Our third article in the Approachability series focuses on the learning curve. Once you’ve gotten your players in the door and interested, how do you set them loose on the game? Can someone learn the rules and strategy over the course of the game, or does it take three games, or ten? And what’s the best design strategy to make the replay as approachable as the initial play?
To answer those questions, we’ll revisit the clarity and navigation axioms of approachability. Clarity, as we established at the beginning of our approachability series has two forms. The first type, as we covered in an earlier article, dealt with turn-scale clarity and how players were able to understand how their actions lead to results. The second type which we will cover in this article is game-scale clarity which is how a game can ensure a player comes away from a game with an understanding of their performance and how to improve in future sessions.
Clarity, Transparency, and Opacity
One of our favorite bloggers Gil Hova recently wrote an article that ties into our clarity axiom as it relates to the learning curve. The article is about information availability and the ideas of to transparency and opacity. Transparency, he writes, is a property of a game that describes the players’ ability to figure out what will happen next. Chess is an example of a perfectly transparent game because there is no hidden information, while poker is characterized by high opacity because it’s difficult to guess what cards other players are holding.
Neither transparency nor opacity is inherently good for approachability. If a game is too transparent, it can appear as an onslaught of choices with no indication of which one might be the best one to pursue. Alex talked about how the uncertainty gap can reduce approachability in several recent articles. On the other hand, too much opacity leads to a poor understanding of the game’s mechanics. This week’s article will focus on using design to create a relationship between a game’s mechanics and its outcomes.
For certain games, low transparency is actually a positive attribute. Most of the fun in Fluxx comes from not knowing how the rules are going to transform from one turn to the next, and the intrigue in Apples to Apples arises from the various (and hidden) scoring criteria that each player uses. Scenario-based games including Betrayal at House on the Hill and Tales of Andor use a lack of transparency as a source of suspense, as the players are uncertain of how the game will progress or even what the victory conditions are.
Even though all of those games are deliberately low-transparency games, none of them suffer from a lack of clarity. In Fluxx, it’s apparent early on that the way to win is to get Keepers in play–even though it might not be decided until the final play of the game which Keepers you need. So Fluxx has a gentle learning curve–it has enough clarity to be approachable–despite its lack of transparency. Of course, Fluxx is famous for its chaotic mechanics and light strategic weight; in general, games with more strategic depth and complex mechanics need more transparency to retain their clarity.
Iconography is an interesting case of clarity in that it represents a significant time investment before a game is able to be played. Whether to include icons or other symbols on game components or player aids, and how extensive to make them, is a question of opportunity cost. The thought is that, by making the game setup and introduction a bit more complicated, the actual game play will go more smoothly.
Race for the Galaxy is the undisputed poster child for complicated iconography, and an entire page of the player aid is devoted to explaining how it all works. Among new Race players, there is often a sense that the iconography reduces clarity and therefore damages approachability. We’ve defined clarity as the property that addresses if players can understand what options are available to take on a turn. If a player can’t even read the symbols on a card, then it’s going to be nearly impossible to understand how to play that card.
Then, why use iconography at all? The answer relates to another axiom of approachability, parsimony. If players pay the opportunity cost required to learn what all the symbols mean, then the game actually becomes easier to play because each component looks and acts consistently with each other. (Iconography can be a uniting element of aesthetic design as well.) In Race, a card specifying “draw one card when you develop another card, but that bonus does not apply to this card” might initially have more clarity than one with the symbols “II, diamond, +1,” but the second card is a parsimonious and graphically pleasing way to present the same information.
Icons are not always a good design choice. If the goal is to make a game that a new player can pick up and play with a minimum of invested time, then complicated symbols may not be the best way to achieve that goal. The iconography question raises an important point about approachability: decreasing approachability, at least in the short term, may well serve a greater design purpose. But designers should be aware of that trade-off and include such complexity deliberately.
Clarity and the Learning Curve
One method to increase the likelihood of transitioning new players into returning players is to improve the opportunity of a person to grasp an understanding of the game. Without this comprehension a portion of the potential audience for your game will be lost simply because they lacked enough information to make a knowledgeable decision.
Games like 7 Wonders and Race for the Galaxy contain enough complexity and display distinct iconography that they almost require a “first attempt at flight” for new players or (to use an alternative avian adage) have an experienced player take them under their wing and literally walk them through each decision. In either case first time players will fall crashing to the ground which can make for an unpleasant experience.
So how can this be mitigated to provide a better experience to beginners? One method is one we’ve touched on indirectly several times in our approachability series:
Encouraging optimization of scoring:
In our earlier coverage of clarity we looked at the idea of examining the distance to the objective in games. To help reduce the likelihood of the dreaded “first game disaster” we can focus players on a few key items to get them started. This idea is a blend of clarity and navigation suited for moving players from stage zero to stage one on the learning curve.
In Castles of Burgundy players are encouraged (via the scoring structure) to do three things:
- Complete large areas.
- Complete areas quickly.
- Complete areas before other players.
There are several other ways to score points in the game but the central mechanics drive these three motives. Players learn to juggle turn order while selecting from a common pool of tiles. As players progress in experience they may begin to string a group of buildings together for efficiency or adapt to target the tiles needed by opponents. Eventually players begin to develop their planning and see what tiles they can leave on the board for later and what tiles they should prioritize ahead of opponents. Even after the first game of Castles of Burgundy, a player can have a rough idea why they performed the way they did relative to their opponents.
The objective of Ingenious to score points in a variety of colors, and a player’s lowest-scoring color is the player’s overall score. A player may need some time to adapt to this scoring principle but can observe during and after the game why their lowest color held their score back.
Players may begin by targeting the colors most available or they may target their weakest color gradually. As players become more fluent in the gameplay they will seek to boost their score in available colors quickly and spread their focus to many areas. Even more experience will reveal they really only need to create a scarcity in an opponent’s weakest area by sealing off a color and blocking it at every opportunity.
In summary, orienting the focus of gameplay on straightforward scoring opportunities will allow players to develop the learning curve of a game. Optimization is a frequent area of enjoyment for gamers and allowing them to do so will quickly translate into a game that is accepting of new players.
Navigation and the Learning Curve
We’ve defined navigation as a tenet of approachability that guides player actions toward some ultimate goal or objective. Clear navigation is essential to making a game learnable because it gives an idea of what actions should be taken and why. For games that reward creating a strong engine, giving players some idea of how and why to build the engine is helpful to smoothing the learning curve.
- In Puerto Rico, experienced players know that the 10-cost buildings form cornerstones to the game’s most basic strategies. But new players wouldn’t necessarily realize that, much less have any idea how to get there.
- How is a beginner supposed to understand what can work together to create a well-designed deck in Dominion? It is usually a bad idea to take low value VP cards immediately, but at some point, it’s important stop adding money/action cards and start taking VP cards. Without having played before, though, it’s difficult to realize where that threshold is.
- 7 Wonders suffers a similar problem: the optimal strategy is usually to acquire resource and commerce cards early and build toward wonders, monuments, and guild cards, but it’s difficult to know at first where that turning point is, and what resources might be important later on.
- Both Dominion and Agricola use different subsets of cards every game. Depending on what cards are selected, the game can play very differently: Agricola with an abundance of plow minor improvements rewards farming far more than one with no plows at all. The inclusion of different cards every game creates vast replay value but makes navigation more difficult; it can be tough to recognize the best way to use the fishing pole if the fishing pole only shows up in one of every three or four games.
Again, none of the above games suffer from clarity issues. It’s obvious in Dominion that you score points by having Victory cards in your deck, and you win the game by having points. And the mechanics of how to acquire Victory cards are simple to understand. Where Dominion’s approachability is damaged is in navigation: there is a definite learning curve to knowing the right moment in the game to shift from Treasure and Actions to Victory.
A strong navigation component isn’t the same thing as leading players by the hand. Good navigation can be as simple as having consistent rules and mechanics throughout a game, or it can involve active techniques to nudge players toward optimal strategies.
- In Ticket to Ride, incomplete routes at the end of the game complete with a score penalty. The penalty disincentivizes players from continually wasting turns on drawing route cards, a great example of loss aversion. Absent an incomplete route penalty, an inexperienced player may (unwittingly) spend too many turns drawing route cards over the course of the game and would likely finish the game with an uncompetitive score. Such a player could potentially routinely finish 10-20 points behind in the game and always feel dissatisfied with the game when they have actually eliminated themselves from the game long before scoring.
In summary, games with good navigation ease the learning curve because they allow players to develop an intuition about how to proceed in any given situation.
