[Part two of Trzewik's massive designer diary for Stronghold, originally published in thirteen parts on BGN and divided into two on BGG only due to the size limit on blog posts. Check out the first part here if you haven't already. Now, to wrap up this tale... —WEM]
#12 - Replayability
March 2009, Gliwice, a boardgame convention: I sit with Nataniel and Widlak from the Rebel.pl store, and we play Stronghold. They're testing. They're assessing. They'll tell me what they think. They'll say whether there's potential in here. They'll say whether it's worth putting a lot of cash into this game. After all, they work for the biggest games store in Poland, and they have a good perspective.
Nataniel is complaining about the catapults hitting with six, Widlak is commenting, the game runs in a cheerful atmosphere. "The catapults definitely need to change. I'm worried about replayability, the invader has this board with actions: catapult construction, ritual, dispatch, after a number of matches it's going to be a pain. Otherwise it's okay, this may become a hit," says Nataniel after the match.
Phew... I wasn't worried about replayability. I've been fascinated for years with the idea of the "plots" used in FFG's A Game of Thrones. Players use one special card from the plot deck, and this card changes a piece of the rules for that turn. Everyone receives more gold, for instance, or duels become more lethal, or no duels are allowed, etc.
I had similar plans regarding Stronghold, so a few days later I brought two sets of cards to the office, one for the Invader, one for the Defender. Each of them would draw three cards before the game, and they could use them anytime during the game, each time changing the game itself. "Poisoned water" in the castle caused the hospital to stop working. "Riots in the barracks" meant all the goblins in the Invader's camp were killed. I feel – I've always felt – that I'm strong plotwise. These cards showed the life inside both players' camps, events which were unexpected and demolished strategies and plans. Exactly.
The cards introduced chaos. They changed a serious, static strategy game into mutual incident swapping right out of a card game. A game about siege planning, a game about planning a castle defence, a serious game – and then bang, a card from the sky and your hospital is out. Bang, and a bleak fear falls on the men on the walls, and a few soldiers have to be sent back to the barracks... A discord. A dissonance. Wrong fairy tale.
This screwing each other up with cards didn't fit the game at all. Plotwise it was interesting, plotwise it gave the game a kick because it told about events in the castle and the enemy camp in a colorful manner, but in terms of game mechanisms it was artificial, awkward. If that was supposed to be a means to ensure replayability, well...it was bad. I took the cards home.
A new day, a new idea. The same set of cards. The same rule of each player drawing three of them, but with one change: You don't keep the cards in hand; instead they lay face-up on the table. The opponent sees what's up your sleeve. He can prepare, he can expect that at some point you may play one of them and slap him. Plotwise? Much weaker, for how do you explain that the castle Defender knows a "Fire in the Chapel" is about to come? Mechanism-wise it was a bit better, with no sudden bangs, but instead something you can prepare for.
And Stronghold-wise? In the case of this particular board game?
Another failure. The players would paralyze each other with cards. Instead of thinking Stronghold-like, as in previous games – building cauldrons and catapults, preparing for the fight – they would wheel and deal to defend themselves from the cards. Stronghold slipped into the background. The main course was trying to screw up the opponent with an event card. We're not going to play this way. I took the cards home.
A new day, a new idea. A new set of cards, specifically one set of cards instead of two. Instead of cards designed to screw up the other player, I now had a set of Events with a capital E. "Mighty Downpour" – the archers in the castle can't shoot; what's more, the roads have softened up and troop movement toward the castle costs an Hourglass more. These plot events affected both players, and plotwise it was splendid. The world around the castle has come alive: black clouds would gather above and the sun would shine through, armies of mercenaries would wander around, and a sound of distant earthquakes could be heard. Rule-wise it was good enough in the way that it didn't boil down to hitting the other player like in a card game. The events were huge – epic! – thwarting both players' plans.
And how did it affect Stronghold? Better than all previous solutions. Good enough for us to grind the game with these rules for over two weeks. Eventually, tired of the whole card mess, we gave it up.
Big events turned the game upside-down and made the whole work on game balancing worthless. Three months of balancing the speed of troop movement together with the archers' firing effectiveness on the walls went down in the mud as soon as we drew the "downpour" card. When we weakened the events so that they were interesting plotwise but without too big of an effect on the rules, we questioned why we bothered with them in the first place; they didn't ensure replayability at all because their effect on the game was minor. It was bad either way.
Days and weeks passed, with tens of versions and types of event cards created – they'll make a mighty interesting showcase one day – and we in the office felt that the cards were malfunctioning, that they were artificial. These cards were beside the game, beside its basic engine. I threw the cards away. Brain restart. I'm looking for a new idea. It's a day-to-day routine when working on a board game. Restart and again from scratch... Restart and again from scratch...
