Narrative Arc in Boardgames

In game design, mechanics on September 23rd, 2010 at 11:26 am

The concept of narrative arc in board games was (to my knowledge) first described by Jonathan Degann in an essay he wrote for the (sadly moribund) Games Journal as part of his Game Design 101 series. It didn’t find a lot of traction for some reason, though, and it doesn’t come up much in the analysis of board and card games. I think it’s an important analytical tool so I thought I’d dredge it up from the depths, clean it off, and see if I can offer a few extra notes about it.

The Nature of Narrative Arc

Over the course of many boardgames, the nature and feel of the decisions you have to make changes. This evolution is the game’s narrative arc. The classic example of this is Chess, familiar to millions from the body of chess literature. There are three major phases of the game, with a turn or two of transition between the phases:

  • Opening – players focus on activating and coordinating their pieces, primarily by opening lines they can act along.
  • Middlegame – the main struggle, where players maneuver to enhance their advantages (in a closed position) or engage in open warfare (in a more open one) or first one, then the other.
  • Endgame – the armies have been decimated, and the game settles down into a (typically) slower and more incremental battle of maneuver until one player manages to put the other player’s back to the wall or both players realize they lack the tools to force a decision.

Because of the Chess’s victory condition, this arc can be cut off suddenly at any point if one player leaves himself exposed to a sudden checkmate or loss of a major piece. In the absence of such blunders, though, this same arc is followed, even though the different stages may take different lengths of time each time you play it.

Go follows a superficially similar arc – opening, middlegame and endgame – but the meat of those three phases is quite different:

  • Opening – players sketch out a broad positional framework – typically building in the corners first, then along the sides and finally into the centre — which will be the home base future strategies are launched from
  • Midgame – the players try to establish real control of the territories they maneuvered toward in the opening while simultaneously trying to constrain the other player’s gains
  • Endgame – crossing the Ts and dotting the Is as players try to extract a little extra territory at the edges of their opponent’s holdings or provoke their opponent into occupying space they could otherwise score for

Like the description of chess’s arc, this is an extremely broad sketch, and sometimes games are resigned before they are played out, but it nails down the basic ideas.

One final example for now that will be familiar to most eurogamers. Puerto Rico and its offspring (like the card games San Juan and Race for the Galaxy; Saint Petersburg is pretty close too) follow a typical arc, especially when played by experienced players:

  • Income – players begin the game trying to develop a reliable income of money or cards (which are used as currency in the card games)
  • Engine Building – once income is reliable, money is used to build a means of steadily producing victory points.
  • Conversion – once engines are built, they are used in the final stages of the game to accumulate victory points as quickly as possible before the game end conditions are met.

If you’ve played one of these games, you may realize that the phases that I describe are artificial, since there are rarely clear boundaries. One phase of play tends to move into the next over the course of two or three rounds or turns, and these transitional times mix the characteristics of their neighbouring phases. Different players can go through these transitions at different times, too, especially in longer games. I’m using them as a convenient abstraction that helps to illustrate arcs. They can also be useful when playing some games, though, as a guide to what kinds of problems you should be solving during different times in a game.

Not all games have an arc and many others have very little, and those games tend to produce a feeling of lather–rinse–repeat when you play them. This may not be a problem if the game is short enough or if other aspects of the game’s design are engaging enough, but how much you can tolerate is largely a matter of taste. A lot of gamers love Clans’ quick pace and unconventional tactical problems, for instance, but I (and many others) find it wears out its welcome about halfway through because there is little evolution in the nature of the game’s decisions.

Mind you, games that appear to be lather–rinse–repeat may have a surprisingly strong arc. When you go through most of the deck during a game of High Society, there is an opening phase where players are trying to pick up an item or two on the cheap, then, once positions have been established, there is more maneuvering for position, and toward the end of the game players are trying to manage their remaining money so they can move up the ladder without ending up the poorest (and therefore ineligible to win). It’s not the most sweeping arc in boardgaming, but it suffices to keep play fresh for 15 minutes.

Generating Arc

The best gaming experiences usually come when a game’s narrative is organically produced by the game’s processes. That is, the designer has not included any rules that are specifically intended to shift the landscape on the players, changing the nature of their decisions artificially. Puerto Rico starts players off with very little income, but offers expensive buildings that can act as force multipliers and inexpensive buildings which can improve a player’s income (including the production buildings). This makes it productive to focus on income early because that income can be invested in buildings that provide a return on investment that easily compensates for the time spent developing income. You can probably overcome an investment strategy with a little luck, the bonus ducats you get from role selection and a heap of corn plantations, but it’s tricky, to say the least.

