Back in the early days of Oath’s design, I made two structural decisions that I never really questioned. The first determined the size of the game’s deck, roughly how many cards were in play, and the general size of the game’s initial card-base (as well as its rate of change).
The second choice had to do with the players. I had had a mixed experience with legacy games in my own playgroups. The biggest problem was that they demanded a consistent set of players to get together for all of the episodes. Oath was, in many ways, a reaction against those games. It was also a reaction to those games in a very specific way: I didn’t want the game to write any kind of history for the players—that was their job. The world’s development could be handled by the players. Family history was a strictly personal matter.
This was a mistake.
Because I was so worried about the liabilities that legacy player positions might introduce into the design, I robbed the game of a set of tools that could have helped player situate themselves in the game’s world and develop a sense of ownership over their own actions. I always imagined players would develop vendettas and form alliances on their own and that the game didn’t need to mediate those relationships outside of a few systems (like citizenship). But, as with so many things in tabletop design, if the game doesn’t enforce a behavior, there’s very little guarantee that the players will.
So, from the very start of this project, I started looking at the question of what it would mean to let players develop family histories. I hoped that by giving players a bit more sense of personal (and familial) progression, I could help them extend the strategic horizon of their thinking beyond the limits of a single game. Certainly my children helped me change my perspective about the world! At the same time, I knew that questions of personal progress extremely well-explored in gaming. So, to limit my own investigation, I drew up a few core principles which would help me keep my own work focused on the peculiar problems and opportunities of Oath.
For instance, it was important that the system not replicate the kinds of character progressions that are common in the role-playing space. I didn’t want steady linear progress. Players might unlock powers, but there would be costs associated with them. Here, I was thinking of the many ways that the status effects (and even the wealth system) informed a player’s position in the old game Tales of the Arabian Nights. In that game, if you amassed great wealth, you would gain certain advantages, but your fortune would create drawbacks, including changing how your character could move through the world.
This led me to my second critical concept: scale. Families, generally speaking, don’t live as long as other institutions. To capture this difference, I wanted them change at a faster rate than the rest of the game world. A family might grow to be quite powerful over a couple generations and then disappear into obscurity. Generational constructions of history often operate in fairly short cycles (e.g. “the first generation creates wealth, the second maintains it, and the third squanders it”, or “hard times create strong leaders who create easy times that create weak leaders who create hard times”). I don’t doubt that there is some wisdom to be gained by thinking about history in this way, but these cliches have always struck me as a little simplistic and prone to opportunistic framing. If I usurp a ruler, I’m certainly going to characterize him as weak in comparison! But, from the perspective of Oath and it’s many slow changes, it seemed like introducing a system that had shorter cycles could help broaden players strategic thinking on the scale of 1-3 games in the future.
Finally, both of these principles would be subject to a critical restriction. I had very little space both physically and mentally in which to work. Oath is a big game with a lot of special effects that players need to keep in their mind. For this reason, the game doesn’t have traditional player hands or a Pax-style card market. These limitations are a big reason why the majority of the expansions new material is used only in-between games. The chronicle phase has plenty of room for cuts and growth. But, when it came to in-game systems, I knew that I couldn’t ask players to do much more than they were already doing. With this restriction in mind, I budgeted myself a small strip of territory beneath each player’s player board. This was roughly enough space for 4 or 5 relic-sized cards. Whatever system I designed would have to sit in this space or less during play.
If you want to see me talk at length about Oath's physical and mental constraints, check out this talk I gave as part of the Lake Superior Design Retreat back in 2022. I should say that this was was addressed to a room full of architects, so the first half is mostly about offering a broader context for the work, so you can all safely skip it.
This led me to my second critical concept: scale. Families, generally speaking, don’t live as long as other institutions. To capture this difference, I wanted them change at a faster rate than the rest of the game world. A family might grow to be quite powerful over a couple generations and then disappear into obscurity. Generational constructions of history often operate in fairly short cycles (e.g. “the first generation creates wealth, the second maintains it, and the third squanders it”, or “hard times create strong leaders who create easy times that create weak leaders who create hard times”). I don’t doubt that there is some wisdom to be gained by thinking about history in this way, but these cliches have always struck me as a little simplistic and prone to opportunistic framing. If I usurp a ruler, I’m certainly going to characterize him as weak in comparison! But, from the perspective of Oath and it’s many slow changes, it seemed like introducing a system that had shorter cycles could help broaden players strategic thinking on the scale of 1-3 games in the future.
