A Tool for Game Accessibility and Depth Analysis – Part 1

This is the first part in a three-part series describing a method for viewing, analyzing, and comparing depth and accessibility of games. The first part I provide an overview of the method, the second part goes into more detail, and the third part applies the method to a specific example.

The Problem

Your new game has been greenlit, a casual tactical shooter with enough depth to keep the hardcore players for years. Though everyone is confident in the game at the moment, you and everyone else know that it’s a big bet for the studio, and the consequences of not delivering could be dire. As the project continues, the question is; does the game you are working on a year later still fit the original plan for depth and complexity, or the one you are preparing to ship two years later? Is it actually casual enough not to scare off new players, does it really have all that depth that you promised it would have? What will happen if you add feature X? Will that make the game too complex? If so, is there another feature that you could drop that could bring the game back to the realm of casual? Not to mention, how does your game compare to the competition?

These are all very difficult, yet crucial, questions that you must answer in order to deliver the right game to your customers. You can use your experience and intuition, but it is often hard to accurately assess your own game when you’re deep in the trenches, and what you really want is a more precise and ideally more objective way to perform the analysis. This series of articles presents a method to analyze and compare games, and to track how your feature set affects your game’s accessibility and depth.

Definition of Concepts

I will begin by defining the terms that I will use in this series of articles more precisely, as many of the concepts in game design are vague, subjective, and/or lack an agreed up on strict definition. These articles concern a game’s complexity, accessibility, and depth, and the concept of mastering a game. By complexity, I mean the complexity created by the rules of the game together with the corresponding space of possible actions that the players can take that those rules give rise to. Note that the complexity of the rules and the space of possible player actions are not dependent on each other. A game like Go has very simple rules, yet the number of possible actions a player can take is quite large, hence the complexity of Go as a game is large. The depth of a game is closely related to its complexity, but rather than measuring complexity of the rules, it is about the complexity of play. A game that can be played using a simple algorithm lacks depth, while a game that allows for and necessitates a large variety of strategies and counters, and great variation of play have a large depth. Complexity and depth are not the same thing, though games that are not complex tend to lack in depth too. Mastering a game means both understanding a game’s complexity and its depth. This includes understanding the relevant rules for any situation, as well as knowing what strategies and tactics are applicable for that situation. Mastery, of course, is a matter of degrees, and a good way of thinking of a certain player’s mastery of the game is at what level she can compete. Finally, a game’s accessibility is a measure of how much effort a player needs to put in, in order to reach a certain level of mastery. The accessibility of a game is dependent on both its complexity and its depth, but also on things such as how easy the rules are to understand, and how much of the game has to be learned before a player can start playing.

Method Overview

The fundamental idea of the method presented in this series is to measure how much effort a player has to put in to fully master the game, that is to get a complete understanding of the game, its rules and mechanics, as well as its strategies and tactics. By mapping this out in the appropriate way, you can with a quick glance at the data determine how two different games compare and also see how the game you’re developing is tracking against its goals. Before we begin, if you are looking for some truly objective data, particular data that could stand up to academic scrutiny, then there are no short cuts. To gather data of that quality, you will have to gather a lot of unbiased data, and perform rigid statistical analysis on it. I believe that using the ideas presented herein would useful if you wanted to perform an analysis of that quality, but I expect most readers don’t need that kind of rigor. Instead, my suggestion is that you use this method as a lens through which to view games, a lens among many through which to analyze and compare games in order to find out what makes them work (or not).

In order to facilitate this second, and most likely more prevalent, use, it is important to present the data in a way that lets you compare two games at a glance. This is facilitated by ordering the data in normalized stacked columns with a consistent color coding[1] (see the first figure). What these graphs shows is a mapping from effort to mastery, that is the accessibility of Game 1 through 3. Each band of color represents a certain amount of effort, and where on the column that band ends represents the mastery achieved by that amount of effort. Note that since the Y-axis represents mastery, all columns are going to have the same height, which means that they are going to end at the point where the player has fully mastered the game, that is at 100%.

