Version History

• 1.3: Prisoner's dilemma mechanism added to wars/quests, guessing mechanism added to disasters
• 1.2: Streamlined and simplified rules, all cards open, disaster cards strike specific chivalries
• 1.1: Closed knight cards and open damosel cards instead of vice versa
• 1.0: First edition

The complete rules are available in the PDF file to the right. In the following sections, I will describe how they came to be.

Implemented Rules

There are many ideas that came to live in Knights & Damosels^©. On this page, I would like to present some of them and explain the reasoning behind them.

The Arthur and Merlin cards

As described under Game, Arthur was introduced to remove random elements from the game and Merlin to handle the secrecy of the damosels behind the chivalry cards. Arthur's power to choose the outcome of failed war/quests and disasters was not really turn dependent so it was natural to give his role to the 1st player in turn.

Merlin on the other hand, would be able to use his wisdom to either play defensively and pick his own damosel's chivalry card or offensively and pick a chivalry card not that his knights needs but that increases the probability of a certain knight to take his damosel's card. Both option would disrupt the damosel game of playing carts attractive to certain knight but by giving his role to the last player in turn. This turned out thematically as well, in some versions of the legends, Merlin could see his destiny but not do anything about it.

The starting conditions

In the 1st turn, no event card is drawn. This is simply to provide the knights with some chivalry cards before the action starts. Without this rule, an early war or quest would be difficult, particularly in the 3 player game where they would risk entering the war or quest with only 2 types of cards or even only 1.

The knights are also given 2 worship points at the start. This is to allow for an early disaster or an early joust loss and still give them 1 worship point left to joust.

The balanced tactics

Vassalage is the "safe" way to earn worship points while the arms and the virtues are more risky. Nevertheless, they still have to be balanced so that one is not more optimal than the other. The expected return is easy to calculate:

In addition, cards may be lost during the game and that probability needs to be taken into account as well. Vassalage cards may be lost during disasters (25% of the events) if a player chooses the same card as King Arthur (33% of the times if Arthur chooses randomly). Arms or Virtue cards may be lost during wars or quests (also 25% of the events) if the 3 players participating don't choose the same card as a randomly drawn card (also 33% of the times as the table below shows).

Naturally, the probabilities will differ depending on which cards are in play. If Arthur has only 1 vassalage card himself, he's unlikely to choose that card to be removed. If 1 knight only has 1 virtue card, the other knights may choose other cards to increase the likelihood of success. However, as a first calculation, the above were fine and tests and simulations showed that the types of cards were fairly similar both in terms of return and loss.

The number of cards

The number of cards dictates elements like the game length and the victory conditions. I wanted the game to last no more than 30 minutes but also allow each player to play Arthur and Merlin at least once. This gave me a lower limit of 6 turns to allow for a 6 player game. It was then natural to set the upper limit to 12 turns, a number that can be divided with 3,4 and 6 players. That led me to 12 event cards, 3 of each kind. To keep the card drawing interesting, I only wanted two thirds of the cards to be drawn in an average game. Since the knights get 1 card in the 1st turn, without any event card being drawn, but 2 of the 3 disasters are likely, without any cards being given, I ended up with 8 + 1 - 2 = 7 cards as a good victory condition. With 9 cards in total, the knights would still be able to build either a mixed deck or focus on 1 or 2 card types before the game ends but not to build a complete deck.

The scaling

Scaling a game to different number of players is often an issue in game design. In Knights & Damosels, one option would have been to increase the number of event cards with the number of players but this would have increased the game length. Instead, I scaled the victory conditions.

For knight victory, the object of 7 cards was not affected by the number of players as all players get on average 1 card per turn. However, the object of worship points needed to be scaled. Considering that worship victory must be attained in five worship scoring events (the first card gives no worship and the seventh card means knight victory), that each scoring event rewards 10 points (including bonus) and that there won't be enough cards in play the first turn to earn all 10 points, I calculated with 8+10+10+10+10=48 points for an average 6 player game and 2 less for each player less since 2 less enter play in the 2 first turns. This gave the following victory levels:

• 3 player games: 39 / 3 = 13 points or 2.6 per turn
• 4 player games: 42 / 4 = 11 points or 2.2 per turn
• 5 player games: 45 / 5 = 9 points or 1.8 per turn
• 6 player games: 48 / 6 = 8 points or 1.6 per turn

Adding 2-4 "spread" worship to allow for different levels between the players, I ended up with the levels 12-13-14-15.

