margin.
When each card pulls ahead
Average victory points ahead of Big Money after each round. Below zero = behind.
- Witch
- Courtyard
- Mill
- Smithy
- Duke
Show the numbers
| Round | Witch | Courtyard | Mill | Smithy | Duke |
|---|---|---|---|---|---|
| 1 | 0.0 | 0.0 | 0.5 | 0.0 | 0.0 |
| 2 | 0.0 | 0.0 | 1.0 | 0.0 | 0.0 |
| 3 | 0.1 | 0.0 | 1.6 | 0.0 | 0.0 |
| 4 | 0.2 | 0.0 | 2.3 | 0.0 | 0.0 |
| 5 | 0.5 | 0.0 | 2.8 | 0.0 | 0.0 |
| 6 | 0.9 | -0.0 | 3.1 | -0.0 | 0.0 |
| 7 | 0.9 | -0.7 | 2.7 | 0.1 | -0.0 |
| 8 | 0.2 | 0.2 | 1.5 | 0.9 | -0.0 |
| 9 | -0.4 | 1.8 | 0.6 | 1.8 | 0.0 |
| 10 | -0.1 | 3.1 | 0.5 | 2.5 | 0.0 |
| 11 | 0.7 | 4.0 | 0.7 | 3.0 | 0.0 |
| 12 | 2.0 | 4.5 | 1.2 | 3.2 | -0.0 |
| 13 | 3.5 | 4.9 | 1.7 | 3.4 | 0.0 |
| 14 | 5.0 | 5.3 | 2.1 | 3.6 | -0.0 |
| 15 | 6.5 | 5.6 | 2.5 | 3.7 | 0.0 |
| 16 | 8.0 | 6.0 | 3.0 | 3.9 | 0.1 |
| 17 | 9.3 | 6.5 | 3.7 | 4.3 | 0.2 |
| 18 | 10.5 | 6.9 | 4.3 | 4.6 | 0.3 |
| 19 | 11.7 | 7.3 | 4.6 | 4.9 | 0.4 |
| 20 | 13.1 | 7.5 | 4.9 | 5.1 | 0.5 |
| 21 | 14.1 | 7.6 | 5.0 | 5.2 | 0.6 |
| 22 | 14.9 | 7.7 | 5.1 | 5.4 | 0.8 |
| 23 | 15.6 | 7.8 | 5.3 | 5.5 | 1.1 |
| 24 | 16.2 | 7.9 | 5.3 | 5.6 | 1.4 |
| 25 | 16.7 | 8.0 | 5.4 | 5.6 | 1.7 |
A surprise: Witch is not ahead early. After round 10 it is level with Big Money (-0.1 points), because its owner spends $5 hands on Witches instead of Gold while Big Money starts on Provinces. Then the Curses bite and the lead runs away: +6.5 after round 15, +13.1 after round 20. Courtyard is the opposite kind of strong: ahead from round 9 and steady after that. Mill is ahead from the first turns (every Mill is a point) and Duke does nothing until the Duchies come out at the end (+0.5 after round 20). A card that only pays off late can be raced. That makes Duke a fine card, but no ban candidate.
What about engines?
Experienced players know Big Money is not the end of the story. We also tested a few classic two-card strategies. Fair warning: engines are hard for bots, and a good human plays them much better than ours do.
Two-card strategies
Win rate of each strategy against Big Money, against Big Money that takes Gardens when it sees a rush, and against Big Money + Witch. 10,000 games per bar.
- vs Big Money
- vs Big Money (takes Gardens)
- vs Witch
- Village/Smithy: Villages so you can play several Smithies in one turn.
- Mining Village/Torturer: Chain Torturers so the opponent discards or takes Curses again and again.
- Throne Room/Witch: Throne Room a Witch: 4 cards and 2 Curses.
- Festival/Library: Festivals give actions and money but no cards; Library draws you back up to 7.
- Workshop/Gardens: Gain a card every turn with Workshop and pile up Gardens.
- Ironworks/Gardens: Like Workshop, but Ironworks also gives a bonus for what it gains.
Show the numbers
| Strategy | Big Money | Big Money (takes Gardens) | Witch | Courtyard | Torturer | Patrol | Masquerade |
|---|---|---|---|---|---|---|---|
| Village/Smithy | 51% | 52% | 4% | 22% | 23% | 23% | 27% |
| Mining Village/Torturer | 51% | 48% | 50% | 54% | 79% | 80% | 14% |
| Throne Room/Witch | 85% | 28% | 85% | 64% | 60% | 57% | 42% |
| Festival/Library | 46% | 38% | 42% | 71% | 72% | 10% | 38% |
| Workshop/Gardens | 93% | 70% | 96% | 56% | 84% | 66% | 65% |
| Ironworks/Gardens | 80% | 80% | 98% | 82% | 97% | 84% | 91% |
Village/Smithy, the first engine most people try, wins only 51% against Big Money, while one Smithy in a money deck wins 73%. That matches long-standing player wisdom: Villages cost you money now and only pay off when you draw them together with your Smithies. Throne Room makes Witch worse, not better: Throne Room/Witch wins 28% against a plain Witch deck.
