Castles in Their Proper Places (ish)

A long story about a small game design decision.

I've not done anything with The Castle War for a little while, but have been thinking about it, particularly in light of some feedback I received.  If you didn't know, or don't remember, this is a two-player card game inspired by the 12th century civil war in England between King Stephen and Empress Matilda, which eventually led to the crowning of Matilda's son as Henry II. The game is based on a period of that conflict where there was essentially a stalemate characterised by a series of sieges and exchanges of castles, where there were no significant open battles.

Game play involves the two players deploying forces and using tactics and events across six castles that were all significant during the period. If one player ever controls all six castles, they immediately win; otherwise, whoever controls the most castles at the end of the game (when the card deck runs out) is the winner.

But with six castles, a tie is clearly possible. How we break that tie is a question that has taken some thought.

I drew a sketch map to show the (approximate) locations of the castles.

Originally, the castles were nameless and numbered, 1 to 6, to tie in with a dice rolling system that was used to open up opportunities to the players. Then if there was a tie at the game end, you added up the numbers on the castles you controlled and whoever had the highest score won. This meant that the high numbered castles were more valuable than the low numbered ones, but if you concentrated on defending the two highest scoring castles, the other player could win by capturing the rest, even though their combined score was not as high. This was fine, but somehow a lot of players seemed to have difficulty internalising that the scores were only useful for tie breaking.

To give the game a little extra flavour, I wanted to name the castles, so a little research yielded the six locations depicted in the rough map above. (Fun fact: I live in between Faringdon and Wallingford, and have visited the sites of both castles.)  Furthermore, I wanted to have the initial ownership of the castles reflecting history as best I could.  (Another fun fact: Wallingford castle is distinguished in having not changed hands throughout the conflict, despite being sieged and attacked with some vigour.) The problem here is that, in order to keep the initial scores balanced, I had to distribute the names of the castles along the line of six with no real attention to geography. This was fine from a game play point of view, but as this is inspired by history, it bothered me a bit.

So now a slight diversion. We'll bring this back in a bit.

Before this stalemate stage of the war, Matilda was in a very strong position and very nearly got herself crowned as Queen, it seems there were effectively three power bases that needed to be brought onside in order to become officially the monarch: the Church, the royal treasury (based in the old Wessex capital of Winchester), and the people (or at least the leaders) of London. Matilda had gained Church support and held the treasury, but never managed to convince the leaders of London to back her cause -- partly, it seems, due to her own arrogance, and partly due to the actions of another Matilda, this one the Queen, and wife of the then-imprisoned King Stephen.

Eventually, Queen Matilda and her allies managed to retake Winchester, capture some key prisoners, and arrange for an exchange, thus freeing the King and bringing us to the Castle War's stalemate situation.

Anyway, three factions that could be influenced... that gave me an idea...

If there was some form of way of tracking influence with these three factions or power bases, then whoever has influence over the most of them wins an otherwise tied game. There are a number of ways this could be handled, but I have decided to make each faction a simple binary: they are supporting either Stephen or Matilda, and I chose to represent this with a set of three flippable cards, similar to the way I have players flipping the castle cards to show ownership.

I then added an icon representing one of the factions to each card in the game, with an even distribution of these icons throughout the deck.  The plan now is that each player can now take an action on their turn to flip one of the faction influence cards by discarding a set of three cards with matching faction icons on them.  The set of three decision is intended so that, with a full hand (of six cards) you stand a decent chance of having a set, but using it will dramatically reduce your options for the next turn and you may have to give up on some other useful cards.

With the numbers on the castles being irrelevant for game end purposes, I can now arrange the row of castles in an order that follows a plausible route between them that you may see in the map, from Wareham to Wallingford, and as there are some minor elements of adjacency in the game, that makes more thematic sense now.

You wouldn't put the faction cards there for an actual game, but you get the idea.

A common question from players is about how to cycle out unwanted cards in hand, so this system will give a way to address that, assuming those unwanted cards can be combined into a set.

So far this seems to work OK, but I am a little concerned that collecting three matching icons to gain influence in something might just feel boring and unthematic to a lot of players.  I'm hoping that it's one of those elements of the game that if you totally ignore it you may be at a disadvantage, but if you focus on it too much, you will almost certainly lose.  It should give a small, but meaningful, additional option at some points in the game, without dominating.  Only more playtesting will show if this is actually a decent decision, and if the set collection doesn't work out, I am certain there are other options to allow us to do something similar in the game.

The other real issue here is that this setup gives Stephen an advantage, winning the tiebreak if nothing else changes, so Matilda gets the first turn, which also appears advantageous.  Again, playtesting should help reveal if one side has a significant advantage here and if there is, there are plenty of ways to mitigate that.

Castle, Castle, Castle...

I've been making a little more progress with my 2-player game about castle sieges and battles during the 12th century.  I had previously done a little light testing with a small, hand-made card set, and then spent a while setting up a nanDECK script and data file to build a "full" deck that should allow a proper play through.

By way of a sanity test, my daughter, Miss B, had a play of the new set, and this seemed to work reasonably well (and earned the coveted accolades of "not terrible" and "better than the last version"), so I felt okay about taking the game out of the house for more testing.

At the monthly weekend playtest meetup in London I had two fine people giving the game a shakedown.  We cut the game off a little short in order to allow time for playing someone else's prototype (a very entertaining game about moving cows into a field and trying to get them past a bovicidal rival farmer), but was able to see the game in play and have some very useful feedback discussions.

Version 2, and the second castle from the right is looking in serious trouble.

