Why Portal 2 is Bad and You Should Feel Bad: Tutorial Design & Player Agency.

26 Mar

I find myself time and again in situations at parties where I describe myself as a game designer and people want to talk to me about what I think about Portal 2. When I make a face like someone who has just licked a rotten lemon, I usually get one of two reactions. Either my hitherto friend looks genuinely shocked and even a little tearful that I could do anything other than exclaim in orgasmic ecstasy at the mere thought of the game, or they back slowly away from me, sometimes looking around for a sharp object. One guy actually fell down the stairs doing this. In the spirit of friendly conversation, I’d ask that you hold off reaching for the poker till we’re done here. Yes you. Put it down.


To begin with, while I maintain it is a ‘bad game’, it is not true to say that I dislike Portal 2. When I played it, I was enthralled, engaged, involved at almost every point. I remember one puzzle where I got stalled because I was making some stupid assumptions about the game’s design subconsciously, but otherwise it was a perfect flow experience. The wit was masterful, the atmosphere haunting, the puzzles not so trivial as to lack the piquant joy of finding a solution. Yet, as I sat back from my computer after this apparently wonderful game, all I felt was a deep sense of regret and ‘is that it?’

I was feeling a sense of loss for something wonderful. It is something most of my party friends have experienced far less often and far less poignantly than I and, more tellingly, they don’t know where they should expect to find it. While it may be true that there are different games for different sorts, I felt at the time that there would have perhaps been no better time to introduce these players to the powerful experiences I have had, the ones that made Portal 2 feel like I’d just read a story where the hero finds the seven keys, unlocks the seven gates, travels the seven paths, discovers his kindly father is the dark lord Mortabadassion, struggles to confront him and then, as he reaches the grimdark inner sanctum, opens the mighty gates, sees the demonic creature waiting for him… The end.

I think ‘Blueballed’ might be appropriate terminology.


What the hell does this have to do with tutorial design, I hear you ask? Plenty.

Portal 2 is a tutorial. It is a masterfully written, expertly crafted tutorial. It is a tutorial so winningly constructed that the act of learning the properties and interactions of a variety of given objects and substances in carefully controlled environments optimised for that task (which is a fairly solid definition of the nature and purpose of a tutorial) becomes more engaging than the vast majority of games, be they similarly oriented or not. Yet a tutorial is a means unto an end. A tutorial’s purpose is to reliably generate understanding and mastery of the subject material. To generate the ability to, upon stumbling upon such things in the mess and clutter of everyday life, quickly and deeply understand their significance and what might be achieved with them and to achieve it. At no point in Portal 2 does the game ditch you from the carefully controlled, rarefied environment of the tutorial into the chaos of reality.

When the subjects under consideration exist only in the tutorial, what is the point?

Plenty of games theorists will say “Aha!” at this point: “Learning is autotelic, it is enjoyable in and of itself. It doesn’t need a point.” This is absolutely true, and it is why we all adored Portal 2 while we were playing it, but to say something doesn’t need a point is not the same as saying it would not be better if it had one.

This is where I and my party friends diverge. They do not recognize their trip to mastery of quantum tunnelling devices, friendly crates and moonrocks as such. I do. I yearn for the time when I might be thrown into the wilderness with naught but a portal gun and my newfound ability to outclutch McGuyver to my name. Throw the bathtub at me, the kitchen sink as well. I don’t care. I am ready. I am God on rollerskates today. I am, in fact, the man.

Having discussed this with folks, many get a similar feeling just playing Portal 2. The reason I personally don’t is I’m a lot more aware of the fact the environment is heavily defined, controlled and organised to deliberately lead me to solutions. It’s the age old reason ‘good’ gamers bitch about ‘easy’ games with the hand holding and the hoyvin glaiven. What I wonder is how these other gamers would feel if, thrown into a clearly chaotic, non designed scenario, they devised a solution.

Modern game designers tend to shy away from this because in most cases, when thrown into chaotic, non designed scenarios, players panic and fail. In fact, arguably the strongest trend in mainstream videogame design is to more and more carefully delimit chaos and uniqueness. This has necessitated more and more emphasis on loud noises and bright lights to distract people from realizing that no, they aren’t super soldiers living by their wits and cunning, they’re walking down a hallway pressing brightly lit buttons with ‘press me’ written all over them.

