Something that games have that other media don’t is difficulty. I mean, I guess there are ‘difficult’ books and films and so on, but the difficulty is conceptual rather than forming any sort of actual obstacle to progress. Finnegan’s Wake won’t lock a chapter until you can demonstrate that you fully understand the previous one (and thank heaven for that); only in games is there this dynamic interaction in which the player has to prove their worth before they can access more of the story.

It might seem obvious, but I think it’s interesting to note that there are different kinds of difficulty. I suppose I mean this in two connected but distinct ways, actually. The first sort of difficulty lies in the fact that what makes a puzzle game difficult is going to be very different from what makes a shooter difficult. The key to the difference there is probably the type of skill involved; to overcome progressively difficult puzzles, one needs progressively better problem-solving abilities, whereas to succeed at an FPS at a high standard requires good reflexes and quick thinking.

However, it’s not just in the skills involved that a game can alter the way in which it makes itself more or less challenging. Across any genre, no matter what sort of gameplay it has (and consequently what sort of obstacles it presents), a game can raise the level of difficulty through multiple methods. This second family of difficulty types is, I think, the more interesting.

Perhaps we could refer to these models or families as the skill-based types of difficulty – which make demands of the player’s ability to react quickly in an action or rhythm game, or make sense of a challenging environment in a platformer – and the mechanical types of difficulty, which I’ll get on to in a moment. Alternatively, we could just call them the external types and the internal types. Hopefully the reasons for this will become clear as we go along.

Within the internal family, there are three ways that I can think of in which games can alter how difficult they are without changing the skills demanded of the player. This is why I’m calling them the mechanical or internal types, because the change happens in the mechanics of the game rather than outside the game. Thinking about it, I suppose you could say that the skill-based family ought to be the internal one because it’s to do with things internal to the player, but I’m using ‘internal’ and ‘external’ from the point of view of the game itself and not the player. An interesting point to note is that all external switches (so any time a game alters its gameplay so as to change how difficult it is by requiring completely different skills) are qualitative, whereas internal ones can be either qualitative or quantitative.

The first method of raising the difficulty of a game without changing the skills required to be good at it is to do so statistically. The vast majority of games, including pretty much all those with any kind of combat, will at some point do some sort of number-crunching to work out whether the player is successful. The most obvious example is probably the Bethesda-style RPG: Skyrim or Fallout 4, to give two recent examples. In this sort of game, the player character has expressly quantified attributes; how good they are at fighting, healing, moving or persuading is expressed simply as a number. To make something more difficult is as simple as requiring a higher number. A simple illustration might be something like the lockpicking skill common to several RPGs with stealth elements; if Hero McProtagonist has a lockpicking skill of 45, then a lock rated 20 should be easy while a lock at 80 is going to be difficult if not impossible. A more complex series of calculations go into something like Pokémon‘s battles, which compare attributes like the attacking mon’s attack stat, the power of the move, the other mon’s ability to defend or evade and any other situational variables before expressing the whole thing as a quantity of HP knocked off by the attack.

This kind of statistical variation is probably the one that crops up most in games with a difficulty setting. There are those which tie the nominally simple ‘easy’, ‘medium’ and ‘hard’ (or funkily-named variants) to something deeper and more complex, but in most cases raising the difficulty is likely to give enemies better stats and that’s about it. It’s possibly one of the lazier ways of achieving the effect, but there’s certainly nothing wrong with it; the only way most RPGs are able to function is through statistical manipulation, since if everyone had only 1 point in every stat, all the time, battles would basically just consist of each person dying in one hit until either the heroic or villainous side won by virtue of having more people to fall back on.

Version two of the mechanical approach to difficulty involves adjusting how hard it is to understand what needs to be done in order to proceed. This isn’t quite the same as the skill-based approach, although a game that suddenly switches from bullet-hell gameplay to out-and-out puzzling could certainly be argued to be making itself harder to understand. I’m thinking more about a situation in which the gameplay doesn’t change at all; rather, it becomes harder to work out how to apply the same mechanics. Undertale does this pretty well by progressively making it more and more difficult to figure out what particular sequence of actions the player needs to take in order to spare each enemy; at the outset, a single turn spent having a chat or quickly having a little dance might be enough to have an enemy become amenable to ending the conflict without violence, but by the end of the game it becomes a struggle to determine and sequentially carry out the many actions required before the foe’s able to kill you.