The process of mastering a game is divided up into three phases; the pre-play phase, the playing phase, and the research phase. Pre-play has two components; innate understanding of the game, and tutorial.

If the game is based on a concept that the player has knowledge of from the outside world, then the player understands some of the game, and this is the player’s innate knowledge. For instance, a game with a military setting normally comes with rules and mechanics inspired by how real-world combat and equipment works. Guns have a limited clip and needs to be reloaded, rifles have longer effective range than pistols, etc. Games may also use common conventions in their genre of games like established game controls and that using health packs immediately restores a certain amount of health to the player. The advantage of relying of innate knowledge is that there’s a part of the game that the player doesn’t need to learn, which allows the game to be more complicated without reducing its accessibility. Note that abstract games, like Reversi or Game 2 in the example, leverage no outside knowledge and hence the player has no innate mastery.

The tutorial of the game includes such activities as reading the rules of a board game, the player having the rules taught to her by another player, or playing a tutorial in a video game. Note that some video games have no tutorials and just drops the player into the game, while other video games have their tutorial spread over time, rather than having all learning being front loaded. Also, as a rule, only video games can get away without a tutorial phase, as the game itself can enforce its rules, something that a physical game cannot not. In the example, both Game 1 and 2 have tutorials that give the player a decent mastery of the game.

The next phase is the playing phase. I’ve divided this phase up into four parts; half an hour of playing, two hours, five and ten. They all correspond to the amount of mastery a player is expected to gain after having actively played the game for that long. In the example, you can see that after having played Game 1 for ten hours, the player is expected to have completely mastered it, while for Game 3, the player is barely scratching the surface at that point.

The choice of durations as well as those of the research phase is relatively arbitrary, and is informed by the kind of games that I am interested in analyzing. You may find that other durations work better for the kind of game you want to analyze, but remember that you can only compare two games if you have used the same durations.

The final phase is the research phase. At this point, the player is seeing diminishing returns from just playing the game, and will have to more actively try to improve her mastery. Researching includes activities such as experimenting with new tactics and strategies, reading online forums, books or other sources on the game, finding a teacher, and deep analysis of the rules. The research phase is divided up into five parts; one hour of research, five hours, ten, 20, and “extensive” research. The last category being a catch all for anything over 20 hours.

Now that we have a full understanding of what the graphs mean, we can easily look at them and compare the three games. Even a casual glance will reveal that the three example games are very different. A player playing Game 1 for long enough will have a full mastery of the game, while Game 3 requires the player to spend quite some time in order to reach full mastery. Game 2 is somewhere in the middle, a player will become quite competent from just playing the game, but in order to fully master the game, the player will have to put in some effort beyond just playing.

Of course, Game 1 through 3 are obviously very different games, and you expect to see very different profiles from that. Though comparing very different games can be interesting, a more likely scenario is that you want to compare similar games. Assuming Game 4 and Game 5 are very similar games, looking at their profiles, you can see that even though they both require roughly the same amount of effort of the player to reach full mastery, the effort required to attain partial mastery differs quite a lot, not to mention that Game 5’s tutorial is clearly more effective in teaching the game.

Some Examples

Lets’ look at a couple of examples of how to analyze games in order to get a better understanding of how it works. First up: Tic-Tac-Toe. Tic-Tac-Toe is an abstract game, so just looking at a board and the playing pieces won’t give a player any insight in how the game is played. However, after a rule explanation (that is the tutorial phase), you can reasonably assume that a player has a grasp of the complete ruleset of Tic-Tac-Toe. Knowing the rules is a good start, but there is one more thing that the player needs to learn in order to fully master the game, and that is the optimal strategy of Tic-Tac-Toe. Understanding and employing the optimal strategy of Tic-Tac-Toe guarantees the player a draw at worst. Most players will grasp this pretty quickly, particularly if playing against an opponent who already knows this strategy. The resulting analysis will produce the following chart.

