This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.
In the first part, I sketched out the fundamental idea of my method to analyze gameplay depth and accessibility in games, this second part will go into more detail, as well as point out several pitfalls and subtleties that you need to keep in mind when applying this method.
Two fundamental questions: How do you know that you made the correct analysis, and is the whole method not too subjective to be useful. The answer is that the method is subjective, but that that isn’t a big problem. Part of the point is just to think about these issues and to make conscious decisions regarding depth and accessibility, and to sanity check how the game is presented to your players. It isn’t really a big issue if your analysis is off by 5 percentage points here or there, as long as you’ve done the work.
If you do want more accurate numbers, (and you do want that, right?), you can use user tests to check them, but since user tests come late in a game’s development cycle, and because they are quite expensive, you probably won’t want to rely solely on them. If you do have the budget and access to do early user tests, even partial ones can provide you with good information and allow you to make the needed adjustments earlier in the development process. Another good way to improve your numbers is not just doing the analysis yourself, but rather let several people do it, and identify where the different analyses differ, or just use the average result.
Also, if you go into a detailed enough level, the objective answer may be more apparent. You can know exactly when a specific feature is mastered if that feature is introduced at a given point in your game with a tutorial followed by a couple of scenarios that forces the player to use that feature in order to advance, and that that fully explores the feature in question.
Finally, you may want to err on the conservative side of things. It is easy to think that a system that you designed and work with every day for the past year would be easy for an outsider to understand, and failing to see the peculiarities of your system and just how difficult it may be to penetrate it if you do not have all the knowledge that comes with actually having constructed it in the first place. This is another good reason why you would want others to do their independent analyses to check that there’s nothing you have severely underestimated.
Let’s look at chess. Chess is arguably one of the most successful games in history. Chess as we know it today formed mostly in 13th to 16th century from more ancient roots, and by the end of 15th century the first book on chess was written. Today, chess is played all over the world, countless books have been written about the game, it was long used as the foremost challenge in Artificial Intelligence as the measure to “beat” human intelligence. In order to become a top-level player, a Grand Master, a player must put in an amount of effort and time comparable to that needed to become an elite athlete, and challenges for the title of World Chess Champion are international news.
Performing our analysis on the game of chess then should reveal a game with tremendous depth compared to the games we have looked at so far, and indeed almost any game that we could imagine. However, actually looking at the game, you will find that it has fairly simple rules and a limited branching factor compared to many of the modern strategy games that exists today. For instance, a game like World in Flames probably has a depth and complexity that is many orders of magnitude greater than that of chess. More importantly, if chess was invented today, rather than centuries ago, it is unlikely that it would occupy the same place in our culture, and the game would never be explored to the depths that it has been today, in other words, many games have a potential depth that they will never realize because of the sheer amount of work that has to be done in order to fully explore them.
When analyzing your game as well as comparing it to other games, what you need to do is to pick a time horizon that is reasonable for you game. Computer games have fairly short lifecycles, particularly singleplayer games that few players will go back and play a second time, board games don’t age in the same way, but if you look at individual copies of the game, most games are not played that many times (I’m sure even your favorite board game hasn’t seen that much use). What you have to figure out is how much of the game’s complexity your complete body of players is going to explore. The upper limit for the game’s complexity will be the sum total of that.
There are, of course, many games that will be played for much longer, and/or much more intensely. Singleplayer games that are meant to be replayable, like many strategy games, will be much more thoroughly explored than a game based around the escapism of a story driven adventure. The kind of games that are likely to see the most intense exploration are long lived competitive multiplayer game, and if you are making one, you can reasonably assume that there will be a large number of players that will put a lot of effort in to understanding your game.
It is also worth noting that unless your game is truly unique, it will have constituent parts and features that are similar to other games, and the knowledge that has been gained from those games by your players will be brought to your game when they are trying to understand it. That means that if you assume that there is a limit to the total amount of time people will spend on understating your game, then that time does not have to be spent on understanding already well-established patterns that you have used in your game.