Finally, both of these principles would be subject to a critical restriction. I had very little space both physically and mentally in which to work. Oath is a big game with a lot of special effects that players need to keep in their mind. For this reason, the game doesn’t have traditional player hands or a Pax-style card market. These limitations are a big reason why the majority of the expansions new material is used only in-between games. The chronicle phase has plenty of room for cuts and growth. But, when it came to in-game systems, I knew that I couldn’t ask players to do much more than they were already doing. With this restriction in mind, I budgeted myself a small strip of territory beneath each player’s player board. This was roughly enough space for 4 or 5 relic-sized cards. Whatever system I designed would have to sit in this space or less during play.
If you want to see me talk at length about Oath's physical and mental constraints, check out this talk I gave as part of the Lake Superior Design Retreat back in 2022. I should say that this was was addressed to a room full of architects, so the first half is mostly about offering a broader context for the work, so you can all safely skip it.
A Failed Approach
It took about a month of rough design work to produce a system that worked within the principles I outlined above. However, this system ultimately proved to be a failure. Normally, I wouldn’t the details of a failed system—these design diaries are long enough!—but, almost every concept in the current working system has its genesis in this system. I thought it would be a good exercise to fully outline those failures, so that you could all get a sense of how sometimes a good idea can at first seem like anything but.
My initial impulse was do have the lineage system work as a deck builder. This mechanism is good at showing gradual change and also creating game-to-game variance. To test the idea, I drafted something like 20 or so different “trait” cards and printed them in triplicate so that I could see how the different decks felt.
Quickly I encountered an obvious problem. A deck’s identity in a deck builder manifests itself very slowly, usually over the course of several full rotations through the deck. If players were going to only have a couple active traits each game, then it would take a long time to cycle through a deck. On the other hand, if the deck was fully shuffled each game as new traits were introduced, players would have a harder time forming a sense of ownership over their families.
To fix this, I decided to add an element to the mix. I designed a character sheet that would help power the deck-building system. The basic idea was that the sheet would allow players to develop different ways of engaging with their decks. This stuff only mattered when the deck was being referenced, so it seemed naturally suited to a slim piece of paper that could be marked up, then tucked behind a player board when it wasn’t in use.
It took about a month of rough design work to produce a system that worked within the principles I outlined above. However, this system ultimately proved to be a failure. Normally, I wouldn’t the details of a failed system—these design diaries are long enough!—but, almost every concept in the current working system has its genesis in this system. I thought it would be a good exercise to fully outline those failures, so that you could all get a sense of how sometimes a good idea can at first seem like anything but.
My initial impulse was do have the lineage system work as a deck builder. This mechanism is good at showing gradual change and also creating game-to-game variance. To test the idea, I drafted something like 20 or so different “trait” cards and printed them in triplicate so that I could see how the different decks felt.
Quickly I encountered an obvious problem. A deck’s identity in a deck builder manifests itself very slowly, usually over the course of several full rotations through the deck. If players were going to only have a couple active traits each game, then it would take a long time to cycle through a deck. On the other hand, if the deck was fully shuffled each game as new traits were introduced, players would have a harder time forming a sense of ownership over their families.
To fix this, I decided to add an element to the mix. I designed a character sheet that would help power the deck-building system. The basic idea was that the sheet would allow players to develop different ways of engaging with their decks. This stuff only mattered when the deck was being referenced, so it seemed naturally suited to a slim piece of paper that could be marked up, then tucked behind a player board when it wasn’t in use.
I imagined that families could basically develop along 4 different paths. The first path, ardent, would give them more traits, allowing them to use more than the normal number of abilities per game. The seeking trait would give them a wider selecting, essentially increasing their draw. The diligent trait would give them a baseline power that would be useful outside of the trait system and the covetous trait would allow them to keep relics from game to game.
This approach generated two problems. First, I needed some way for a player to earn these abilities and, potentially, to add new traits into their deck. While the language of Oath gave me plenty of terms of to use (burning favor for instance), I found that I wanted something else that was less disruptive to the game’s economy. This lead me to the concept of Influence.