At a glance, you can see that Tic-Tac-Toe is a rather shallow game. You can count on players grasping most of its complexity from a good explanation of the rules, and that after playing the game for a short period of time, most players will have fully explored the game. This doesn’t necessarily say anything about whether the game is enjoyable or not, but with a profile like this, the game is not likely to have a lasting appeal, at least not based solely on the merits of the depth of its gameplay.

Next, we’ll look at another board game: Pente. Pente is a variant on five-in-a-row that is played on a Go board, and has rules for capturing your opponent’s pieces. The rules of Pente are rather simple, and can be easily taught in a few minutes, and most people will quickly pick up on the basics, like avoiding having your pieces captured, blocking an opponent’s four in a row, and not leaving your opponent with a double open ended three in a row. However, unlike Tic-Tac-Toe, Pente does not have a trivial optimal strategy, and as a player keeps playing the game, she will discover more strategies and be able to identify more opportunities to exploit or dangerous situations to avoid and be able to do so earlier.

As the player spends more time with the game, the player will get diminishing returns in respect to how much her understanding of the game and her mastery of it increases, and at some point, just playing the game normally won’t suffice to further explore it. This is when the player will have to enter the research phase to improve her knowledge further. This could be experimenting while playing the game (e.g. trying different strategies and see what happens), experimenting with specific scenarios of the game (e.g. is there a way to get out of this particular scenario I’m having problems with?), finding a teacher, or taking the short cut and reading what others have written about what they have found out about the game. This may seem like an unlikely scenario that most players would put effort in to exploring the game that deeply, but today, thanks to the Internet, only a dedicated few actually has to do the work. If your game manages to build a viable community, even a small one, some of your most dedicated fans will do the job, and they like nothing more than for everyone to know about it, and will post the details on the forums. More people will read about it, and those people may teach their friends, and sooner than you expected it, the knowledge will be disseminated to a large portion of your player base.

Skill vs Understanding

It is worth emphasizing that what we are trying to assess is what it takes for a player to get a complete understanding and mastery of the rules and strategies of the game, that is the theoretical knowledge needed to play the game at the highest level. What we are not looking at is how to acquire and perfect the physical skills necessary to play the game. Fully mastering the gameplay may for some games make you a master, and for other games it’s not even necessary, and not fundamentally important, in order to win. The 100-meter dash would be an extreme example of a game where understanding nuances of the rules, and the best strategies will do little to improve your chances of doing well. There’s no doubt that a lot of planning goes in to the performances of the runners competing in the Olympics, but even if you intensely study these strategies to the point where you could coach the world’s top athletes, it won’t help you at your local track meet unless you practice actually running.

An example from the world of computer games is the differences between Quake III Arena and Counter Strike. I’m not a particularly good FPS player, and back in the day, I would get destroyed on a Quake server, but I could frequently get in the top 3 on a Counter Strike server, despite both games using the same fundamental mechanics to defeat opponents. The reason for my very different experiences was of course the difference in the rules of the games. Quake focuses on hand-eye coordination, movement and aiming, while Counter Strike has a different set of rules and mechanics, and because of those, positioning yourself on the map becomes crucial to successfully playing the game. By anticipating my opponents’ moves, I could place myself in such an advantageous position that I could defeat my opponents despite them often being much more skilled at aiming and shooting than me. This leads to the observation that what makes a game interesting is not purely dependent on its depth, but that you can make a game interesting and challenging by having more of a focus on skill and physical ability, rather than the ability to come up with a good tactic given the situation and the game’s rules. In fact, there are many popular games, that almost entirely focuses on physical skill or ability, for instance archery and weightlifting. At the same time, this example also leads to the interesting observation that a game focusing more on tactics and strategy will be less skill-based. Of course, in order to compete at the highest level, a player needs master both tactics and strategies, as well as having the physical skill to compete, as in the case with RTSs, and the exact mix is different from game to game.

Additional Factors

There are a few things that aren’t immediately captured in the analysis and the graph above that are important in order to fully understand a game. First of all, the actual size of the game, the rules and the strategies that a player needs to master differs quite a lot from game to game. Clearly learning Tic-Tac-Toe is a much smaller task than understanding Europa Universalis III. This may seem like a significant thing to leave out, but in the end, what we really are interested in is how long time it takes to learn the game, not how much one has to learn, and that is an important distinction. Smaller games are more likely to be learned quickly and with less total effort than a more complex game; similarly, if you find that two games of different complexities end up having similar profiles, then that may be an indication that the less complex game is bad at explaining to the player how it works.