#13 - Replayability Replayed
I believe that every game has at its core a rule representing the concept of the whole game, a philosophy which can be presented in one or two sentences. Tens of lesser rules are built around it to enrich the fun. Citadels? "On your turn you're building a district; additionally you're using your character's special ability." Samurai? "On your turn you're positioning the troop tiles on the board, fighting for an advantage around the towns." Settlers of Catan? "On your turn, you roll dice to see whether you gained resources, then you trade them and use them to build roads, settlements and towns."
In Stronghold such a sentence would probably be: "On your turn you assign your troops to take actions like machine construction and troop training. Depending on the number of assigned troops, your opponent receives an equal number of action points for his defensive actions." The event cards I had tried to use to strengthen replayability were an addition to the actual game, not its integral part. I had to ensure replayability on the core level, the game's main part – not by using an extension. I'd wheeled and dealed. The solution was born eventually by damn accident.
The Invader's actions, as I've said before, were written down on a sheet with the Resources field, the Machine Construction field, the Rituals field, etc. This sheet annoyed me for various reasons, such as when a rule changed, because I then had to hand copy the whole sheet. Or a new rule would appear, so I had to rework the whole thing to make it fit the sheet. Or beta testers would whine that once a phase (e.g., Machine Construction) was completed, it couldn't be marked.
So I finally waved the A4 sheet, which was supposed to represent the Invader's camp, and split it into separate cards: the Resources card, the Machine Construction card, the Training card, the Rituals card, and the Dispatch card. This change allowed me to make quick modifications based on what came out of test matches, and the cards were much clearer for the players at the same time. They had six cards in front of them and could tick one as completed after each phase. It worked.
And then bang! A revelation came. I had the Machine Construction card together with a Catapult and Ladders in front of me: "Hmm, maybe I should do another, alternative card, with a Trebuchet and a Ballista...?" I looked at other phases like Training and Rituals. If I made the effort, I could think of more additional options for each of them. Then, before a Stronghold match, a player would draw one card for every phase. One time he would draw Catapults, another time it would be a Trebuchet. One time Saboteurs, another time Marksmen. One time Blood Stones, another time Possession. A different match every time.
Multidej was the first person to hear this new idea. He listened and immediately got enthusiastic about it. "It's like there are different commanders and engineers in the game. Their profiles could be drawn on the side, such as some bloodthirsty shaman with his murderous rituals, or a different shaman – the tactics master – with completely different rituals on another card... These cards could be given a human face to show there are different commanders with different ideas for war..."
Oh, yes, Multidej fell in love with this idea straight away. And I fell in love with his vision of different people representing different concepts and attack philosophies. You draw an engineer who believes that the castle should be taken by building trebuchets, or one who likes to build ballistas. Superb. Arduous fun had begun...
Phase one, resources. I took coloured pencils and went wild. I drew a lake on one card, a thick forest on another, a forest and quarries on yet another. It would represent the setting for the siege, one time by the lake, another time in the forests... Each card would give different resources, one time a player would have lots of wood, another time a bit of wood and stones, or lots of wood and food. This set-up would determine the whole course of the game.
Similarly the second phase, War machines, was changed. Ballistas, Trebuchets and Ladders were added to the already existing Catapult. The work took long weeks. I created seven different options for one of the phases, seven different war machines, seven types of training, rituals, seven different dispatch types. I divided them into basic and special. Basic, like the Catapult or the Saboteurs, would appear on three out of five Invader's cards. Special ones, like the Trebuchet or the Quartermaster, would appear on only one card. Those additional options appeared gradually. Their numbers were increasing. One time the testers were playing with a battering ram, another time with a siege tower. At one point someone moaned: "C'mon, I didn't draw any Saboteurs. How am I supposed to win?!"
Separate matches started to differ significantly from one another. There was a moment, sometime in early June, when at last I started hearing what I had wanted to hear for long months: "Shit, unlucky. I made a few mistakes in this match. I would play it differently now, and I really want to play again."
I smiled and said, "But you do know that each phase has five different cards, so you're sure to have a completely different set next time? Today there was no siege tower, there were no saboteurs, no bloodstones in the game. There are seven different machines, seven different equipment types, seven types of training in the game... There's a lot for you to learn; you need to learn how to play with the Towers, the Ballistas, how to use the Battering Ram... You will be making mistakes in many coming matches, before you learn it all."
Eyes wide open. Dreamy face. A question: When's the rematch? I had worked for that for seven long, tough months. I was in heaven.
Stronghold demo table at Spiel 2009