The combo that beats Witch
The Gardens rushes are the real surprise. A Gardens is worth 1 point per 10 cards you own, so you take a card-gainer (Workshop or Ironworks), gain a card every turn, then empty the Gardens and Estate piles to end the game before your opponent has their Provinces. Against Big Money that works 96% (Workshop) and 98% (Ironworks) of the time. Against Witch it works even better (93% and 97%), because every Curse is one more card for your Gardens to count.
That is the naive test, so we went one level deeper and built bots that fight for the Gardens.
Can anyone stop a Gardens rush?
Win rate of each rush against bots that take a Gardens over Gold as soon as the rusher owns one, and against the other rush. 10,000 games per bar.
- Big Money (takes Gardens)
- Witch (takes Gardens)
- Big Money + gainer (takes Gardens)
- The other rush
Show the numbers
| Rush | Opponent | Rush wins | Turns per player |
|---|---|---|---|
| Workshop/Gardens | Big Money (takes Gardens) | 56.7% | 17.0 |
| Workshop/Gardens | Witch (takes Gardens) | 70.4% | 16.7 |
| Workshop/Gardens | Big Money + Workshop (takes Gardens) | 85.7% | 12.7 |
| Workshop/Gardens | Ironworks/Gardens | 30.4% | 11.8 |
| Ironworks/Gardens | Big Money (takes Gardens) | 82.3% | 16.5 |
| Ironworks/Gardens | Witch (takes Gardens) | 85.8% | 16.5 |
| Ironworks/Gardens | Big Money + Ironworks (takes Gardens) | 79.1% | 13.8 |
| Ironworks/Gardens | Workshop/Gardens | 69.9% | 11.8 |
Fighting back works against Workshop/Gardens: Big Money that takes Gardens holds it to 57%. It does not work against Ironworks/Gardens. Ironworks gives you a card when it gains a Gardens, so that rush is faster, and it won 79%-86% against every counter we built, including Witch that takes Gardens (86%). It even beats the Workshop rush (70%). With both cards on the table, it is the strongest strategy we found. A strong human engine player might race it; our bots could not.
Should you ban it? You can't ban a combo, and you would not want to ban Ironworks or Gardens: alone they are ordinary (62% and 56% with Big Money). In a random 10-card kingdom both show up together only 3.4% of the time. Learn to spot it, and be the one who rushes.
So, should you ban Witch?
Witch shows up in 19% of random kingdoms, and every time it does, the game is about Witch: no other card beats it, so both players have to buy it. If that bores your group, banning Witch is the one ban in the base game and Intrigue that the numbers back up. If you keep it, buy it early. Don't count on Moat: our Moat bot wins only 16% against Witch. Masquerade is the real defense (48%).
At the table
- Witch out? Buy one with your first $5, even over Gold, and add more: our best bot ends with 3.0 on average.
- Masquerade is the best card against Witch: pass the Curses back.
- Ironworks and Gardens on the same table? Rush them, or take Gardens early yourself.
- Village and Throne Room need other Actions. Don't add them to a money deck.
Which box first? Dominion vs Intrigue has the answer, and the expansion ranking covers what to buy after that.
How we tested
We wrote our own rules engine for the second editions of the base game and Intrigue, with checks for the awkward cards (Throne Room on Mining Village, Masquerade passing, Swindler, Torturer choices, Diplomat reactions, tie-breaks). All bots share one set of decisions for playing cards, in the style of the open-source simulators Dominiate and Geronimoo's DominionSim; what differs between them is what they buy.
The baseline is the well-known "Big Money Ultimate": buy Gold and Silver, Provinces once the deck holds more than $18 in money, and Duchies and Estates near the end. None of the 216 variants of those thresholds we tried beat it. For each kingdom card we tried up to 24 buy rules in 2,000-game runs, replayed the best four with 8,000 fresh games each, and measured the winner with 20,000 more. Every number on this page comes from fixed random seeds. A tie goes to the player who had fewer turns, as in the rulebook. Going first matters a little: in Big Money against itself the first player wins 52% and 8% of games are tied.
Limits. These are two-player games with only the cards being tested in the supply; a real game has ten kingdom cards, and a Moat or Masquerade on the table changes the Witch picture. Our bots play money decks well and engines poorly, so cards that shine in engines (Throne Room, Village, Festival, Conspirator) are undersold. "Best buy rule" means the best of the rules we tried, not the best possible. Five cards from later expansions are tested in a follow-up. More on how we test.
Image credits
- Photo: Shannon Prickett (cropped by the uploader), via Wikimedia Commons (CC BY 2.0, Shannon Prickett (resized)): A Dominion supply laid out on a table: piles of Kingdom cards, treasure and Victory cards
- Box art: Rio Grande Games (publisher box art via retailer product image (philibertnet.com), shown to identify the product): Box of Dominion (2nd edition): a castle on a green hill above a river valley, riders on the road in front
