Part 3 - Case Study

In part 1, I described the method, in part 2 I went in to more detail, and in this last part of the series, we look at how to apply the method to an actual game. We look at how we can use this method both to see how a feature affects a game, and how to use the same analysis to identify opportunities for potential improvement. We will use the Happiness feature in Civilization V as our example. As we focus on Happiness, we will construct a graph for what it takes to master Happiness, and then see how that fits in with the greater analysis of Civilization V as a whole.

First, let’s look at Civilization V’s total. Civ V is a very complex and deep game, and the graph shows it. Fully analyzing Civ V is well beyond the scope of this article, but focusing on one feature alone will illustrate how we can go about analyzing each part of a game, and how to build a full analysis from that.

The steps necessary for analyzing a feature is first to get a thorough understanding of the feature and how it fits in the game, then build a time line for when and how the player learns to understand that feature. With that information gathered, we can construct the graph for the individual feature. The final step is to collect all the features together and build the final graph for the game as a whole.

Civ V Happiness

A fundamental element of all Civilization games is Happiness, and in Civilization V[1], Happiness has far-reaching effects on the player’s success. In order to analyze Happiness’s impact on the game, we have to first understand how it works, but also try to estimate at what point a player is likely to learn any specific aspect of how it works.

Even before the player starts the game for the first time, she will have an innate understanding of Happiness, just due to the name, in the sense that Happiness is a good thing, and the more of it the better, and conversely that Unhappiness is a bad thing. If the game was designed by someone from a different cultural background, they may not have called this concept Happiness, but even so, they are likely to understand the most fundamental aspect of Happiness in Civilization V: Happiness is good, Unhappiness is bad. Unfortunately, the player is unlikely to have little other relevant knowledge of how Happiness affects the game. Even players that have played previous Civilization games don’t know that much more, they know that Happiness is an important concept, but they don’t know how it will work in this version. One advantage veteran players do have over new ones is that they may feel that they need to learn about Happiness, and hence are more likely to be on the lookout for anything that will help them do that. Though these players may not be able to understand how Happiness works from a mechanical point of view, there is one more piece of information that they are likely to have before they start playing the game, and that veteran players are almost certain to know, and that is that entertainment-type buildings will improve Happiness. For the analysis, all this information goes into the bucket of innate knowledge.

The developers of Civilization V made the decision not to teach the game through a tutorial, but rather let the player jump straight into the game and teach the player through information in the UI, helpful tips, and trial and error. Since exploration and discovery is central to Civilization, letting the player explore the game mechanics is arguably staying true to the game. Also, due to the scope of the game, teaching all facets of Civilization in a tutorial would make that tutorial prohibitively long, and frequently interrupting the game for a new tutorial as new aspect of the game is encountered would probably be quite bad for pacing and emersion. As we’ll see, this will greatly affect how much is learned from tutorials.

Please see the appendix if you don’t know Civilization V and how Happiness works in it, or if you want to get a feel for what is required of a more detailed analysis of a feature.

The Analysis

Once we understand the role of Happiness in Civilization V, and how the player learns about it, we can analyze its impact on the game’s depth and accessibility.

The first step is to figure out how much understanding each aspect of Happiness makes up the total understanding of Happiness needed to fully master Happiness. This can be a straight forward process for some features. For instance, if your game contains 5 unique weapon types, mastering each means mastering 20% of the all the weapons, assuming that each weapon is equal in complexity and importance. This is unfortunately much more complicated for a feature like Happiness, and lacking any better method, I will be using my subjective judgement. The goal is to create the table below, it shows how much mastery the player is expected to achieve for a specific amount of effort. The method I use for doing this is to first sort each component of the feature into the appropriate bucket, and then decide the impact of each bucket relative to the whole.

