Idle (?) Thoughts on Two Player Diplomacy

(This originally appeared in Diplomacy World #144)

People involved in creation of something out of nothing really do get their ideas in odd places, sometimes. I get a significant “input” to my game design when I’m in the shower and while I’m lying awake in bed. This idea popped into my head at “Oh Dark Hundred” recently.

A little introduction might help. My most well-known game is Britannia (1986 and later), and there’s going to be a reprint with plastic figures in the next year or so. To go along with that, the publishers wanted me to make a two player version of the game that lasted 60 to 90 minutes (Britannia itself is 4 to 5 hours.) I’m surprised and pleased at how well it has come out. It uses a new board, lasts 65 to 75 minutes, and is recognizably Britannia-like.

So it’s not surprising that now my thoughts occasionally turn to creating two player versions of games for more than two (Britannia has four players). Usually this is my own games, but this morning it was Diplomacy.

Insofar as the essence of Diplomacy is playing against the other players, a two player game cannot be Diplomacy. In other words, Diplomacy is about the psychological part of the game and much less about the game system. Yet the “Gunboat Diplomacy” variant has been popular, and that’s a game where any negotiation is prohibited. With two players, much of the psychology is gone.

So, I said to myself, if we’re going to abandon the essence of the game anyway, what can we do to change the game to make it more interesting for two players? Because with two players it would be a sort of a chess match that depended on who guessed best in the strategic/tactical part of the game, and would be devilishly difficult to balance fairly.

By removing the multiplayer aspect we remove much of the uncertainty of the game: with two players you can minimax it, you can assume the other player is perfect and play accordingly to maximize your minimum gain as in the premise behind the mathematical theory of games. Chess, Go, Checkers survive the situation because they are too complex to be solved by humans, though all three are played better by computers now than by the best humans. That’s not desirable, so I would replace the uncertainty of more than two players with two things: dice in combat and event cards.

Now I can hear many Diplomacy people sucking in a deep (dismayed?) breath at the idea of overt chance elements in the game, but I’ve explained why I think it’s necessary, and I have a dice combat system that would only mildly affect things but would provide an element of unpredictability. That method is that you roll one die per Army or fleet in the combat, including supports. The side with a higher sum wins the combat, with ties going to the defender most likely (or rerolled if both sides are attacking), but that’s something that would be determined in testing.

For example, a supported army (two) attacks an unsupported army (one). Rolls are 4,5 for the attackers, so the defender cannot win (can’t get more than a 6).

Occasionally a two on one would not dislodge the defender because the defender wins (or ties) the dice rolls. And in rare instances even a 3 to 1 attack might fail. On the other hand, a two versus three attack would occasionally succeed. The biggest change here would be that one-to-one attacks would sometimes succeed. (One vs one, 15 wins for each side plus 6 ties. If ties go to defender (assuming there is a defender rather than both moving), that’s 21 vs 15 (7 to 5).  Two vs one results: 15 ties, 21 wins for the weaker, 180 wins for the stronger. If ties go to weaker, it’s 180 to 36.)

I use this method in Eurasia (name likely to change to something like Surge of Empires), which is scheduled to be published sometime.


Another way to provide variance in combat would be to use combat cards rather than dice. Each player would have the same set of cards, but different ones in hand at different times, and it would be a guessing element involved in whether you play a strong card or weak card to add to the combat (there are also some special cards). I use such a method in several games but I’m not going to go into it here.


I don’t know if event cards would be necessary, and I haven’t tried to come up with any kind of scheme. But event cards are a way to add interest and variation to a game that the players can control in a way that they cannot control the dice, though with dice they can play to take account of probability.


The other point of uncertainty/variance would be in selection of the sides. While lying in bed I tried to think of an entirely fair three versus three and didn’t get very far. I’d probably use a combination of selection and chance to assign countries. The first player would choose a country, the second player would choose two countries, the third player would choose a second country. The third country that each received would be determined randomly from the three remaining. And for the one that was not controlled by either player, we could use a method known in some Diplomacy variants, where the players write orders (say, five of them?) for the units of the uncontrolled country. They can allocate all five (identical) orders to one unit or spread them amongst the units. If a unit received a majority of the same order then it would execute that order. Of course, you could go further and do that for all three countries that the players had not themselves selected.


How long would this game take the play? I should think it would hit that magic 60 to 90 minute length that is commonly desired nowadays in wargames, if not to the victory criterion then certainly to a point where one player resigns. It would be quicker, of course, if you had some electronic method of giving orders/moving the pieces. Handwriting orders for two or three countries takes a while.