Wednesday, October 26, 2011

The Automation of Player Skill

In a recent feature on Gamasutra, "The Abstraction of Skill in Game Design", Josh Bycer created a framework for understanding the degrees to which games abstract or "filter" the abilities of the player through the skills of an in-game avatar or the game mechanics themselves.  Upon reading this piece, however, I found myself wanting to more deeply explore the nuances involved in the player-skill equation.  Consequently, in this article, I'll be discussing some of the more specific ways in which developers automate or even apprehend player skill, much to what I believe is the detriment of the gameplay experience.

Let Me Get That For You

The most obvious way in which games automate player skill is quite literally by, well, automating things for the player.  Whether this manifests as auto-aim in a first-person shooter, your squad automatically taking ideal positions and weapons in a strategy game, or your character automatically gaining skills in preset areas in a role-playing game, the end result is always the same: the player's direct control over the action and the game mechanics are suspended and superseded by the game designer's own will.

I'll take the most common example of this, the auto-aiming in a first-person shooter, most commonly seen in console games.  Auto-aim is typically considered a necessity by shooter developers, whether or on console or PC, mostly in order to provide newer and less experienced players with the ability to get through the game without necessarily possessing a high skill level.  As an accessibility tool, this is all well and good - after all, it makes little sense to deny a new player or a disabled gamer the "privilege" of finishing the game.  However, on consoles, auto-aim is more or less a universal constant: finding a game without it is impossible, and finding one which lets you disable the feature in the first place is often quite difficult.

Shooters on consoles can still be fun and complex games, but taking out the aiming reduces skill involved and cheapens the play experience
Shooters on consoles have rarely worked well for me.  Although I can enjoy the spectacle, the visceral thrill of firing at enemies (or friends in multiplayer mode), as well as the story and characters, rarely do I find myself enjoying the gameplay in particular, the fundamental "bread and butter" gameplay mechanics which all shooters revolve around.  I think, at least in part, a lot of this comes down to the fact that so much of my play experience feels automated as a result of auto-aim, and, in more recent games, the snap-to feature when aiming down iron sights.  Sitting down to play Call of Duty: Modern Warfare 2 with a friend in split-screen, it often feels as if there's a third presence in the room - an invisible game designer sitting between the controller and the television, who anticipates my own movements, says "oh, you want to shoot the guy on the left?  I see, let me get that for you!" and pulls my crosshair in the right direction.  Actually winning in such a scenario feels less about my skill in the fundamental act of shooting, and a lot more to do with my ability to simply select the right targets at the right time.

Now, don't get me wrong, shooter developers already know this.  Bungie, pioneers of the modern console shooter themselves, were smart enough to anticipate this problem in the original Halo, and designed everything from the game environments to the enemy AI around auto-aim, the player's more limited degree of movement, and slower aiming speed.  Furthermore, in downplaying the skill importance in shooting, they were required to compensate by emphasizing other areas of the game, such as more distinctive weaponry with strong balance, the addition of vehicle sections, AI comrades who fight capably alongside the player, etc.  I don't mean to say that it's not possible to make a fun game with auto-aim, either, because clearly I'd be in the wrong, and I've enjoyed many shooters with degrees of auto-aim built into them over the years, both on consoles and PC.  Despite all this, though, the movement of auto-aim from an accessibility and difficulty feature into a standard feature which games are designed around feels, to me, like the epitome of dumbing down the game experience for the player - rather than having to master the game mechanics in order to win, now I am made master by default, with a silent, invisible hand guiding my way all along.
The Reduction of Player Judgement

Another recent example of this I ran into, in a completely separate genre, was in Neverwinter Nights.  I've always had a strained relationship the game, having played it since its release on and off but never really getting far into it or particularly enjoying it; I've always felt like I've been missing something, as if there's something about the game that everyone loves that has flown right over my head.  Now, I believe I've cracked the code, and it lies in the removal of the player's judgement in utilizing the skills he or she might have.