The first row, innate knowledge, is based on the general understanding of happiness that it is reasonable to assume that players have before they already started playing the game. In this case, they understand that Happiness is good, and that an Unhappy populace is bad. Their innate knowledge also enables them to more easily learn things such as that entertainment and luxuries improves Happiness as those rules aligns well with a player’s general knowledge of how to make people happier. Tutorial is what a player gains from the tutorials in the game. In the case of Happiness, I don’t think that there is much to learn from the advisors, but they will at least contribute something together with the other hints in the UI. Assuming a relatively good pace, the player should encounter Happiness in the game more directly within the first half hour of play. This is the point that the player will be forced to look into Happiness, learn that there are consequences of being Unhappy, and the basic methods to deal with it. After two hours of playing, the player should be well acquainted with the basic effects of Happiness, understand that with more and larger cities comes more Unhappiness, and understand the most direct ways of increasing Happiness. After about five hours of play, the player will have seen most things in the game that can be used to create more happiness. At the ten-hour mark, the player is likely to have finished her first game of Civ V and already started her second. This means that the player will have seen the end-game, including some of the effects that Ideology has on Happiness, as well as having seen what influence through Tourism can do. But a second game also means that the player will be able to go through a new game with the knowledge gathered during the previous one and notice more details and options. Soon though, the player will need to start experimenting, turn to online resources, or in other ways put in a more concerted effort in order to come up with different strategies for dealing with Happiness, as well as testing the pros and cons of various civilizations. To truly get to the bottom of the deepest ramifications of how Happiness works will require the player to invest significant time and effort. All this is reflected in the table below.

Using the above table, we can make a graph to show how Happiness affects the game in isolation. This is useful as it allows us to study depth and accessibility of this particular feature, and to compare it with what we originally envisioned. Looking at the graph, it is easy to see that Happiness has considerable depth, but at the same time Happiness is accessible enough for a player to be able to effectively play with Happiness after a game or two.

The next step is to figure out how much Happiness itself affects Civilization V as a whole. This can be done on the game in its entirety at once, or for larger games, dividing them in to individual features[2] as we did with happiness, and perform the analysis separately for each, and then merge those results into the final result. The advantage of the second approach is that concentrating on just one feature at a time makes it less likely that you accidently forget about or gloss over any particular detail. However, when analyzing each feature separately you do run the risk of either missing complexity where the features interact, or accidentally counting parts twice. It also does require you to merge them into a whole once you’re done. This isn’t as complicated as it might first seem, but is a straight forward mathematical operation. What you need to do, is to decide what impact mastering each feature has on mastering the game, normalize each feature’s contribution and add it to the whole. The only thing that makes this process a little more complicated is that you have to remember when you analyzing each feature that you do so in the context of the player playing the whole game. You will get a skewed graph if you decided that each feature in isolation can be learned in say 2 hours, and then add them together. It may be true that if what the player had to learn was just one feature, that this could be done in the 2 hours, but you need to take into account that the player has to learn not just one feature, but all the features in the game, and that that may be too tall an order to accomplish in those 2 hours.

Below is a graph[3] showing the Happiness feature in isolation next to the graph for Civilization V as a while. The final graph show’s Happiness’s impact on Civilization’s total[4]. The brighter areas show Happiness’s contribution, while the dulled areas show the complexity contributed by the rest of the game.

Now that we have the results of our analysis, we can start using it. From the graph it’s apparent that Civilization is a game with significant depth that takes a long time to master. Happiness, compared to the game as a whole, has a similar profile, but is somewhat more accessible than at least some of the other components of the game. Having gained a good understanding of how the game’s profile look, we can now start to ask the important questions. If we were making Civilization, is this the kind of profile we want? Does it meet our goals? Does it have the depth we want, and can a new player start enjoying the game without having to spend an unreasonable amount of time learning the game first. Is it accessible enough that a casual player that is not willing to spend the time needed to gain a deep mastery of the game can still enjoy it? Is the Happiness feature of an appropriate complexity, and is its impact on the game as a whole the right one? Civilization is a game with many parts, so it is important that their individual impacts on the game’s total complexity is well balanced. A minor feature that was meant to be just a bit of flavor should not have a major impact on the game’s complexity as a whole, while something central like war, should. Other important questions are, are the important concepts of Happiness being learned at the appropriate time and is the accessibility of Happiness well matched with the game’s other features? Since accessibility is a measure of how easily understood a given feature or game is, it can either be improved by making the feature or game simpler, or by making it easier to understand. This means that adding a good tutorial where there is none may be a good low impact option if accessibility needs to be improved. The reverse, too, is of course also true, but it is rarely desirable to make something harder to understand, so often the only option is to increase complexity if the feature or game is too easy to master.