I think that combat which becomes harder not simply by raising the enemy’s health and defence, but by having them be more proficient, harder to avoid taking damage and harder to attack back, also falls into this category. It might involve raising the speed or range of their attacks, but the end result is that the player has to be careful and tactical about how they approach a fight. Dark Souls isn’t a bad example of this; the statistics of a fight, while they can make a difference, aren’t as important as each enemy’s unique moveset. Grinding to raise your stats might help a bit, but it won’t get you nearly as far as studying your opponent’s patterns and becoming better at staying away from their attacks entirely. Ultimately, although every enemy can be defeated using the same fundamental skills, some are indubitably harder than others because those skills are going to need to be applied more cleverly and carefully. It’s a tactical sort of thing. I think that strategy games probably fall into this category for this reason; one of the few RTS games I’m familiar with, Civilisation V, adjusts its difficulty by making the AI opponents smarter and therefore requiring the player to apply their own gameplay skills more cleverly. (There might also be some statistical manipulation in there, which goes to show that it’s totally possible to utilise more than one of these methods.)

This second type of internal difficulty is more involved than the first, since the player needs to improve their own abilities and not just their character’s stats, but it can still be misapplied. Dark Souls practically made its name from being considered unusually difficult, to the point that some thought that it was unfair. I actually don’t think that it does ever cross the line from ‘hard but fair’ into ‘unreasonable’, but I can certainly see that it would be possible to make a game so demanding on the player’s ability to comprehend it that it would become totally opaque. (Undertale, as I’ve discussed before, has been criticised for making it unfairly hard to figure out what to do in order to get the ‘good’ ending and therefore succeed in the way the game clearly wants you to.) I Wanna Be The Guy might well be another example of a game making it so ridiculously difficult to understand how to apply its own basic mechanics that it becomes nigh-impossible. I mean, the only actual skills involved are running and jumping, but it’s so stupid hard to run and jump in the right way when things that look like background objects start falling and killing you on the very first screen.

Method number three is a bit nebulous, but the idea is broadly that the player is put at a disadvantage in some way which is neither purely statistical nor strictly related to their ability to discern how to apply their skills and then do so. The main thing I’m thinking of here is something like the trick the Kingdom Hearts games pull when they cause the player to unlock new abilities at different levels depending on the difficulty. Yeah, changing the difficulty level will also put the player at a quantitative disadvantage because enemies will have better stats, but KH makes sure that it’s not just a question of changing the numbers by giving the player helpful abilities earlier or later in the game. In the Final Mix version of the first KH game, for example, the skill Leaf Bracer allows the player to be briefly invulnerable while casting healing magic. This is an incredibly useful skill; without it, enemies can interrupt the spell and deal further damage, which could be fatal. Depending on the player’s choices at the beginning of the game, Leaf Bracer unlocks at either level 27, level 39 or level 69. You can see how this completely alters the difficulty of the game; you might be able to reliably heal yourself without fear of interruption by the time you reach the mid-game, or you might never unlock the skill at all unless you spend hours grinding (the game can be completed at well below level 50 on lower difficulties, even on a first run).

I can’t think of many other examples of this third type, though I’m sure there are plenty of games which start the player out with better equipment or abilities on lower difficulties, even if only as an Easter egg (think Resident Evil 4, which allowed players who’d already completed the game to unlock an infinite-ammo rocket launcher and in doing so basically provided a New Game+ at the lowest possible difficulty). Actually, come to think of it, I Wanna Be The Guy kind of does this too. The gameplay itself will never change, but higher difficulties mean fewer save points, which means that the player has much further to go in a single run without dying.

So that’s my theory of difficulty. Quick recap!

• Two families of difficulty types: internal (or mechanical) and external (or skill-based).
• External types of difficulty boil down to genre and the player possessing different skills in order to succeed at the gameplay.
• Internal types change the game itself, not the player, which can be achieved through statistical manipulation (including better or worse odds in the RNG if applicable), tactical depth or other disadvantage.
• Games can use as many of these types as they want, and in fact I reckon those with the most ‘dynamic’ or ‘interesting’ challenges are probably the ones that employ a bit of everything.

So what do you think? Are there other ways games make themselves harder? Maybe you think that horror games have a unique sort of difficulty because they actually make it hard for the player to want to continue in case something scares them! Let me know.