Likewise, Chess’s arc arises from a couple of organic factors. The more obvious one is that pieces are removed from play, never to be replaced, producing an inexorable march towards the meticulous endgame I mentioned (except for the rare games where queens stay on the board throughout most of the game). The early part of the game’s narrative is generated by the initial position, though. The setup position puts the players’ pieces in about the least active configuration possible, so you have to mobilize your forces before you can conduct effective operations. Mobilizing your forces requires (aside from the knights) moving your pawns out of the way which, more or less coincidentally, also threatens to restrict your opponents’ pieces in a way she wants to avoid.

Ensuring that your game design naturally produces a strong arc can be a tricky business, though. In fact, plenty of commercial games – even long or “epic” ones – have little to no arc because it is such a hard thing to create organically. It’s easy to understand why designers take the easy way out and hardboil game-changing events into the rules. This feels hamfisted, though, unless it has a strong thematic basis.

Amun-Re’s Flood, at the midpoint of the game, can seem weird the first couple times you play, even though it presents some interesting questions in the second half of the game. It takes away all of the players’ on-board holdings, and it wipes out almost every farmer in the Nile Valley. The pyramids that were built in the first half remain, however, and that has a big effect on how the auction plays out in the second half of the game. These changes in the strategic landscape help refresh the game, but it also affects how you play the first half, since you must make your plans knowing that your pyramids will be put up for sale in the next half.

Pandemic’s Infection rate is just as artificial, but it feels less obtrusive. It happens several times during each game, so it does not come across as tacked on or forced. Each instance of it has a small impact on the feel of the game, too, so it gathers steam with time instead of causing one massive earthquake. It is one of a handful of mechanics that contribute to Pandemic’s arc – curing (and eradicating) diseases, the depletion of the cube supply and the outbreak meter also contribute – so you don’t devote as much attention to the Infection rate as you do to Amun-Re’s flood, either.

Another example of a mechanically forced game arc – this one more annoying because it is unnecessary – is the availability of buildings in Caylus. Players can only build wooden buildings at the start of the game, but, once a particular wooden building is constructed, stone buildings become an option and special buildings are available once the right stone building is bought. This brings stronger buildings into play in waves, which limits the options new players need to consider, and creates an arc for Caylus by providing more, and more powerful, options as the game progresses. The same effect could be produced less obtrusively by keeping key building materials out of the game until certain buildings come into play, instead, though. In fact, Caylus does exactly this with gold. Functionally, this would pretty much work the same way the actual design does, but the lack of direct connection between the trigger and the new wave of buildings is removed, making it less obtrusive. You get hit by a rubber mallet instead of a metal one.

Carefully structured building costs are an even better solution. They remove the obvious mechanical hand that says “hey! there are new things to consider now!” completely and it eliminates the waves of new buildings in favour of gradually adding new options as players acquire more resources, letting the game’s arc progress more organically. You can see this structure in action in Puerto Rico, although, lamentably, a couple buildings’ prices are wonky. Early in the game, players only need to concern themselves with the first column and a half of buildings, but as their income increases their options expand. Strategy-defining buildings like the Harbor and the Factory become available, and the 10-cost capstone buildings become a viable option somewhere between half and two-thirds of the way through the game, becoming the main target when building during the last few rounds. Puerto Rico also demonstrates the interesting decisions that can emerge from offering players a choice between taking a few small buildings early or saving up for a larger, more powerful building before the other players can snap it up.

There are practically an infinite number of elements of gameplay can produce an arc, and most games that have a strong arc combine more than one of these elements. I doubt anyone could produce a complete list of the elements that feed into a game’s arc, but some common examples include:

  • Expanding play options; often new play options are stronger than older options (e.g. Caylus, Agricola, To Court the King,)
  • Existing play options become more powerful (through synergies or timing effects) (e.g. Puerto Rico family, Goa, the opening in Chess, Magic: the Gathering)
  • Diminishing play options (e.g. Chess, Go, some wargames)
  • Increased purchasing power (e.g. Magic: the Gathering, Puerto Rico,
  • Increasing payoffs (e.g. Princes of Florence, Modern Art, Tower of Babel)
  • Increasingly difficult challenges (e.g. most co-operative games, Kingsburg)
  • Bombs – events which suddenly and radically change the state of the game (e.g. Amun-Re’s Flood, Tigris & Euphrates’ external conflicts, various stock market crash mechanics)

Bits and Bobs

There is a common meme floating around, probably perpetuated by writers of various sorts, that human beings are storytelling creatures by nature, and out brains are designed to see narratives everywhere, even manufacturing them where they don’t belong. I don’t know of any research to back up this theory, but it is fairly well established that people are designed to spot patterns, to the point of perceiving them even when reality deviates a bit from what the pattern would dictate. The aesthetic pleasure of narrative arcs in board games probably plays into both of these tendencies if they exist.