We would ask other questions if instead we are making a Civilization competitor. If we want to make a competitor in a very similar vein, then there are good reasons[5] to try to make a game that has a very similar graph. If we want to make a more casual game, like The Battle for Polytopia, then we want to concentrate on making the game easier to learn, in particular you need to ensure that mastering a significant part of the game is easy. This doesn’t preclude a significant amount of depth on the tail-end, but this should cover a smaller part of the game. If, on the other hand, you would like to make a deeper game, like the Europa Universalis series of games, you need to ensure that there is a significant depth to master.

To be clear, no specific profile is optimal. It is all about what is suitable for the game. If you want to make a game with more depth than Civilization, then it is important that your game’s profile is supporting that. If you do your analysis, and the result is not what you had planned for, then either you need to make some changes to your game, or how you think of your game. Which solution is best depends on your particular circumstances.

Appendix – Deconstructing Happiness

When the game start, the player will be prompted to found her first city as soon as possible, and when she does, some information will show up in the menu bar which includes her empire’s current Happiness.

The smiling face with the 7 next to it is the current Happiness. At this point, if the player moves the mouse cursor over the smiling face, the game will present her with additional information about the Happiness of her empire.

Though the meaning of all the numbers here may not be something the player understands at this point, it is still a great resource for understanding that there are multiple things that affect Happiness, and the player can be on the lookout for them. Over all, this is an excellent help for the player to learn and understand how Happiness is calculated. Unfortunately, it has a number of weaknesses too. The foremost one is that there is nothing in the game that encourages the player to move the mouse cursor over the little smiley face in the first place, a simple prompt to explore the information in the status bar would have gone a long way. Secondly, there is nothing here that explains why you should care about Happiness, other than that it affects city growth, though that alone should be plenty motivation for a player that understand the importance of city growth in Civilization. Thirdly, there is no second layer to this information. The player can’t dig deeper into the information if for instance she wanted to know more about the 1.8 Unhappiness generated by her number of cities.

The next big event concerning happiness happens after about 30 minutes to 2 hours of playing depending on how fast the player takes her turns. If you are an experienced player, you might object that this will happened well before the 2-hour mark, and possibly even before 30 minutes. This is a good example of the difference between an experience player and a beginner. We need to look at these times for beginners, not for expert players, as we are trying to discern how long time it takes for someone who is new to the game to learn it. An experienced Civ V player knows all the options available to her, is capable to quickly assess the early game state, and will have well established build and move orders for the start of the game, and hence will move through this phase at a very different pace than someone who is new to the game. Regardless of how long it takes the player to get to the next event, when it happens, one of the game’s advisors pops up in a window to offer the player some advice on Happiness. This window contains less information than what you get from hovering your mouse over the happy face in the status bar, but it has links to information from which you can learn more. The problem is that the player at this point has seen many of these pop-ups, and though they may be helpful to a novice player willing to do a lot of reading, they are interrupting the game, and mostly just telling players things that they already know. These advisors are very wordy, (generally considered a problem for keeping players’ attention), and my suspicion is that the vast majority of players have started to ignore the advisors at this point, and very few have taken the time to explore the linked text that they provide after having seen the first few.