Tests and simulations showed that those levels were high enough to make a worship point victory about equivalent to the knight victory.

The damosel victory, finally, was scaled based on expected distribution. With a knight victory condition of 7 cards, I expected the following distribution of damosel cards per knight, given that damosels are fairly successful in playing the cards their knight wants. (damosels being labelled A-F where A is the knight's own damosel):

• 3 player games: BBBB CCC
• 4 player games: BBBB CC D
• 5 player games: BBB CC D E
• 6 player games: BBB C D E F

This meant that a damosel would have to get 1 more card with a knight in 3-4 player games compared to 5-6 player games. Tests and simulations were required to find a level comparable with the other victory conditions and I finally ended up with the levels of 4 and 5.

The jousts

The jousts were needed to add more action and interaction in the struggle for cards but it was also important to prevent them from being used so often that they disrupt the game. One way to do so was to add the risk of losing 1 worship point and another to give the challenger less probability to succeed. To make a joust worth the cost, an expected return of more than 1 worship point is thus needed. In the common scenario, a knight may earn 2 points from going on a quest and another 1 point from the joust victory itself so a probability to succeed of 33% would be adequate. The current "tie break" mechanism gives exactly this probability.

The Battle of Camlann

Playing for the simple reason of acquiring victory points is often perceived as an artificial way of motivating player actions. The idea of the Battle of Camlann is to give the players an opportunity to actually use the cards and the points they've gathered during the game. Cards could be used for the actual battle and points to tell the winners apart. But wouldn't such rules result in all-against-one battles against the winner, since everybody will want to be on the winning side? Not necessarily, since less player on one side means greater chance for each player on that side to win the lottery afterwards. But this requires that your side has a fair chance to win even if it is smaller. How important would then a player majority be in the Battle of Camlann? Calculations would have to tell.

In the tables below, I have assumed a winner with 8 cards (7 to obtain victory and an 8th bonus card for the victory) against 2-4 enemies with 6 cards each. The winner's card are listed in the top row and the opponents' cards in the left column and the results are listed with the cards defeated by the winner listed first.

1 player with 8 cards against 2 players with 12 cards

The one player's 8 cards will defeat 12,8 opponent cards.

1 player with 8 cards against 3 players with 18 cards

The one player's 8 cards will defeat 17,6 opponent cards.

1 player with 8 cards against 4 players with 24 cards

The one player's 8 cards will defeat 22,9 opponent cards.

The conclusion is that the lonely winner is likely to win against 2 opponents, draw against 3 and lose against 4, although the battle is even in all 3 cases! Perfect! Now a player's choice between the sides is not dictated by the side that offers the greatest chance of to win the battle but also that offers the greatest chance to win the following lottery. The last thing needed to ensure an exciting final battle was thus the rule that a lonely player gets 1 extra card in the 5 and 6 player games to avoid too uneven battles.

The 3 player version

One of the main tensions of the game is the agony over whether to take a card or not. However, in a 3 player game with only 2 opponents, it's easier to conclude who gives which card. Moreover, by memorizing one players' cards, you can play similar cards to be sure that the they can only be taken by the other player. Consequently, if all players apply that strategy, all players will win a damosel victory at the same time. The simple solution was to add a 4th set of chivalry cards from a non-participating damosel. That created a 3rd "neutral" opponent that prevented the forced taking of cards as players can take the neutral card and discard the player card.

Finally, why does Arms beat Vassalage but not Virtue?

I needed a rock-paper-scissors relation between the cards and why not? Virtues were considered more important than brute force for a knight but would simple farmers and hunter care about virtues?

Rules (Video)

Rules (PDF)

... and Rejected Rules

There are of course also ideas that did not make it and here I explain why.

Arthur's war/quest decision

Arthur is the one who decides who will fail a war or a quest rather than using a random mechanism. How about letting him decide which card that is needed to succeed rather than drawing a random card? Although this would remove the last random mechanism, it would make the Arthur role too powerful. Unless the participating knights would have one of each card type, they would never be able to succeed without the King's good will (and since he also plays a competing knight, why would he want others to succeed?).

Fields and reserves

The first version of the Battle of Camlann rules made a difference between played cards (on the field) and not played card (in the reserve) in a turn. The idea was that the object of the guessing game would be to play a card not played by the other side and having that card defeat all cards in the reserve. The other cards would simply only defeat 1 card on the field. However, not only was this complicated but it also made the 1st turn determining the entire battle. The current version, where all cards on the field can defeat each other in rock-paper-scissors style is both simpler and allows for fluctuating battle luck.