We Should Stop: Simply Calling Games "Difficult"

We Should Stop: Simply Calling Games "Difficult"

In We Should Stop, The Head Pirate looks the things we all say and do as gamers, and picks apart the things we all do but really shouldn't.

Defining difficult is … what’s the word for it …

When you put a simple game of chance under the microscope, you can learn quite a bit about the most common mistakes we make when speaking to a games difficulty.  Picture a game where you role a 20 sided dice and “win” on a roll of 1-10 while you lose on a roll of 11-20.  In mathematics we call this type of game “fair”; you win as often as you lose.  There are a number of ways could modify the game to be “unfair”, the most obvious is to change the range of winning and losing values. We could also use a “loaded” dice where the probability of landing on any given value if not uniform, changing the games mechanics, or could add additional rules to the game, like saying you can never win on the same number twice in a row.  But have we made the game any more or less difficult?

Although we've modified the chance of winning and losing we haven't changed what has to be done to complete the game. The TASK required to reach a result is the same, and this task serves as a meaningful definition of game-play .  The only thing you have any control over is if you successfully manage to roll a dice such that it lands on a face and can be read.  As anyone who has ever had friends over for board games on far too small a table can attest, it’s hardly a guarantee … but few would call this task difficult.  As such, we need language to separate a game that requires you to do something that requires a high level of skill, like tossing a dice between a series of rings for the toss to count as a way of making you win less frequently, from other games that ensures you win less frequently by saying you only win if you roll three 20s in a row.  In the first instance, game-play is far more difficult, while in the second who wins or loses is still determine entirely by chance*

All your base(line)

When we move over to Video games it’s often less obvious what is the task we have control over and what is left to chance, and frequently they overlap.  To understand this, we’ll take a look at an action sequence that requires reactive button inputs like a boss fight in Devil My Cry, Bayonetta or Dark Souls (or any other brawler type game).  In this scenario the “boss” will use a combination of abilities with various telegraphs which require you to use correct counters, each mapped to a different button on the controller.  If you are successful, the boss takes damage and if you fail you take damage instead.  The encounter plays out in a set order; that is to say the boss uses the same attacks in the same order every time it is attempted.  This task is entirely skill based and has no random component which allows us to assign it an arbitrary rating of difficulty: let’s say 10. We’ll very quickly notice an interesting anomaly.  Although the encounter never changes, people who have played though it a number of times will have a higher success rate as they are able to apply both reflex and memory to the task.  Therefor any meaningful definition of difficulty must allow for that success rate to increase with repetition.  Practice might not make perfect, but for a task to be defined as “difficult”, it has to have some measurable effect.

With a baseline defined, we can look at ways of changing the experience and what effect that has on the difficulty using the arbitrary scale given above.  We already understanding that difficulty refers to a task, and that the task here is reacting to a telegraphed attack with the correct input using both reflex and memorization.  By lowing the amount of time we are given to input the correct response we can made that task more challenging without changing the structure of the task itself.  If, for example, we have 1 second to react in the game we gave an difficulty rating of 10 to, with only half a second might have a difficulty of 15, while a game that gave you 2 seconds to react would have a difficulty of 5.  This limits a meaningful discussion of “difficulty” to platformers, brawlers, fighting games, and other games where the primary challenges come from some primarily reflex based task.  Within these categories, this difference is rather obvious.  It’s easy to argue that Super Meat boy is more difficult than Super Mario brothers, or Street Fighter is more difficult than Mortal Kombat or Skull Girls.  But what about changes that effect something other than difficulty but still make it less likely we’ll complete the encounter?  What can we call them?

Getting Subjective

Most commonly, we would see this type of encounter move away from scripted attacks towards random patterns that are different every time it’s attempted.  Although this gives a perception of making the encounter more challenging by forcing the players to use reflex alone, that’s not really what happens.  The player is still building and applying memory to the individual telegraphs, and that’s good news for our definition; the encounter is still difficult.  Only now instead of a set difficulty of 10, it would be a range of values based on what exactly the random sequence ends up being.  This can be a problem when trying to review or talk about a game that utilizes this technique.  One reviewer might end up getting a random difficulty (using our arbitrary scale) of 5 and say the encounter was too easy, while another might get a random difficulty of 15 and say it’s too hard.  Both are correct, as the experience is now subjective. CCG and puzzle games are the most common example of this type of subjective difficulty, although it can be quite prevalent in RPG games, particular in boss fight.  It’s clear that the type of gamer who enjoys playing Magic That Gathering is generally different from a gamer who enjoys competitive Street Fighter, so this distinction from games with objective or set difficulty is extremely important.