Most players playing Civilization will notice that sometime during the early game, somewhere between 30 minutes to an hour after the previous advisor popped up, they run out of Happiness. This is because in the beginning, the player’s ability to produce Happiness is lagging behind their ability to create population and cities (the main sources of Unhappiness), and eventually the player will find that their empire’s Happiness is now in the red; that is, it is Unhappy. At this point, the same advisor will pop up and explain Unhappiness and how to address it. This advisor again, is suffering from the same issues as the previous advisor, only that by now the player has seen even more of them and is even less likely to read what they have to say (it should be pointed out that there is no way to tell how important any particular advice is, they look the same from the mundane and to the critically important). Also, worth noting is that though developing Happiness resources, as the advisor suggests, is a good suggestion to gain more Happiness, it is far from the only option, and given the player’s situation may not even be a valid one.

Once the player’s empire is Unhappy, the player will start to notice some effects. Just how obvious these effects are to the player depends on how much effort the player puts into the game, and how much the player pays attention. There are numerous pop-ups, mouse overs and other ways to gain information that will give the player a hint of how things are affected by Unhappiness. The player is, however, never encouraged to seek this information out, or even told that it exists, and there is also no place that will give the player a unified view of the full impact Unhappiness will have on their game.

Unhappiness affects an empire in a number of seemingly minor ways: It reduces the empire’s production, taxes, population growth and the fighting efficiency of its armies with small amounts that grows linearly with increased Unhappiness. In addition to this, there is the concept of a Golden Age, which greatly improves an empire’s production and taxes for a set duration. Golden Ages are triggered by accumulating a certain amount of excess Happiness over time (Unhappiness will only prevent Golden Age points from being accumulated, not subtract from what has already been accumulated). There are also additional effects when the empire’s Happiness drops below -10 and -20 respectively, but it is very unlikely that that would happen unless the player engages in a war with an opponent that targets the player’s Happiness (which can be a very effective tactic).

The problem with the effects Unhappiness has on a game are that they are seemingly small, short-term effects, but they actually have major long-term ramifications that will be hard for a player to fully grasp unless they spend a lot of time in the game and study the rules and their impact closely. If a player allows her empire to become Unhappy, the effects of that Unhappiness makes it harder to improve the situation. Often, what is required is for the player to build or buy something that improves Happiness, and the player’s ability to do so is diminished by Unhappiness. Civilization V as a whole can be viewed as the player accumulating capacity to win the game over time, while the opponents do the same. The capacity growth follows a particular trajectory, and any player that gets stuck too long dealing with one crisis or another will be flatlining their trajectory, effectively being left behind. In order to later catch up, that player needs to increase the pace of her empire’s development to outpace that of their opponents’. This is obviously a tall order, particularly if we assume roughly equally competent players, hence truly understanding, planning for, and managing Unhappiness will have a major impact on a player’s ability to win the game. Due to Unhappiness’s often subtle, but long-term effects, fully understanding Happiness in the context of the game is both difficult and requires a good understanding of the game’s other systems.

In addition to understanding how Happiness affects the player’s performance in the game, as well as how it interacts with other systems, the player also needs to understand the tradeoffs that the she has to make in order to gain and maintain happiness. There are several resources that can be traded for happiness such as the time needed to develop a luxury resource, the time spent building and the additional upkeep to maintain a Stadium, or what you have to give to an adversary in trade in order to gain some resource she needs.

Finally, the player needs to understand how all wonders, technologies, special civilization abilities, civics, religious tenants, and ideology affects Happiness to fully master it. This is a task that will take most players a significant concerted effort to fully understand.

I would like to thank Caitlin Bever, Wenceslao Villaneuva, Matthew Langer, and Matthew Makuch for the great discussions we’ve had and for all the proofreading they did.  

[1] This analysis was done on Civilization V with both expansions Gods & Kings, and Brave New World installed.

[2] A “feature” in this case can be as large or as small part of the game as you like, use what works best for you. Smaller features are likely to give a more detailed analysis, but at the same time will take more time.

[3] Please note that the graphs are a qualitative estimate made by me using the described method, not the results of any actual scientific study.

[4] Please note that I have exaggerated this impact to create a more readable graph.

[5] This is given the assumption that you actually think Civilization V is a good game and doesn’t have serious flaws in its depth, complexity, or accessibility. If instead you have identified such short comings, then of course, you should adjust for that.