This may actually be a trickier question to answer. Are we looking at someone who hasn’t played many games at all? Someone who is new to the genre, someone who is well versed in similar games, or someone who has played the two previous installments in your series? The answer is that you may want to create a profile for more than one of these players if not all of them. You want to make sure that a player that doesn’t have much knowledge from other games can gain that knowledge in your game, while at the same time ensure that you don’t bore the expert by forcing them to learn about things that are close to second nature to them. For example, Tic-Tac-Toe has a very different profile if your player is a 5-yearold rather than an adult, and you may discover that for that player, the game is less accessible, and that is won’t be fully explored in half an hour, which gives it a more desirable profile.
An interesting special case is DLCs and expansions. Most DLC packages only add limited complexity to a game, but they are at the same time frequently bundled with the full game for new players. These new players will now have to not only master the original game, but also the additional rules of the DLC, potentially turning your easily accessible game into something else. The same observation also applies to games as a service where a game gets updated over time, with more and more features added to keep the interest of the existing players.
The actual effective depth of your game is not only determined by the complexity of the mechanics of the game, but also by how much of that complexity is necessary to master in order to master the game. If there is an optimal strategy, as in the case with Tic-Tac-Toe, and if the player does not need to understand the full complexity of the game in order to derive and use that optimal strategy effectively, then the rest of the game’s depth is irrelevant and hence the game’s effective depth is limited by the depth of the optimal strategy.
Though optimal strategies can occur in multiplayer games, these are usually quickly patched by the developer, for our purposes, they are much more of a concern for singleplayer games. Human opponents are able to adapt to a strategy, particularly if they possess a deep enough understanding of the game, but most AI opponents in singleplayer games are static, and the game will be effectively mastered if the players develop a strategy that the AI cannot beat. In fact, for singleplayer games, the player does not even have to develop an optimal strategy to win, just a strategy that the AI cannot defeat. Such a “good-enough strategy” is for our purposes the same as an optimal strategy and will have the same effect of limiting the effective depth of the game. Obviously, few people want to consciously design such an optimal strategy that undermines a good portion of the more advanced mechanics into their game, and much has been said about how to try and avoid it. If you analyze a competitor’s game that has an optimal strategy, then it might be worthwhile to do a second analysis in which you disregard the optimal strategy to see what the game was actually intended to be like. Also note that many games have other mechanisms to work around a limited challenge provided by their AI, like grading the player on how much time she needs to win, or how big a margin she has to the runner up.
Of course, if there is an optimal strategy, but it requires a full mastery of the game, then the existence of the optimal strategy, though not necessarily desirable, doesn’t change the effective depth of the game.
Once you have performed the analyses of your game, and determined its particular profile, you have to decide if it is the profile that you desire. There is no right or wrong profile, instead the question is if the profile is suitable for the game. If you are making a competitive multiplayer game, then you want most players to be able to grasp the basics quickly so that they are able to compete, but depending on your game, you may want to make mastering the game take months or even years if the challenge comes from complex mechanics, while on the other hand, your game may be more skill-based in which case you might want to limit the depth of your game to allow the skill to come to the fore.
Also, note that accessibility as the term is used here has a direct mapping to time and effort needed to become good at the game (in addition to whatever physical skills the player needs to acquire). Depending on your game, the amount of time you can ask a player to invest in learning to play your game varies, that is a deeper game is not always a positive. The appropriate depth of a casual word-based game meant to be played in one minute chunks during the player’s commute needs be a lot easier to master, at least to the level where the game becomes enjoyable, than a historically accurate simulation of the Thirty Years’ War.
In singleplayer games that are not meant to be re-played, you may want to create a profile having the player exploring and learning the mechanics up until the very end. Spreading learning out over the entire experience puts less pressure on the player to learn a lot of new things at the beginning, and for the same reason, it allows more features to be introduced over the duration of the game. It also helps keeping things fresh for the player if the player is continuously presented with new things to learn and explore, and the analysis can find potential dead zones where nothing new is presented. At the same time, you can use the profile to check that features necessary to have been mastered to progress past a specific point in the game, are indeed likely to have been mastered at that point. It might be a good idea to alter the tool when using it in this way. Progression can be a more useful measure than time spent if you intend to use the tool to verify that the player has the enough time to learn a feature before not having mastered it becomes a progression blocker. This is particularly important in games with a significant number of optional side quests, where different players can have spent very different amount of time with the game before reaching a specific gate.
You can find the third and final part 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.