The Unforgiven

Adding a life bar, or same amount of variability to the number of times we need to succeed at entering the correct inputs or the number of times we can fail to do so is another common way a developer might modify difficulty.  By forcing the player to “pass” a given challenge (set of tasks) 3 times in a row, for example, the developer will significantly lower the number of times the attempted encountered is completed successfully.  In addition, the developer could force you to reply content before you get the opportunity to retry the encounter you failed, both increasing the amount of time that needs to be invested in each attempt, and limiting your minds ability to form muscle memory (as you are forced to shift to other tasks to fight your way back to the failed encounter).  Some games may even take it a further as a single mistake forces you to restart the game from the beginning.  Even in these extreme cases, however, the difficulty of the gameplay remains unchanged; we perceive the game as harder because the time investment needed to successfully complete it is increased.  It’s quite literally the oldest trick in the developer’s handbook.**

Going back to our definition of difficulty, we were careful to separate things the player has control over from random events.  The trick in this type of modification is that it misrepresents player’s short term (immediate) skill level as controllable when it isn't.  Although we can get better at things over time, how good we are at something right now is a set number which can be defined as a probability of succeeding at a given task.  For example, if you are able to complete an encounter with a difficulty of 10 (using our arbitrary scale from above) 50% of the time, how often you succeed at that take 3 times in a row is a simple function of portability outside your control.***  More importantly, we've learned nothing about this players ability or skill. He may simply have “lucked out” and preformed the task at the absolute limit of his abilities multiple times and be completely unable to complete a task with a higher difficulty.  On this plus side, this can lead to an extreme feeling of gratification when a player achieves a sequence of successes which, to them, seems far more difficult than it actually was.

Games that utilize this technique are so popular that it’s spawned an entire genre (roguelike) and games like BloodBorne  or Dark Souls have received almost universal praise for their use of this technique.  This is why it’s so important to be able to talk about them using the right language.  Given the vast variation in core difficulty (or difficulty in task) of different rougelike games which are enjoyed by the same type of gamer, it’s obvious the sought after property is something independent of it.  In the business and legal worlds we call being able to fail (or default) at something without penalty “forgiveness”, and so it makes sense that we label these games as unforgiving.

Is this getting too complex?

Last but not least, addition rules and button combinations could be added to the encounter.  There is an appreciable difference between pressing “x” to react to an attack and pressing x, y, x, x, x while holding left, a and b at the same time.  Alternately, the rules could change throughout the encounter; if you pressed “x” three times in a row, the next time you should press “x” you need to press “y” instead.  Unlike difficulty, which focuses primary on reflex but allows for memory as well, this type of game focuses almost entirely on memory and cognitive ability to define a new way and encounter can present a challenge.  This is even clearer when you look at turn based and strategy games which eliminate reflex skill although while still maintaining meaningful differences from each other in terms of challenge.  I don’t think anyone would say that Civilization is any easier than Crusader Kings, but Crusader Kings has a much larger number of systems which the player needs to keep track of.  These games are best evaluated by the number of systems and mechanics the player is required to track and master, best labeled as the level of complexity.

The Final Word

In a perfect world, we call Super Meat Boy a game with extremely high objective difficult that is very forgiving and not complex, while we say BloodBorne has moderate subjective difficulty and complexity, while being extremely unforgiving.  Civilization has low subjective difficulty and a high level of complexity.  Only then do we all mean the same thing when we talk about a game’s challenge.  As it is now, we boost about clearing a game on nightmare or making it though hard-core mode while that doesn't necessarily measure skill.  More importantly, reviews and media lack a powerful tool in matching games with the people who enjoy them.  I personally don’t enjoy unforgiving games or subjective difficultly, but I love complexity and games that are objectively difficult, while my wife will spend hours perfecting a subjectively difficult, insanely unforgiving game, but can’t stand complexity.  As far as any gaming review site is concerned, however, we both like “hard” games.

I think we can do better.

* We seem to understand this and use different language outside of games.  Although the chances of winning the lottery are far smaller than the chances of becoming a millionaire by going to school to be a doctor and developing a successful practice, no one would ever say winning the Lottery is more difficult.  We clearly see the “task” of becoming a doctor being more difficulty then the “task” or buying a lottery ticket, and don’t confuse chance of success the same way we do with games.

**When Video games first came into being, this was the first trick developers used to make games more challenging to increase their length.  A game which offered no continues was perceived as “harder” then one that didn’t regardless of the difficulty of the tasks the player was being asked to do.

***You would succeed .50 * .50 * .50 percent of the time, or 12.5%