Yet Portal 2 provides an example of something amazing. It’s a tutorial, sure… but… it’s a tutorial. Thanks to Valve’s peerless ability to design controlled situations which both teach players how to do things and then test that they actually understand what they’ve learned, players coming out of Portal 2 are, knowingly or not, the aforementioned God on rollerskates with a portal gun. Put them in a random, chaotic situation and give them a portal gun and I guarantee you only the tiniest fraction will panic and fail. They know what they’ve got, they know what to look for. They know that if this is a puzzle that can be solved, they can solve it. Portal 2 turns the sort of gamers game designers have to design for into the sort of gamers game designers want to design for.

And then it doesn’t take that last, incredibly important step.

It doesn’t give them a chance to feel the apotheosis of agency, the intense, godlike sensation of knowing that you, and you alone, mastered this challenge. Your brain, your guts, your victory. It’s a sensation we don’t get to feel that often. It’s why professional athletes often quite literally beat their own bodies within inches of being permanently crippled. It’s why mathematicians work for decades on some inconsequential logic riddle. That certain knowledge that we had a unique effect upon the world, that we made our mark in a way no one else has yet done. This is the foundation of agency and one of the most powerful gifts a game can give.

In the earnest and admirable desire to seek ways of getting players to the point where they can feel that, we may have lost sight of that fact that, despite the wonderful discovery that the process of getting there is incredibly fun and rewarding in itself, there is a point to be got to.

I imagine I don’t need to talk in too much detail about this to veteran gamers, players from the era where most gameplay was emergent, fluid and clearly generated in such a way that the world you inhabited was volatile and unpredictable. Either that or, while the scenario was fairly rigid, the player has an extremely broad set of tools  to create a plethora of unique solutions- this is perhaps the most iconic quality of Pokemon and other JRPGs. In either case, player agency was clearly established and maintained. Your solutions were your own.

In Portal 1, this was not the case. The game was limited, constructed to very carefully give a limited set of solutions within designed parameters to the player. The genius of Portal was that it subtly acknowledged and played with this concept. You can’t say ‘this is a controlled environment’ much more directly than placing players in minimalist ‘test chambers’ with limited resources outlined by infographics and light trails. So when towards the end you find yourself on the highway to hell and spot a way out, the huge and immediate visual change between environments, pacing, tone and puzzle style tell you powerfully that you’re not in Kansas any more. This is not the case. The last few areas of portal 1 are just as carefully controlled and designed as the rest of the game, but they more strongly convey the illusion of agency, which if successful is just as good as the real thing (for a while, anyway).

Portal 2 does not do this or, at least, not as well. Even in the test chambers, Portal helps convey agency more effectively in a limited sense with its open tiling style. Portal 2 often provides only a single possible action to solve puzzles- there’s only one tile you can portal to, you just have to find it. The subtle distinction here is that any player will subconsciously understand that their solution will be the same as everyone else’s and after a few examples of this their sense of agency will begin to waver and perhaps even shatter. The problem is the game will still be incredibly fun. Without a comparison piece, these players will not be able to criticise the game in such a way the designer can improve their experience.

(This is an incredibly important thing practicing game designers must come to understand- you can’t just playtest a single stream of iterated prototypes. You need to prepare dual concepts which isolate particular design choices. Do controlled experiments, and I don’t mean limiting the amount of game they get to play, I mean controlled experiments. One of my career goals is to establish a system or group through which such experiments can be successfully executed by ensuring unbiased controls, significant evidence pools and so on. For the field of game design to develop professionally it needs to happen.)


So, what can we glean from all this analysis?

