For a guy who ostensibly thinks dungeon-crawling is stupid, I sure have played a lot of games in that genre of late. The latest is Pathfinder: The Adventure Card Game, a cooperative game from Paizo Publishing, and it's not bad.
The game is a quite faithful port of the Pathfinder/D&D3 roleplaying experience, minus the actual roleplaying (which is traditionally optional anyway). You've got a character with strength, dexterity, wisdom, and so on, each rated as a die size (d4, d6, d8, etc., but ironically not a d20). Encounters (which can be monsters, barriers, allies, treasures) have a target number which you need to beat to successfully navigate. You can play cards from your character's personalized deck and use your inherent special powers to boost your skills, and occasionally your allies can help you. Track down and kill the episode's Villain, usually cutting a swathe through his or her Henchpeople on the way, and you win. After the game, you can rebuild your character's deck using cards you've acquired during the adventure to make him or her more potent next time.
The mechanics of this are simple and nicely done, but not particularly noteworthy. What I think is interesting is looking at how the game approaches the question of how to balance narrative scripting against gameplay variability.
Cooperative games usually need to provide some kind of narrative experience to be successful; they can't just be intellectual puzzles. There are obviously a lot of ways to do this, but the general idea is to give the series of challenges the players must overcome (and rewards they receive for doing so) some sort of structure designed to engage them. This can be entirely narrative, with the challenges having some attached title or flavor text which is read aloud with the story becoming emergent as the texts are read (as long as they are coherent enough that players can improvise logical connections). Or the structure can be much more constructed and explicit, with challenges and rewards designed and ordered to produce an intended overall emotional story arc.
Examples of games which use the first idea are easy to find; successful examples include Robinson Crusoe, Tales of the Arabian Nights, and Ghost Stories (or Arkham Horror, Shadows over Camelot, or Defenders of the Realm, if you consider those games good). You have a huge supply of little storylets, which are pulled out more or less randomly and translated into game-mechanics form. A windstorm hits (reducing your shelter level), your lack of Courtly Graces offends the nobility (and you become Scorned), or whatever. As they are read they form a timeline you can create a story out of.
This has the gameplay advantage of making the tasks you are facing varied and unpredictable, and differ greatly from game to game. It also allows the players to do their own storytelling when the events remain within the bounds of the somewhat plausible. The huge disadvantage, as anyone who has a basic understanding of literature or music will tell you, is that we have a pretty good understanding of how compelling narratives are built, and this is most definitely not it. Stories have build-up, carefully managed cycles of tension and resolution, anticipation, and suspense. None of which you can reliably do if you’re just pulling random storylets.
Still, I think there is nothing inherently wrong with this way of doing things. For example, while Nuclear War or Fluxx aren't particular good games by 2014 standards, they do have delightful anti-establishment or satirical aesthetics that are both completely coherent and tied up with their total randomness (and, it bears mentioning, their brevity). Or a game like Once Upon a Time, where the players' attempts to create signal out of noise and find ways to creatively link events is what the game is. So clearly it's possible to do great work this way. But it's also an easy and unfortunate default pattern when a designer is unskilled, or when a game doesn't have a strong creative vision or anything particular to say. If you look at a big and intricate game like Battlestar Galactica, where the fictional world it's designed to emulate has a clear authorial style, it's hard to see the merit in having the players interact with a simple, random, unstructured throughline.
The opposite end of the spectrum is Knizia's highly structured Lord of the Rings. Here, the story events and the challenges associated with them are laid out in a strict order. You're going through Rivendell to Moria to Rohan, and that's all there is to it. You face the same challenges (and narrative elements) in the same order each game. There is this still quite a lot of randomness in the timing of the events and resource flows, as random draws from a bag of tiles trigger various game elements, but the story events that drive the narrative are scripted.
This strong structure gives the gameplay itself the ebb and flow required to make the story engaging. The designer can directly tweak and manage the flow of challenges and rewards to manipulate the moment-to-moment game tension, hopefully giving us both high-tension action scenes and rewarding us with moments of rest and refresh after we get through. This can, when well executed, give us a far more visceral engagement with the game because it goes after our emotions very directly. Pandemic does the same thing: the structured way the decks are manipulated (pre-stacking the player deck, the stacking and re-stacking of the infection deck) alternates high-risk and high-tension periods where you are firefighting crises with lower-risk infrastructure-building and research-gathering periods.
Even though for all these reasons I think of the structured narrative as "the right way" and the random event firehose as "the wrong way", in truth it's a continuum and structure is certainly not an end in and of itself. The goal is to modulate the players’ sense or risk, to feed the dread of anticipation and allow the relief and accomplishment of a challenge faced down, and that requires both a degree of predictability as well as significant risk and therefore uncertainty. Clearly you can go too far in trying to organize your narrative – making the story predictable and boring – just as you can make a game too random and disjointed. It wouldn't be hard to argue that Britannia, for example, is too well-organized and that it needs more uncertainty to maintain tension. My experience though is that cooperative or narrative-driven games almost never err on the side of being too structured.
The interesting thing about Pathfinder is that from the long view it resembles classic, unstructured, firehose-driven games. You have a box containing a very large number of cards that the characters can encounter, and you randomly pull some of them out and deal them into piles at different locations to explore. When you explore, you just draw a card from a location deck and do what it says, with perhaps minor assistance from the other players. The Villain is dealt into one of these piles at random and you just need to plow through the decks to hunt him down. If your goal is hunting the bad guy, there are no percentages in going to the Apothecary before you hit the Treacherous Cave; the Villain is equally likely to be anywhere. It’s eerily similar to Arkham Horror's "go to a location and random stuff happens for no particular reason".
But Pathfinder combines straightforward gameplay with just enough structure to make decision-making and task allocation interesting and have a real but measured sense of risk. Each location has a clearly specified mix of cards that go into the deck: monsters, barriers, weapons, armor, spells, items, and allies. The mix is listed on the top of the location card, where you can always look at it and know what you’re getting in to. So unlike in Arkham Horror, when you go to a location you have a pretty clear idea of what you might get out of it and which character is best suited for the challenges it might present (the Thief for the location with the barriers, the Fighter for the place with the monsters, the Sorcerer for the place with the allies). Still, while the Fighter may be the best person to take on the monsters in the Desecrated Vault, there is still usually the possibility that he’ll run into a barrier or trap that’ll hose him, so there is almost always still some risk. And there are balancing factors; maybe you really need to find a better a weapon, so a trip to the Garrison is worth the risk of facing monsters. More likely, you don’t have a character who is ideally suited to exploring a location, but someone has to do it, so you need to figure out who is going to sign up for the increased risk (because you always have to face the card you draw, teaming up is actually not particularly useful). Additionally, once locations have been cleared of Henchpeople, they need to be “closed”, secured against the Villain’s return. This involves another test, and the character best suited to exploring the location may well not be ideally suited to closing it. Opportunities to close a location are infrequent and valuable and you want someone who is able to do it there when the opportunity presents itself, which is another matter of risk management. This all adds up to a significant amount of nuance and randomness, but because the general contours are spelled out and what needs to be done is clear, it’s interestingly tractable. You always know what you need to do to make forward progress, and you can make judgements about risk and reward that can pay off or not.
However, what this structure doesn’t do is give you any overall sense of pacing or drive. Some locations are more dangerous than others (sometimes significantly so, often not), but the game never modulates its moment-to-moment tension. You’re never forced to run the gauntlet before you want to or go into panic defense mode, nor are you given a moment of respite to recover and gear up after facing something particularly dangerous. Pathfinder's time pressure is just a 30-turn clock you need to beat – an arbitrary, inorganic limit. Compare to Pandemic, with its beautifully organic ebbing and flowing threat and pressure, where you need to win before the diseases do. By comparison, Pathfinder just has a time limit because if it didn’t there would be no game. Given Pathfinder’s source material this is fine, time just isn’t a dimension of traditional D&D stories; for structural reasons D&D-style RPGs in general have a difficult time managing time as a storytelling pressure. But this is a boardgame, not an RPG, and there is no need to be bound by a stricture of the original format.
Interestingly for a game that lacks any kind of strong overarching narrative, Pathfinder eschews any sort of explicit textual elements. Cards have illustrations and more or less descriptive titles but no flavor text. There are no “event” cards which add dramatic twists or change the rules or environment. The only real explanation of what you’re trying to accomplish comes up front, when you select the adventure to go on and get a few perfunctory sentences of flavor on the card that also outlines the locations, Villains, and Henchpeople involved (location cards also have some static descriptions, but they are in practice invisible because they're on the back of the card). This makes the experience somewhat generic. The box says it’s the “Rise of the Runelords Base Set”, with the “Rise of the Runelords” being the long-form adventure arc which wends its way through the base game and 5 expansions. But there is no sense that this is taking place in anything other than just a generic D&D fantasy world. If the premise of the story is that there are Runelords and they are rising, the game doesn’t exactly go out of its way to fill you in on what’s up with that.
What this long arc does capture, though, is very distinctively D&D: the slow, grinding out of improvements to your character and his or her equipment. You wade through monsters and challenges and maybe you’ll find a longsword to replace your short sword. The upgrades to your character and availability of new cards to add your deck are sporadic; after four games, you may have a better weapon or one more spell, or you may have basically the same deck you started with and one minor skill improvement. After you make it all the way through all the adventures your character will have accomplished quite a bit in the end, but that will be a lot of hours of gaming and the rewards for risking death each time out are very incremental. That’s fine, it’s the D&D tradition, but in the context of a boardgame it feels wrong. If this is the route we’re going to go, I’d like more intense pacing. Personally, I’d much rather have multiple, complete 6-episode arcs which have a quick pace and you can play a character through and then move on to the next story with a fresh character. The 36-episode monster arc just seems like a huge time sink. This feels to me like a back-port from MMORPGs, and not really appropriate for a boardgames.
Still, when all is taken into account I do like Pathfinder: The Adventure Card Game: Rise of the Runelords well enough. The pacing of the long adventure arc is probably too slack to keep me interested for that long, but the individual adventures are playable, quick, simple, and are structured well enough to provide both meaningful decisions and some tension. It’s certainly not in the same league as Robinson Crusoe, Pandemic, or Lord of the Rings, but you can’t play those games all the time and some of them require a significant energy investment while Pathfinder is more lightweight. Besides, D&D is more than a game now, it’s become something of a cultural touchstone. While the Pathfinder Adventure Card Game may not exactly be a work of game design brilliance, it is a workmanlike game that has a huge weight of tradition behind it.