Another important factor that the analysis so far has ignored is that it is not necessary for a player to have a complete mastery of the game in order for her to enjoy it. In fact, many games require only an understanding of a small portion of the game to be enjoyable, and very few players will ever approach anything near complete mastery of most games they play. For instance, it’s likely that you have tried Texas Hold’em at some point in your life, and enjoyed it, yet most people are far from a spot at the featured table at a WSOP tournament. I suggest that two additional points are important:

If the player can quickly start enjoying the game, then the player is more likely to continue playing and to spend the time needed to learn the game enough to reach the second point, where the player gets the full experience that you have designed for. With the full experience, I don’t mean a complete understanding of the game, but understanding all major features of the game, and knowledge of how to use them. For example, in Battlefield 1942, most players could quickly start running around in the game, shooting at the enemy, which is when most players would start to enjoy the game. However, you need to spend a bit more time in the game to master the different vehicles and weapons in the game, and understand how they interact with each other in order to get the full experience of the game. The player doesn’t necessarily need to understand that all interactions in Battlefield are built on a rock-paper-scissors relationship between the various weapons and vehicles, but understanding how to counter different approaches by the enemy is necessary. Also, details that may help an expert player, as the fact that the German tanks lack of sloped armor makes it easier to score a more damaging hit on them are great to know, but not necessary in order to fully access and enjoy the core gameplay.

If we were to look at Game 4 and 5 above, assuming that they are very similar games, and that in both cases the player will start having fun at around 25% mastery and get the full experience at 50%, we would see that it takes a lot less time for the player to start having fun in Game 5 as well as getting the full experience. This is something that should worry the makers of Game 4.

How to Use this Tool

How do we use this tool to improve our games? You can use this tool both to analyze the competition, and analyze your own game.

Analyzing your Own Game

Analyzing your own game allows you to test if the game you are actually making matches the assumptions you have made about how accessible your game is. You can investigate if new features are properly explained to the player as they are implemented, and crucially that they are properly taught to the player before the time the player needs to understand them. Generally, this is more of a concern for games that most players only play once, than games that are played repeatedly, as the player often isn’t expected to be an expert after the first game.

Another use is to keep track of how the complexity of the game develops over time. As you develop a game, you learn what works, and what doesn’t, and you will discover new and interesting features to add to your game as well as features that did not pan out and got dropped. This is a natural process, but it also means that since your game is changing, so does its complexity, depth, and accessibility. Tracking the current depth and accessibility of the game throughout development is very useful to avoid making a game that is too complicated or too simplistic. If you are tracking the game throughout its development, then it becomes easy to see what a proposed new feature would do to the game’s accessibility, or what a cut feature would do to its depth.

Analyzing the Competition

You can of course come up with many reasons why you would want to analyze the competition, but the foremost would be to see if there is a common pattern. If you do find a common pattern, then you know what works for that genre, or if you want to try something different, you can see if it has already been done and how well that went. You may also find that there is more than one popular game in the genre, and that they have very different profiles, like Quake and Counter Strike mentioned above, and that this is what allows them to co-exist rather than directly compete with each other since they are attracting different players, or at least offering different experiences to the players. If you do find that your game has a very different profile compared to all the games that you are going to be competing with, then recognizing this is obviously important. If you want to keep the design of your game, then recognizing that it is different will be important for your pitch, identifying your target audience, and when verifying your design, or it may be a good indication that your game has features that aren’t entirely suited for your chosen genre, or that it is missing some crucial part that other games have. Drastically changing the complexity of a game isn’t inherently a bad idea, tower defense games are essentially RTS games where most of the elements have been removed which leaves the player able to focus on just one aspect.

You can find the part 2 here.

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] Dubbed a Tobias strip by my first proof reader.