Despite being a solid implementation of the Dungeons & Dragons 3rd Edition rules (a considerable step up from the "Advanced" D&D rules of earlier CRPGs), Neverwinter Nights is also extremely automated.  Rather than building one's own party and then directly controlling each of them in order to overcome tactical combat encounters, instead the AI takes over everything from the inventory management, equipment, leveling up, and fighting in everyone but the "main" character.  Supposedly to speed up the game pace and give the companions more personality, all it does is utterly sap the life out of the game by turning combat into a boring, repetitive slog where you spend 95% of your time watching your party auto-attack their way to victory.  Even as a spell-caster with lots of active abilities, ultimately gameplay comes down to simply hitting a few hotkeys when necessary, and looting the occasional chest.

Neverwinter Nights loses much of the inherent depth in its ruleset by automating many aspects of play, from combat to feat and skill use.
Furthermore, when the AI breaks down, it often takes the player with it.  I recall a recent attempt at a battle against a high-level spider enemy, wherein my AI companion, rather than taking the fight and absorbing some of the hits for me, instead decided to run off to a nearby trap and attempt to disarm it, while my player character was forced to take on the spider alone.  As if that wasn't annoying enough, my companion apparently failed in disarming the trap, and the resulting fireball blew both himself and my own player character to kingdom come.  I was not amused.  At least I had a quicksave on hand.

Then there are the more subtle ways in which the game automates player actions.  Going back to traps, rather than having to search the environment manually, either by using visual cues, logic and reasoning, or even just entering a "search mode", instead the function takes place automatically (at least when one has an AI companion capable of the feat), and almost never fails.  This might seem like a small thing, but it makes all the difference.  In a CRPG, the player's skill comes into play not in terms of twitch reflexes, but in terms of making the right decisions in succession, based on strong judgement and knowledge of the game mechanics.  The Infinity Engine games such as Icewind Dale turned traps into a sort of resource management of their own.  Usually it was good practice to scout an area with a stealthy rogue or character made invisible by way of a spell, both for enemies and for traps.  However, often traps would be placed near enemy encampments, and disarming them would break stealth and expose your squishy mage or rogue.

Hence, trap disarming was an interesting risk and reward decision tree - one that a player could ignore entirely should he or she want to put up with the occasional inconvenience traps cause, or could handle intelligently.  Indeed, much of the fun in playing a rogue or magic-user is in using those unique skills to one's advantage to tackle the game's obstacles in interesting and creative ways.  In Neverwinter Nights, the entirety of the nuance of detecting and disarming traps has been completely and utterly removed, replaced with a skill-less automation which not only reduces the depth of the game considerably, but also makes all those complex feats and talents a player can invest in a lot less interesting as well.  Rather than taking advantage of the inherent depth in the D&D rules to provide the player with interesting skill-based gameplay (reasoning, logic and risk-management), instead, Neverwinter Nights decides for the player how to play the game, and ends up actually being a less faithful implementation of the tabletop game - even without changing the rules themselves.

Cinematic Experiences

One of the most common points of praise for modern videogames is when they are compared favourably to movies.  Many game franchises have made it their key selling point in order to provide the player with a "cinematic experience", to place the explosions, characters and dramatics all around the player as the universe's centre.  I certainly won't deny that a cinematic game can be enjoyable - as I said, I love spectacle, and despite not being much more than an on-rails shooter, I can't help but be visually stunned by a game like Call of Duty, whose sole mandate appears to be to provide the player with a roller coaster ride of action and excitement.

However, rarely do we really see how this focus on cinematics can hamper the game experience itself - in fact, most players and developers alike tend to assume that those cinematic elements are good in and of themselves.  Eurogamer's Simon Parkin recently posted a review of Uncharted 3: Drake's Deception in which he gave the game a solid score, but provided a strong critique of some of the game's limitations - most notably, as stemming from developer Naughty Dog's attempts to provide the player with an action movie's worth of excitement.
"But it also reveals another truth. Uncharted 3 is the most exciting game in the world, but only until you deviate from the script. Even in this chase the conflict between the developer's theatrical choreography and player-controlled interactions is clear. In order to ensure each set-piece is set off correctly, the game commits the cardinal sin of insinuating you have full control of your character, but in fact tugging you towards trigger points - making sure you're in the right spot to tumble over the bonnet of that braking car, for example."
The point of criticism is clear: while Uncharted 3 might be an incredible thrill-ride, it's also, in some senses, the worst game in the world: control is apprehended so that the choreography of the piece is maintained, and the game's pacing moves not  out of the player's own actions, but out of the developer's want to adhere to an extremely (and exceedingly) specific flow of events.

