7   Putting It All Together

In the introduction, I posed this question:

How does the designer create the desired player experience? How can the emotions of the players be steered onto the desired path?

Loss aversion is a surprisingly versatile tool to achieve a wide variety of experiences. The core of loss aversion is simple: losing feels worse than gaining feels good. But designers can spin this simple core into a vast array of player emotions by working with or against people’s natural inclinations:

There’s an important caveat for everything discussed in this book: these studies describe tendencies, not absolute truth for all people. Think back to the first formal study question posed:

Which would you rather have?

(A) An 80 percent chance of winning $4,000.

(B) A 100 percent chance of winning $3,000.

If 80 percent of people answered B, that means that 20 percent of people answered A—or one in five. So there is an excellent chance that in almost any multiplayer game, there will be people who don’t respond the way the designer expects them to—or don’t get the intended emotional impact.

And most of the studies had results in the 65 percent range, meaning that one out of three people responded the other way. The car wash study resulted in 34 percent of people redeeming one coupon, compared to 17 percent who redeemed the other. But 66 percent of people still never bothered to get the free wash. My twelve-year-old niece recently played Portal and tossed the Weighted Companion Cube into the fire without a second thought.

Everything that’s discussed in this book is a tendency and doesn’t affect everyone the same way. The reaction depends on the environment, the type of day the player had, and so many other factors. Incorporating the emotional triggers related to loss aversion may nudge players in certain directions, but, as they say, your results may vary.

The ultimate goal of understanding loss aversion and its associated techniques is not to ensure only positive emotions or only negative ones. No emotion is always right or wrong to incorporate into a design. And though properly tuning the emotional palette of the game will not guarantee that it succeeds or is popular, having an emotional mismatch will certainly result in the opposite.

There are many elements that go into game design, and all of them are critical to the user experience. User interface, story, artwork, sound, game mechanics, difficulty curve, feedback loops—each is incredibly important in its own way. Loss aversion is a piece of this puzzle—and often a hidden piece. Testers may be able to point to why controls were difficult to use or why a particular puzzle was too obscure, but they may not realize why a certain decision rubbed them the wrong way or made them anxious. Designers should be familiar with the reasons behind these amorphous concerns and, even better, to design things in the first place that utilize these techniques effectively from the start.

The Agricola Series

The game Agricola,1 designed by Uwe Rosenberg, has been a top-rated game since its release. Themed around medieval subsistence farming, it challenges the players to use their limited workers (initially two, more if you have children) to expand the family farm and avoid starvation. Rosenberg has released two games building on the theme and mechanics of Agricola: Caverna2 in 2013 and A Feast for Odin3 in 2016. Looking at what was changed in these subsequent games—and the impact these changes have on the player experience—illustrates many of the themes of this book.4

At the end of Agricola, you score points from various aspects of your farm. You get points for having animals, pastures, vegetables, and more. The score summary from the game outlines this, as shown in figure 7.1. Notice, however, that you’re penalized if you’re missing any of the elements. No sheep means you lose a point, and you also lose points by having undeveloped areas in your farmyard.

Figure 7.1

Agricola score card.

The design goal of this somewhat convoluted system is to push players to diversify their farms, which presents an interesting challenge because specialization makes players the most efficient. So there’s a balance there. But as you can expect, these penalties create a negative emotion in players: first-timers often forget about them, and experienced players worry and work hard to avoid them whenever possible. These are clearly framed as losses.

In Caverna, this dynamic is softened a bit. Penalties are only given for missing farm animals and for empty spaces, not for the other elements of the farm. Rosenberg removed over half of the loss aversion elements in scoring.

For Odin, Rosenberg took a different tack. Rather than having players lose points for not being diverse, he represents penalties geometrically on a player board. Instead of growing vegetables and getting farm animals, players have to acquire goods, represented by different colored polyomino pieces of varying sizes. These are then placed on the player board, starting from one corner and building outward (see figure 7.2). Spread across the board are spaces with a −1 symbol. At the end of the game, the player loses points for the −1 spaces that are not covered by goods tiles. In addition, goods tiles of certain colors may not be placed next to one another.

Figure 7.2

A Feast for Odin player board. The −1 spaces can be seen in the grid on the left.

This accomplishes two things. First, by requiring that different color tiles abut, it enforces diversity in what goods players acquire. Next, by putting the negative spaces in front of players right from the start and having those spaces be covered as the game goes on, an end game negative is turned into an in-game positive.

This is a terrific example of framing. By putting the negatives right out in the open, players can only gain points as the game goes on. They are at their worst right at the start, and then they build up from there. It also gives great positive feedback as the game progresses because players are able to purchase bigger tiles and cover a larger number of negative spaces at the same time. This process gives an emotional lift to the players, who can shed losses more easily as the game progresses.

Rosenberg could have taken this one step further and switched the negative spaces to positive spaces, so the players would gain bonus points as the spaces became covered instead of losing negative points. Mathematically it would be the same: players who did not hit a bonus space would still be a point behind their opponents who did. It is possible that Rosenberg kept negative points in Odin to preserve the lineage of negative points from Agricola and Caverna. Perhaps it would have been simply a step too far in the evolution. However, it is also possible that he believed the negative points created a stronger drive to cover them up. Loss aversion tells us that eliminating negative points is more motivating than gaining bonus points and that players will work harder to achieve that goal. So maintaining this feature drives player behavior to remain focused on that aspect of the game, which the designer may have decided was part of his vision for the player experience.