Skill Progression and the Learning Curve
The last area we’ll look at for this article is designing with consideration for the learning curve. There are numerous ways to raise the elevation of the learning curve in your game. The orientation of the learning curve is most heavily influenced by the amount of experience necessary to climb it. Generally a heavier game would have a far more gradual curve as seen in red below as more time is necessary to become fluent with game strategies and advantageous positions. A game with the steep blue curve below would have a large upfront investment but provide little marginal benefit as experience is gained during later games.
The maximum height illustrated by each curve would be tied closely to its overall weight and complexity as well as the intricacies of its advanced levels of play. Games like Chess or Go would have very high maximum elevations with a change in slope similar to the green curve on the graph. In these games players can learn a great deal from early exposure to basic strategies and playing styles. At some point there is a shift in which the slope flattens out as players will require far more experience in order to become knowledgeable about more complex strategies and advanced levels of competitive play. Let’s take at some more applications of these ideas:
Complexity and Depth
As we’ve discussed in an article earlier this year on gameplay complexity, one method is to increase the number of social elements in the game by including social deduction, negotiation or bluffing mechanics. The unpredictability of human opponents will require a significant time investment in order to gain adequate experience. This is evident in many popular cards games driven by these mechanics such as Poker and Gin.
Randomized and Adaptation
A randomized set-up is a common method to increase variety and replay value. These also reward players who have attained the experience and ability to adapt to the changing conditions in each game. Settlers of Catan will reward players with the experience to identify optimal starting positions for cities during the first phase of the game. This experience is probably gained in just a few games, with little marginal gains in experience after this. This would be a steep learning curve as most of the proficiency is gained from a relatively short amount of experience.
Compare this learning curve to a game like Saint Petersburg in which players not only adapt to the cards that come into play but also the pacing of the game players must adapt to in order to optimize their score. A game like Saint Petersburg probably requires far more games than Settlers of Catan in order to become proficient in the variety of card combinations and pacing. When graphed Saint Petersburg should have a more gradual learning curve compared to Settlers.
A game like Le Havre has an even more gradual learning curve. Players randomize the building cards prior to the game meaning the availability of key buildings may be significant in determining the viability of specific strategies. Le Havre randomizes the resource tiles which determines the schedule of added resources each round. A large number of special buildings are used during the game but most of the cards are never seen during the same game.
The turn structure of Le Havre provides players with equal turns but distributed unequally between rounds. Each turn during the game is an interesting puzzle with no substitution for experience as there is a large degree of “one step back to make two steps forward” with finding some of the best scoring opportunities. Le Havre would display a very gradual slope to its learning curve as players must gain far more experience in order to be fluent in its optimal gameplay.
One of the best illustrations Le Havre can provide is that players are aware there are large changes between games and they may not encounter every building for quite some time. Including plenty of areas to explore is one of the best ways to encourage players to continue ascending the learning curve of your game. This design implementation often conflicts with our axiom of parsimony so it can be a challenge for game designers to identify the proper balance of ideas.
Both clarity and navigation are essential elements in creating a reasonable learning curve for a game. The easier it is to figure out how to play a game, whether in the mechanics of how a card is played or what all the goofy little symbols on it mean, the more a player will be enticed to play it. And games that shine some light at the end of the strategic tunnel incentivize players to travel down a defined path rather than fumbling around in the dark.
A successful game leaves players thinking “I didn’t quite get there that first game, but I can see how my decisions can have benefits and repercussions for next time”. On the other hand, a poor game leaves players failing to see the forest for the trees and feeling that they just didn’t know where they went wrong. Games with well-designed learning curves improve approachability by allowing players to think of strategies they want to pursue later in the game, or the next time they play, and they empower players to play in a manner of their choosing.
As a final thought, playtest your game with other game designers and spend a moment to collectively graph the learning curve of your game. Consider what that learning curve means for your audience. How you can get players to commit to climbing the heights of your game design? Are there interesting strategies or ideas to discover even after playing ten or fifteen games? Ideally if your game has advanced levels of strategy that will oblivious to new players you should invest time in including more fun and engaging elements of assurance that will encourage the player to come back and observe something new. If your game requires players to absorb a great deal of information for their first game, consider including further clarity and navigation to assist them during the most challenging periods of training.
How should game designers think about the learning curve of their games? Share your thoughts with us in the comments below.
- Blog – Fail Better – Vision Quest: Transparency vs. Opacity in a Board Game
- Blog – Ludi Berkeley – Approachability and Approachability Redux