In spite of the name, not all arcs are smooth. Some games build their arc through a progression in a particular element – the number of an influential token that is in play, for example, or the players’ ability to accumulate a certain resource – and then deliberately produces a sudden, sharp reversal. It may be an interesting side effect of a mechanic primarily intended to serve another purpose, it may be a deliberate effort by the designer to add texture to an otherwise limited arc, or it may be a means of presenting the players with new variations on previously experienced problems. The (in)famous Flood in Amun-Re may be the most extreme example. This prevents the two halves of the game from mirroring each other too closely, since it shakes up the relative values of the provinces, as does the pressure players feel to fulfill the bonus cards they’ve acquired. Removing the pieces that form a line of four in a row in GIPF – especially as the culmination of a string of high stakes tactical plays – is a less dramatic example of this kind of mechanic.

Every so often, you will read a comment by Chris Farrell or Brian Bankler (or me…) that says a game is longer than it wants to be. Typically, what this means is that the game does not have a strong enough arc to maintain interest throughout the game. Generally, the longer a game is the more important it is for it to have an arc and the more complex that arc needs to be to sustain interest. There’s no magic formula that describes how often the game should change its feel, since other factors – tactical intricacy, for instance – can help offset an weak arc. Ideally, longer games have a richer, more diverse middlegame than shorter games do, though. It’s no coincidence that Fantasy Flight games are famous for their length and their strong narratives. Of course, they are also famous for rules complexity (by modern standards) and heavy randomness, and this is a common price to pay for a rich narrative arc.


Narrative arc is not the be-all and end-all of boardgaming, but it is little understood and rarely discussed with much sophistication. Often a poor arc gets fobbed off as the game taking too long to play, or as a game that is just a Frankenstein’s monster of old mechanics that have been thrown together with nothing new to offer. Indeed, this seems to be the big difference between the great eurogames of the late 90s and early 00s versus the rather flat euros that come out now. Tigris & Euphrates, Princes of Florence, Puerto Rico, Settlers of Catan and Modern Art all have very strong arcs that sustain interest throughout the game without forcing a player to master 20 pages of rules. Generic modern euro X probably has a flat arc that makes the same sort of short rules feel like an exercise in repetitive stress disorders. Indeed, all of the most successful new euros – Agricola and Le Havre, Race for the Galaxy, Dungeon Lords and even Dominion – features strong arcs, too. Somewhere along the way developers – even the greats like Brunnhoffer and Brück – lost track of how important narrative arc is to board games. The result has been a spate of “soulless” games that reinforce the worst stereotypes of the genre, while publishers like Fantasy Flight serve it up in heaping helpings, but with a side order of complexity and randomness.

Ah, the good old days…

  1. I just played Keythedral last night for the first time in well over a year. After setup, the mostly-fixed board seems quite static. Yet the changing–and increasing–resource requirements of the building tiles demands that players change their collection strategies over time (eg getting 2 wood may be inferior to getting 1 wine). In this case, the mechanics of competition for resources doesn’t change. It’s the building requirements which create the arc.

  2. Nice example. I haven’t played Keythedral yet, but I’m hoping to play it soon. I’ll keep an eye out for this.

  3. I’m not necessarily advocating the game. I rate it in the 6/7 range. In its case, the arc is more of a quantitative, mechanical one as opposed to your qualitative examples. I think changes in game quality over time produce a more appealing arc.

  4. Hey, I like Clans….

    And Caylus….

    Generally, though, what you’re saying makes sense, and it helps make sense of the reason that nearly all of my recent game purchases (“recent”=”2009 and forward”) have been from Fantasy Flight.

    Also, it’s a crime that the Games Journal had to cease publication. A crime, I say!

  5. Nothing wrong with liking Clans.

    Caylus on the other hand… 😉

    And, yeah, I really miss the Journal. You don’t see the kinds of pieces they published anywhere now with any consistency. The Geek’s game design forum, maybe, but there’s a lot to be said for editorial oversight.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: