No Such Thing as a Free Action

NOTE: This essay was originally posted as the second Design Diary on the Pax Pamir: Second Edition Board Game Geek page. This essay refers to an older version of the game. 

This diary isn't going to make much sense unless you are familiar with how the action structure works in Pax Porfiriana and the first edition of Pax Pamir. So, one might as well get through that right up front.

In Pax Porfiriana you get three actions each turn. And, each turn, you can pick from a menu of nine actions. Now, there are important qualifiers to these actions—it doesn't make much sense to take the “Get out of Jail” action if you are not, in fact, in jail. But, regardless, nine is a big number. To my mind, the number of actions, combined with the information flood of the cards opening market formed the one-two punch combo that knocked away many players from the design.

Now, I say that as someone who loves Porfiriana dearly. Indeed, if I were to try to develop a new edition of Porfiriana, I don't think I would change anything about the design. A few good player aids and a new card design would go a long way. But, when it came around to designing Pax Pamir, I knew that I wanted to help smooth over what could sometimes be a difficult entry-point. One way I did this was by adjusting the card layout. I'll talk about that more later. The main thing I did was to change the way actions worked in the game.

In the first edition of Pax Pamir, players always have access to a few core actions: purchase, play, and discard. Players will use these actions the entire game, and they probably take up the vast majority of all of the actions that a player will take over the course of the game. In this way, the game is similar to Porfiriana. But, those three actions weren't enough to represent all of the things I wanted players to be able to do. I was at a bit of a crossroads. I didn't want to inundate new players with options, but I also didn't want to rob players of interesting tools.

I fixed this problem by having cards “host” actions. Basically, by buying and playing cards onto your tableau, players gradually added to the number of actions they could take. During the first few turns of the game, players would go from having a few options to many. This made teaching the game much easier than teaching Porfiriana. First you buy cards. Then you play cards. Finally, you put those cards to use taking actions.

However, this simplification came with some downsides. Uncoupled to the standard sequence play, I was free to come up with lots of actions. One early draft of Pax Pamir had over a dozen. At the time, I thought I was actually making the game more elegant! Writing that sentence now, I realize it sounds a little ridiculous to think that adding actions made a game simpler. Let me explain what I mean.

Pax Porfiriana has a lot of rules that nest in one another. In fact, that's sorta Phil's design aesthetic. For instance, when you play a military unit you have to put it somewhere. And putting it somewhere means understanding movement costs, combat values, and retreats. In sum, the rules of combat and troop movement are actually buried inside of the play card action. When I was working on Pamir it seemed easier to instead to make combat its own special action, and to pull all of those rules out of the play card action.

This created another problem though. When I moved combat into its own action, I made an already tight action economy even tighter. The chief consequence of that was simply that players never took all of these cool actions. If purchasing and playing was the main way to get things done, then that's what players were going to do. For many games armies just kinda stood around staring at one another. In trying to fix Porfiriana, I had sucked the life right out of it.

To remedy this I introduced the idea of a “free” action. Basically, each action was coded to a particular regime. The ones that matched the current regime didn't count against your action limit to the turn. This made a lot of thematic sense (in anarchy it's easy to move armies around and fight). It also losened the design, allowing for multi-card combos and interesting board positions.

But, as any CCG designer will tell you, a “free” action is an invitation to ruin a system. So much of the game was so tightly balanced. Now that players could take actions without spending one of their two actions, all sorts of strange things would start happening to the game state. I got around this problem in two ways, first, the victory condition gave players a lot of weird ways to win the game, and the cards that controlled regime/climate always pushed against their own suit. A player could play an army that would give them powerful military actions and an edge in open war, BUT it would also likely switch the regime/climate to a different suit. Secondly, I put in attrition rules that made it so armies had a tendency to destroy themselves. So, even if the stars did align, it was difficult to go on a rampage and clear the board. It worked, and that system pretty much survived till the publication of the first edition.

The reason why I wanted to lay out the old action system in full here is because it took me a long time to realize some of the inherent flaws in that system. So much of the development of the second edition of Pax Pamir has been a study in interdependence. I would see one system I found inelegant, fix it, and then it would slowly dawn on me that the fix I had put in had triggered a cascading design failure. It wasn't simply that one change broke another element of the game. It was that one fix could break something that immediately broke a third and fourth thing.

So, when I decided to change the dominance system to only account for coalition blocks (formerly called empire cylinders), I ended up breaking the design of individual actions AND the whole action economy of the game. Here's how that happened. First, I decided that I wanted the victory condition to be easier to see. Now the game was merely a matter of blocks and players could easily figure out which coalitions were close to victory. Cool. However, the attrition rules now made the game grind to a halt, and dramatically unbalanced the suits and actions. The game dynamic shifted to something like trench warfare, which was utterly out-of-keeping with the period. To fix this, I made the attack action much more powerful, allowing armies to always decide what they wanted to destroy and removing the attrition rules entirely. This felt a lot closer to the period, as the limited responsiveness of armies pre-radio rewarded those that could take and manage initiative.

To stop the military suit from becoming too powerful, I divided the campaign action into “battle” and “march” actions. That way players had to telegraph their intents and defensive players could have some advantage. This worked pretty well, but, the game's climate clicked over to Military Struggle (making military actions “free”), the game quickly devolved from a thoughtful contest for supremacy into Afghanistan: The Arcade Game. Armies flew around with abandon, blasting each other to pieces and usually just rewarding the player who happened to take the first turn once the regime changed. Ugh.

The solution to this problem took a long time to recognize and implement. This was partly because Afghanistan: The Arcade Game was still fun and interesting. I wasn't even sure if this change in feel was a problem or if this was just how Pax Pamir: Second Edition was going to be. In fact, I even sent out review copies with this ruleset! But, when the reports came back, old fans of the series were less than enthusiastic. Many changes were liked, but there was something off about the game. Most were sure that the problem was with the change in victory conditions. This made sense as it was the immediate cause of many issues, but I wasn't ready to roll the changes back.

I watched game after game. Then, while talking to my brother, we started re-framing how we thought about the problem and the development of the second edition itself. Usually, when you try to improve something, you work small, to make sure that the core structure of the game remains sound. However, I needed to be thinking about how I could rebuilt the core to support the changes I wanted to implement. When it comes to my own design practice, I tend to think holistically. I want to bind everything together. This is usually not how a developer should approach their role. But, in this case, I had to lean into that design ethic, even as a developer. 

To that end, a new action system was born. Like the original game, players would get only two actions per turn, plus a number of “free” actions dictated by the climate. However, actions were no longer linked to as specific climate, cards were. So, in an Economic Boom actions on your economic cards are “free.” Players with good markets and economic assets are liable to get rich when the climate aligns. In this way the new system was quite like the old one.

Except, I decided to give actions to cards not wholly based on what suit the card was but instead based on the card's thematic subject. Some march actions remained on armies, but most are sitting on economic cards that represent troop supplies or hazardous mountain passes. Likewise, some battle actions can be found on armies, and other battle actions might be sitting on political cards or intelligence cards—because, after all, many of these figures were military leaders! Of course, a political leader with the betray action is useless unless he has a spy in the right position. Placing units of the right suit still matter a lot, but now the action system had a much stronger narrative element.

The new system opened up a big strategic space as well. Suddenly, players had to think about suit as they built their tableau. There are single-suit tableaus that enable you to take every action in the game. But, if the climate shifts away from you, you're going to find yourself struggling to keep up with players with more diverse interests. All sorts of combos and synergies opened up. And, for every synergy, a point of potential obstruction was made available to the other players. Regime/climate, an idea I was on the verge of dropping, became a means where players could help or hinder the potential actions of others.

All of this decoupling was enabled further by action consolidation. Going through the actions, I tried to break down some of the specificity of the previous edition's ten or so actions and boil them down to just six. Here more lessons from Root were implemented without too much loss of texture. With a few adjustments to combat, it was possible to make armies and spies battle in similar ways. So, when performing a battle, players simply pick a site of battle and then resolve the elimination of opposing pieces. That site might be a location on the map, with tribes and roads and armies or it might be a court card, with spies milling about hoping to find each other out. 

In most instances, the new action organization allowed me to sneakily build in many of the chrome-y rules of the first edition directly into the game's fundamental structure. Take the Baggage Trains rule for example. In the first edition, armies required a matching road in order to move. The Baggage Trains rule allowed players to spend a little money to ignore this requirement. The original rule was one of those things players easily forgot. It was also thematically murky. This is my own fault, as what was being represented was less a literal baggage trains than the spending of political capital (money) to move armies. In the new system, the march actions requires a road, unless you have a special ability (more on those later). However, if you rule either of the two locations on either side of the road, you can spend a little money to build a road, thus allowing your army to move. The thematic connections between political control and operational flexibility was rightly empathized, and a little rule was removed from the game without losing what it had tried to capture. 

Of the elements of the game that got adjusted in this way, the intelligence system was by far the most altered, and, I think, the one that gained the most nuance. But I'll say more about that next week.