For all that New Foundations adds to Oath, there are very few fully new concepts that have been added to the core rules. Influence is one of those new systems. A players influence is a value equal to their favor + the value of their advisors. Each advisor’s value is linked to the favor in its suit’s bank. This was a concept we actually used during testing for an alternative People’s Favor that we couldn’t get to work back in 2020 (but which we eventually did revisit as I’ll talk about in a few weeks).
Using this concept, I was able to design a much wider range of goals which could reward players for engaging in the game in ways beyond victory. For instance, players might gain upgrades on their family sheet by demonstrating that they had reached a certain influence goal (say, 6 influence in discord could mean having a single discord advisor with a bank of 6 or 3 discord advisors with a bank of 2). We played around with these sorts of goals and found that they created a lot of interesting incentives that really opened up the game’s economy, especially in the final act of the game.
The second problem was that the families didn’t have a way to collapse. The same character progression sheet that helped players form a more meaningful connection between players and their families also tended to flatten family differentiation and lead to families that, frankly, we’re just too stable. I needed some kind of chaos agent.
To answer this call, I introduced a new type of trait: the curse. Curses were essentially reverse traits. Instead of costing resources to unlock, they would give players resources to take on, but, in future games, players would have to deal with the consequences of their choice to take the curse. With curses, I was able to introduce a deck integrity check. If a family’s lineage became too cursed, the family would collapse. It’s sheet would become unusable and players would be free to burn it or bury it as their tradition might dictate.
Taken together, this system worked. And, even in its clumsy and overwrought form, it started giving players curious incentives outside of the regular victory conditions. Do I try to become a citizen so that the site I cultivate survives until the next game, or should I try to improve my family’s wealth so I can store an extra relic in my deck?
At the same time, I was tremendously disappointed. For one, the whole thing was too complex. Players had maintain a deck of cards, a paper sheet, and, when cards were added, some kind of market or draw would be needed. (I haven’t bothered to explain how traits were added to families because I went through a few systems and hated all of them—even the ones that worked.) And, on top of that, the system was just too slow. The deck building took too long to express itself.
So, after a few weeks of work, I put the system back into design. I started by just stripping it down to its bare minimum. Maybe, if the deck was a fixed size, it would allow player identity to better form. Perhaps the lineage sheet was a mistake. Then, while playing the game with Clay and Andrea one day, we hit upon the essential problem. Clay at worked hard to unlock a new trait. At this point, we had moved the unlock conditions to the traits themselves, hoping that by linking the requirement and the reward players would form more of a relationship with their lineages. At the end of the game, he asked me if he’d get the trait the next game, to which I replied, “Maybe.” He gave me one of those looks of confused disappointment you try to surprise when you taste some badly cooked meal. I immediately realized that I had made a critical misjudgment.
I hadn’t appreciated the degree to which a family-driven scale was, in fact, the overriding design principle for the lineage system. Families save money so their kids can go to college—not so their kids or grand kids or great-grand kids or great-great-grand kids just might go to college. The system needed a much higher level of positive (and negative) feedback. With that in mind, I started over.
A New Lineage
The basic problem was a mismatch between the deck-building system and the number of traits I wanted to have active each game. I imagined players would only have one or two active traits. This meant that the deck-building system wasn’t providing really any sense of player identity because so few cards were being used from turn to turn. Imagine playing Dominion if you’re hand size was only a couple cards! (This actually ended up being a critical realization that helped me figure out how to get the governance system working, which I’ll talk about in a future diary).
I was worried about introducing too many traits into a game because, going back to my third design principle, I knew that Oath had very little extra space for rules or components. The deck-building had essentially offered a me a way to hide a player’s lineage. This was good in terms of the game’s overall complexity budget, but it also meant that the lineages were being hidden from the players as well.
What was needed was something more direct. So, I started with a very simple rule: if a player earned a lineage card, they would get to use it in the next game. That was easy enough. But, it came with a new problem, I needed to introduce a source of downward pressure to stop players from creating unbreakable lineages. Sadly, I didn’t have the design space to introduce bad marriages and wastrel sons!
This approach generated two problems. First, I needed some way for a player to earn these abilities and, potentially, to add new traits into their deck. While the language of Oath gave me plenty of terms of to use (burning favor for instance), I found that I wanted something else that was less disruptive to the game’s economy. This lead me to the concept of Influence.
For all that New Foundations adds to Oath, there are very few fully new concepts that have been added to the core rules. Influence is one of those new systems. A players influence is a value equal to their favor + the value of their advisors. Each advisor’s value is linked to the favor in its suit’s bank. This was a concept we actually used during testing for an alternative People’s Favor that we couldn’t get to work back in 2020 (but which we eventually did revisit as I’ll talk about in a few weeks).
Using this concept, I was able to design a much wider range of goals which could reward players for engaging in the game in ways beyond victory. For instance, players might gain upgrades on their family sheet by demonstrating that they had reached a certain influence goal (say, 6 influence in discord could mean having a single discord advisor with a bank of 6 or 3 discord advisors with a bank of 2). We played around with these sorts of goals and found that they created a lot of interesting incentives that really opened up the game’s economy, especially in the final act of the game.
The second problem was that the families didn’t have a way to collapse. The same character progression sheet that helped players form a more meaningful connection between players and their families also tended to flatten family differentiation and lead to families that, frankly, we’re just too stable. I needed some kind of chaos agent.
To answer this call, I introduced a new type of trait: the curse. Curses were essentially reverse traits. Instead of costing resources to unlock, they would give players resources to take on, but, in future games, players would have to deal with the consequences of their choice to take the curse. With curses, I was able to introduce a deck integrity check. If a family’s lineage became too cursed, the family would collapse. It’s sheet would become unusable and players would be free to burn it or bury it as their tradition might dictate.
Taken together, this system worked. And, even in its clumsy and overwrought form, it started giving players curious incentives outside of the regular victory conditions. Do I try to become a citizen so that the site I cultivate survives until the next game, or should I try to improve my family’s wealth so I can store an extra relic in my deck?
At the same time, I was tremendously disappointed. For one, the whole thing was too complex. Players had maintain a deck of cards, a paper sheet, and, when cards were added, some kind of market or draw would be needed. (I haven’t bothered to explain how traits were added to families because I went through a few systems and hated all of them—even the ones that worked.) And, on top of that, the system was just too slow. The deck building took too long to express itself.
So, after a few weeks of work, I put the system back into design. I started by just stripping it down to its bare minimum. Maybe, if the deck was a fixed size, it would allow player identity to better form. Perhaps the lineage sheet was a mistake. Then, while playing the game with Clay and Andrea one day, we hit upon the essential problem. Clay at worked hard to unlock a new trait. At this point, we had moved the unlock conditions to the traits themselves, hoping that by linking the requirement and the reward players would form more of a relationship with their lineages. At the end of the game, he asked me if he’d get the trait the next game, to which I replied, “Maybe.” He gave me one of those looks of confused disappointment you try to surprise when you taste some badly cooked meal. I immediately realized that I had made a critical misjudgment.
I hadn’t appreciated the degree to which a family-driven scale was, in fact, the overriding design principle for the lineage system. Families save money so their kids can go to college—not so their kids or grand kids or great-grand kids or great-great-grand kids just might go to college. The system needed a much higher level of positive (and negative) feedback. With that in mind, I started over.
A New Lineage
The basic problem was a mismatch between the deck-building system and the number of traits I wanted to have active each game. I imagined players would only have one or two active traits. This meant that the deck-building system wasn’t providing really any sense of player identity because so few cards were being used from turn to turn. Imagine playing Dominion if you’re hand size was only a couple cards! (This actually ended up being a critical realization that helped me figure out how to get the governance system working, which I’ll talk about in a future diary).
I was worried about introducing too many traits into a game because, going back to my third design principle, I knew that Oath had very little extra space for rules or components. The deck-building had essentially offered a me a way to hide a player’s lineage. This was good in terms of the game’s overall complexity budget, but it also meant that the lineages were being hidden from the players as well.
What was needed was something more direct. So, I started with a very simple rule: if a player earned a lineage card, they would get to use it in the next game. That was easy enough. But, it came with a new problem, I needed to introduce a source of downward pressure to stop players from creating unbreakable lineages. Sadly, I didn’t have the design space to introduce bad marriages and wastrel sons!
Here I hit a bit of a lucky break. At one point when I was working on the unlock conditions, we moved from bespoke conditions to little quest cards. Each game, players were given a new “immature” trait which they could earn by finishing a quest. The quest card would be placed on top of the trait to show it was mature. When you completed a quest, you simply discarded the quest and then flipped over the trait, showing that it would be secured for the next game. This system replaced the little quest conditions I had written for each trait, and I was sad to miss that customization. But, Arcs had taught me the value of random elements. In Arcs, while certain elements of your plotline are linked to your performance, other elements—such as your future plotline choices—remain random. This allows the trope of the chance encounter to sneak into the game’s narrative logic and creates some very silly and sometimes poignant twists of fate. The quest conditions likewise were creating lovely little narratives turns.
Anyway, those little quest conditions were still in the spreadsheet, so they could be repurposed to “maintenance costs.” During a players turn, they could take a minor action to maintain any trait by meeting its condition or paying its cost as the case may be. This means they would lose the advantage of the trait. Critically, any trait that was not maintained would be lost. So, if you wanted to enjoy the benefits of your trait without attending to your families traditions, they would be lost to time.
To this system I added a new curse system. Players would always have exactly one curse card. If a player began the game without a curse card, they would be dealt one. Curse cards have two sides: the offer and the consequence. The offer side is harmless, and, if the player does not succumb to the temptation of taking it, it will be discarded at the end of the game. However, if you take the curse card, you will flip it over to is consequence side. This will introduce some gameplay restriction that will haunt the player until they fulfill the card’s break condition.
Taken together, this approach was dramatically stripped down compared to earlier versions but generated a very punchy lineage system that let players build powerful legacies over just a couple games. At the same time, if you entered the game with a ton of advantages, you’d have to make some hard choices about which advantages you would spend time maintain and which you would let erode. There was also a critical baseline. Players would always begin the game with one new immature trait that would be covered by a quest card. You’re never more than a game from unlocking your first special power. Of course, incomplete quests get discarded between games, so we could avoid the deluge of sidequests that seems to drown my characters in most open world games (and my interest in completing them).
Even in their current, rough form, this system creates interesting incentives that change how players think about the late game and their relationship to their own player positions. We’ve found that this actually goes a long way in untangling some of the analysis paralysis that can sometimes badly slow down the game’s final turns.
Well, that's it for now. I hadn't planned on writing quite so much for this entry, but I wanted to make sure I could take you all through some of the design work that's gone into the new expansion. You can follow along with the Kickstarter here.
Anyway, those little quest conditions were still in the spreadsheet, so they could be repurposed to “maintenance costs.” During a players turn, they could take a minor action to maintain any trait by meeting its condition or paying its cost as the case may be. This means they would lose the advantage of the trait. Critically, any trait that was not maintained would be lost. So, if you wanted to enjoy the benefits of your trait without attending to your families traditions, they would be lost to time.
To this system I added a new curse system. Players would always have exactly one curse card. If a player began the game without a curse card, they would be dealt one. Curse cards have two sides: the offer and the consequence. The offer side is harmless, and, if the player does not succumb to the temptation of taking it, it will be discarded at the end of the game. However, if you take the curse card, you will flip it over to is consequence side. This will introduce some gameplay restriction that will haunt the player until they fulfill the card’s break condition.
Taken together, this approach was dramatically stripped down compared to earlier versions but generated a very punchy lineage system that let players build powerful legacies over just a couple games. At the same time, if you entered the game with a ton of advantages, you’d have to make some hard choices about which advantages you would spend time maintain and which you would let erode. There was also a critical baseline. Players would always begin the game with one new immature trait that would be covered by a quest card. You’re never more than a game from unlocking your first special power. Of course, incomplete quests get discarded between games, so we could avoid the deluge of sidequests that seems to drown my characters in most open world games (and my interest in completing them).
Even in their current, rough form, this system creates interesting incentives that change how players think about the late game and their relationship to their own player positions. We’ve found that this actually goes a long way in untangling some of the analysis paralysis that can sometimes badly slow down the game’s final turns.
Well, that's it for now. I hadn't planned on writing quite so much for this entry, but I wanted to make sure I could take you all through some of the design work that's gone into the new expansion. You can follow along with the Kickstarter here.