A second aspect that evolved was feeding your people. In Agricola, you must have two food units for each of your workers at increasingly frequent intervals. For each food you are short, you need to take a Begging card, which causes you to lose three points at the end of the game. Your workers don’t actually starve and get eliminated from the game; it’s simply a point penalty assessed at the end.

But the feeding phase does trigger emotion in the way it’s framed. Your workers are family (they are called family members). The action that creates a new worker is called the family action, and your new worker can’t work at first but only needs half the food, an obvious analogy to having a child. Despite the workers being flat wooden discs, the game takes pains to personify them as a family on a family farm.

Not having enough food to feed them at the end of a season feels very personal. And sometimes you will need to sacrifice things to avoid starvation, which may still generate feelings of loss. For example, you may have to eat the vegetables you were hoping to plant next season for more crops—or slaughter an animal for the meat, which may prevent your animals from breeding. These are very visceral choices that definitely trigger loss aversion in the players. On the positive side, the need to feed your family does give new players a clear, early goal in a dizzying array of options. You need to have a certain amount of food in a certain number of turns. You have focus immediately, which helps get players acclimated to the game.

In Caverna, the system is similar to Agricola. However, there are more ways to generate food, so it is a bit simpler to get everyone fed—allowing players to focus on the many additional activities added by Caverna, such as mining and exploration.

Odin preserves the need to feed your workers but twists it around, again via framing. First, the feeding phase is now called the feast phase. You are not failing to provide food to your family but simply failing to adequately provide for a feast. And you no longer gain Begging cards. They are now called Thing Penalties,5 which is much more neutral. Begging is very much a loaded term with negative connotations.

In addition, the mechanism for feeding is based on placing tiles on a banquet table, similar to the way goods tiles are placed on the main player board. And these also come in various widths, making it more viscerally satisfying (and intuitively obvious) to see how close you are to fulfilling the feast requirement.

Finally, silver coins, which are a wildcard resource, can be used as a way of making it easier to fulfill the feasting requirement in Odin.

These changes, taken together, make it much less likely that players will fail at feeding their workers. The rules specifically say, “This should not happen too often, because the feast happens after you receive income. Income should always provide enough silver to cover the spaces.” There is no similar assurance in either Agricola or Caverna.

The final element illustrating Rosenberg’s progression are the workers themselves. Called family members in Agricola, dwarves in Caverna, and Vikings in A Feast for Odin, using these is the core way that players do things in the game. For example, placing a worker on an action space allows a player to immediately take that action. In Agricola, placing a worker on a Gain 3 Stone space allows you to take three stone tokens.

At the start of Agricola, you have two workers. This creates a situation that is primed for regret. The main drivers for regret are decisions that are irrevocable, limited, important, and clear. Here we have all those ingredients. You only have two opportunities to pick something from a limited menu. The action is done immediately and cannot be undone. There are some spaces that give a choice of actions, but the player must immediately decide what to do. And with the threat of starvation, the importance of the decisions—and how they impact your ability to get enough food—is clear. This tendency for regret is strong and does not lessen as the game progresses—despite players building up engines to create food, like flocks of animals, and adding more family members—because the number of turns between harvests (when players need to eat) decreases throughout the game.

The system in Caverna is similar, although because food is less of an issue, the individual decisions feel less consequential. But you still start with only two workers.

In Odin, you start with five workers and automatically gain one each turn to a maximum of twelve. Counteracting this a bit, actions cost between one and four workers to perform. In Agricola and Caverna, they always cost one. This dramatically reduces the potential for regret. Having the ability to place five separate workers means that each individual decision is less important on its own. It gives the opportunity to recover from a mistake.

Having more workers also enhances the feeling of endowed progress. Agricola and Caverna, with only two workers to use each turn, make it more difficult to chain things together from micro- into macroplans. With five workers, Odin has the luxury of smaller actions that give the players more of a sense of forward momentum: they have a heightened feeling of having started on a plan that needs to be completed.

Including spaces that require varying numbers of workers also serves to signal to beginning players the relative strengths of certain actions, as well as reduce the rush for the best spaces that is seen in Agricola. These improvements are not directly related to loss aversion but are worth noting.

Taken altogether—creating negative spaces on the player board instead of negative end game scores, replacing begging with feasting, and increasing the number of workers—Uwe Rosenberg has clearly moved in the direction of reducing loss aversion and regret and enhancing endowed progress by reframing or directly modifying underlying mechanics. Although personal preferences will vary, he has decided that softening these edges creates a better experience for the players.

Conclusion

More and more, game designers aspire to sophisticated experiences and advanced narrative techniques that will take the players on a journey. My hope is that by understanding loss aversion, which is so fundamental to the human psyche, designers will be able to more directly tune the game to achieve their goals and create ever more memorable and dynamic experiences.

Notes

  1. 1. Uwe Rosenberg, designer, Agricola (Lookout Games, 2007).

  2. 2. Uwe Rosenberg, designer, Caverna (Lookout Games, 2013).

  3. 3. Uwe Rosenberg, designer, A Feast for Odin (Lookout Games, 2016).

  4. 4. I am indebted to Grant Rodiek for much of this analysis. See http://hyperbolegames.com/blog/farm-odins-table.

  5. 5. A Thing was a Viking assembly.