Overall, the testers felt that the game did offer them interesting decisions, but the combat part of the game seemed complicated and hard to understand (the game tries to combine a rock-paper-scissors contest of tactical orders with a comparison of relative strengths), and we noted a bunch of elements that seemed either over or under powered.  As I often say, in the early stages of making a game, I don't worry much about balance, but it is always worth noting where there is a perception of a problem so it can be addressed later.

Over the following couple of days I managed to get testing done with a couple of local friends and tried out a few tweaks.  My first  try was very close to the version I tried on at the London meetup, which confirmed a load of the issues we'd seen before, plus revealed an additional wrinkle: it didn't really matter how many castles were in play (by this time I had experimented with four, five, and six), players felt incentivised to concentrate their efforts on just a couple of them, leaving the others uncontested.

I had an idea. If we had six castles, we could number them one to six, and then use dice in some way to control which castles could be affected by cards at any time.  I roughed out some rules based on this concept: each player has two dice which they roll, and the castles matching the die rolls can be played to by either player, but the player owning each die has a small combat advantage at the matching castle; dice get rerolled after battles.

Version 2a, now with added dice. I should change this blog name to "Later, add some dice."

So this had the effect that I had hoped for and resulted in the game spreading out more.  The downsides were that players (and I include myself in this) often missed which locations were allowed for play, forgot to reroll at the appropriate times, and sometimes felt that the dice made it difficult to plan beyond the next battle.  Notwithstanding these issues, I liked what the dice did for the game, so I'm planning on keeping them in for the time being and seeing if I can knock off the rough edges.

The other recurring issue was my combat system, which was based around a 3x3 table that gave an effect for each combination of the three tactical options chosen by the attacker and defender at a castle.  This was still causing more confusion than interest, despite me telling myself that it was pretty straightforward, could be streamlined with decent graphic design, and could be internalised after a few battles anyway.  Sometimes it takes a while to get something into (or out of) my thick head, but eventually I decided to scrap the table, write some simpler rules onto the tactic cards, and give it another go.  This was another definite improvement.  It lost a bit of the subtlety I was hoping for, but it's just a first pass and the tactic effects can certainly be improved later.

I am also not convinced about the supply cubes I have been using, and have some ideas about replacing them with cards, meaning that I will almost certainly need more cards in the game to compensate, but I'm feeling that about 100 cards plus four dice is probably a reasonable component list.

What now? I'm working on updating my prototype again, and have booked in for a playtesting session with this game at Dragonmeet next weekend: if you're there, I'm playtesting from 4pm, and helping out at the playtest zone for the earlier part of the afternoon.  Hopefully I'll get the game into a reasonable shape in time, and maybe even get a little sanity-check testing beforehand. I have my work cut out for me for the next few days...

The Castles of Comparatively Sane King Stephen

Another month, another new game... This one has been stewing in my brain for a little while now as I have been reading a book about the civil war in England (and Normandy, but the narrative in the book focuses on England) between King Stephen and Empress Matilda in the mid 12th century.  This is the period often known as The Anarchy.

There were a lot of twists and turns in the war before it finally reached a settlement which allowed Stephen to remain as king for the rest of his life (only about a year after the end of the war, as it happened), but to be succeeded by Matilda's son, who became Henry II.  Lots of material for a game in there, but the bit that caught my imagination was the period of comparative stalemate through a big chunk of the 1140's, which the author, Jim Bradbury, describes as the Castle War -- I have no idea if this is a widely used term. During this period, there were no major battles, but both sides built, besieged, and captured many castles in a shifting game of chess that Stephen seems to have done better at but, while Matilda ended up withdrawing to Normandy, it was not enough to stop resistance from her faction, or prevent the later campaigns of her son, Henry.  Much of this phase of the war also took place close to where I live (seemingly most of it within an hour's drive), making it of some local interest too.

I have finally reached the stage where I have a playable prototype, albeit one missing a few elements that I am hoping to add later, and also having some rules that are just a bit vague and woolly.  Still, my daughter, Miss B, played the game with me and helped decide a few rules that I wasn't sure about.  It looks like we do actually have the basics of a workable game, though the balance is certainly off, dynamics are ropy, and it could all collapse very easily.  This is fine, and where I wanted to be: it felt like we were actually playing a game.

Hand drawn cards for the win!
I actually had a load of fun scribbling bad pictures of motte and bailey castles.

The game as it stands is a pretty simple card game.  There are a row of cards depicting castles in the centre of the table, and the objective is to either control all of the castles, or to have the majority when the deck of cards runs out.  You play cards, which can be troops, leaders, or tactics cards, face-down on your side of the castles, and when you use a turn to draw fresh cards, you must nominate a castle to "resolve".  When you resolve a castle, all cards played by it are revealed, and a conflict either takes place or it doesn't, as a result of which the castle may change ownership, and cards may be discarded or returned to their player's hand.

This seems to work reasonably well at the moment, at least when both players are playing in the spirit of the game and not trying anything crazy.

So, with the proof of concept holding so far, what now?  The main elements I want to add in, taken from warfare of the period, are to consider "counter castles" (fortifications built to limit the operations of a more established castle, and to act as a base for siege), the sometimes shifting allegiances of the barons involved in the conflict, and prisoners and hostages taken from the opposition.  With all of these I need to be careful to not add too much complexity, as I want to keep the game fairly light and fast flowing, but if they can add to the theme and the strategic decisions available, I'll give it a try.

I also need to consider the overall form of the game. At the moment, the components are a small deck of cards and a few tokens to indicate the supplies held by a castle.  One possible issue is that there being a single deck of cards might lead to games where a poor distribution of cards might result in a massively unbalanced game, or one that is just boring.  It might be worth using separate decks for different types of cards, so players can access the types of cards they need, or perhaps each player could have their own deck.  All things to think about as I move forward...