Uncharted 3 provides enthralling visuals, but in automating much of the gameplay, it can feel as if the player is just getting in the way of the action.
There are two problems most notable with Uncharted 3 as I can gather (and apologies if I'm wrong, as I haven't played it myself).  The first is that oftentimes the player doesn't feel in direct control of the experience.  As Parkin states, often he found protagonist Drake running and jumping around in ways which clearly defied the (in-game) laws of physics, and his own expectations for the character, in order to make a cinematic leap forward or avoid the game over screen.  Not only does this break immersion and the player's understanding of the game world's rules, it also takes control away and subtly tells the player "hey, you're doing it wrong, move over and let me handle it."

Second, oftentimes the game simply refuses to work in a way the player might anticipate or expect, especially in cases when the player defies the wishes of the designers.  I've seen this plenty of times in the past, with absurd instant-death scenarios for stepping outside the dotted line, though much less common in modern games, and most developers have got pretty smart about providing realistic, natural, and coherent explanations for why the player can't do X or Y.  Yet Parkin writes,

"Your freedom of choice risks ruining the shot. Indeed, throughout the game, if you jump into an area you are not supposed to visit, Drake will crumple on the floor dead, Naughty Dog switching role from movie director to vindictive god. That is not your predestined path: Game Over."
Uncharted 3, put simply, doesn't allow the player to experiment, to explore, or solve problems in creative ways.  Instead, everything must adhere to the tightly controlled script that Naughty Dog has laid down, and deviating from that path is met with an instant failure and reload.  The most absurd thing about such a limitation is that Uncharted has always, to a degree, presented itself as a game about adventure and exploration - the protagonist, Nathan Drake, is a modern-day Indiana Jones, who gets ahead by the skin of his teeth, jury-rigging one solution after another to just barely beat out his enemies.  Yet when the player tries to behave as Drake, in a way not anticipated by the developer, the result is not a reward, but the most absolute of punishments.  It's an inherent contradiction, and one developer Naughty Dog seems content to leave be, even if it ultimately harms their game in the long run.


There are, of course, plenty more ways in which one could observe or categorize the ways in which player skill is mediated by game mechanics, and plenty more examples out there to put under the microscope.  I hope that in this article I've been able to both shed light on some interesting examples of what I think is a disturbing inclination, and also that I have made the case as to why they ultimately hinder the play experience rather than help it.  Videogames should be all about providing the player with interesting mechanics, tools and scenarios, and, in my opinion, cutting into that via automation is contrary to the very foundations of the medium, and is something that deserves more consideration and scrutiny than it is given presently, by both developers and gamers.

Thanks for reading, and please share any thoughts in the comments!


  1. I agree. I will post more later if I remember as a response.

  2. Hey Eric, I am in love with this post. Excellent response to Bycer's article and methodically laid out on its own merits.

    Second reason for commenting: I'm an official community advocate for Gamasutra's member and expert blogs and would love to get this featured over on the site. Could I persuade you to get in touch with me over email (kris.ligman@gmail) or Twitter (@KrisLigman)? Hope to hear from you.

  3. On the other hand, a measure of automation is often very welcome. As a kid I remember playing Spyro the Dragon and being so very thankful that my fairy companion swooped in to pick up any jewels and collectables I couldn't be bothered to grab for myself. Indeed, automation is what makes videogames work in the first place - automation of rules, automation of maths, automation of that whole process in a board game where you all gather around the instruction leaflet and argue about the wording of rule 17.2. As with the industrial revolution, automation frees us from the necessity of minor tasks to focus on higher-level work. For example, the automation (or filtering) of accuracy and suppression in Brothers in Arms is what requires the player to focus on tactics, position and movement.

    I think this illuminates what may be the real problem in your examples. Most of them focus on games where the skill or task in question would APPEAR to be within the player's remit. The leader of a party of adventurers should get to decide how to handle a situation. Someone playing a game which is all about shooting should get to decide where to shoot. Someone playing the role of an adventure hero should be responsible for the heroics. In these games, automation doesn't seem to effectively change the emphasis of the player's job. Instead it's intruding on what the player assumed to constitute her job. That said, Neverwinter Nights is clearly ATTEMPTING to use automation to change the emphasis - but failing because the new emphasis turns out to feel incomplete and be frustrating.