  • A tutorial does not need to be designed as ‘a tutorial’. It can be woven into a larger game/experience seamlessly and in most cases these days it is. Yet it needs to be recognized by the designer as a tutorial when it is such, and at the end of the tutorial there needs to be a clear space in which agency is re-established for optimum engagement and pleasure.
  • Constricted, learning focused play is autotelic, it is enjoyable in and of it’s own right. It can form the foundation of a game, but once learning has been established, a good question to ask is ‘how can I give the player an opportunity to demonstrate and revel in their mastery’
  • Modern game design creates clever, capable players within the skillsets that games develop. These players need to be catered to. As videogames increase in scope and use, this demographic will increase in both raw numbers and percentage of overall gamers. This may create a counterswing to the ‘casual revolution’ and mark an increase in demand for more polished equivalents of the mastery games of old. I get the feeling this is already happening.
  • Even within a tutorial (such as Portal) if you can successfully provide the illusion of agency while ensuring that the requisite skills are learned, player experience can generally be improved. Agency creates engagement and personal investment in an experience, which can be used to empower other aspects of play- investment in NPCs and other aesthetic pulls for example.
  • Portal 2 is not actually bad, but it is cowardly. It fails to achieve its potential. It fails to reward gamers with a peak experience. It is great, but it could have been magnificent. You should still feel bad. So there.

7 responses to “Why Portal 2 is Bad and You Should Feel Bad: Tutorial Design & Player Agency.

  1. Caphriel

    March 26, 2013 at 3:56 pm

    I’ll preface this by saying that I never played Portal 2, because it made me violently, horribly motion-sick. I managed to suffer through the first game in very, very short sessions, but Portal 2 was unbearable.

    That being said, did you play the 2-player puzzles in Portal 2? Everyone I know who has played the game considers them much more interesting and challenging. From that perspective, the single-player campaign is played for the story, and functions as a tutorial for the multi-player campaign.

    This is pure speculation on my part, of course, and I’m not trying to defend Portal 2. I’m just curious if you considered it from that angle.

    Cheers on the blog!

    • thereisnosaurus

      March 26, 2013 at 11:46 pm

      I enjoyed the portal 2 co-op most of all and they are more interesting and challenging, but this is mostly due to social interaction as opposed to agency, though I suppose a lot of social interaction is about reinforcing the illusion of agency, so it’s not entirely unrelated. The overall structure is still very much defined, limited environments engineered to show you new capabilities you get from having two portal guns and bodies to stand on switches instead of one.

  2. Nahil

    April 14, 2013 at 6:47 am

    I agree with what you’ve said here, but wouldn’t it apply to just about any puzzle game? Puzzle games present a series of puzzles– controlled challenges with specific solutions. They’re all basically tutorials. What you’re asking for would be games in which the player has to use his/her knowledge to adapt to unpredictable circumstances, right? Not a whole lot of games present unpredictable circumstances. Trial and error and constant, structured teaching are kind of the norm these days. And there’s also a big push for games with interesting, deep stories. Stories are *much* easier to tell in controlled environments.

    • thereisnosaurus

      April 14, 2013 at 7:00 am

      It’s a good point, but not entirely accurate. Puzzle games cover a wide variety of stuff and I think I could fairly say that you can’t apply the same logic as I do to portal to a game like Professor Leyton or Carmen Sandiego. The particular characteristics of portal that make it so powerful in this regard is the composition of the puzzles. They’re made from a set of common challenges with common solutions, and the particular joy in solving them is combining these common elements successfully to get through. The game’s methodology is introducing you to individual attributes of your environment until you have mastered different ways of perceiving and manipulating them in relation to other attributes in the environment. It’s this modular/procedural approach to puzzle generation that gives Portal such potential to expand beyond the limits provided. A game which provides a series of conceptually unconnected puzzles doesn’t do that.

      • Nahil

        April 14, 2013 at 7:43 am

        Ah, good point. I took the fact that Portal 2’s puzzles build on the same core system for granted and forgot about other puzzle games with unconnected puzzles. Still, I’d say puzzle games with interconnected puzzles are quite common (puzzle platformers, for example, are everywhere) and are worthy of this same criticism, along with a lot of non-puzzle games (some of the more linear stealth games come to mind). Regardless, this article has helped me articulate my problems with the Portal games and their kin, and I’d like to thank you for that.

      • thereisnosaurus

        April 14, 2013 at 9:17 am

        indeed! I’m glad I can be of help. As a quick follow up, when analyzing a puzzle game for this property, look for the properties of puzzle elements to vary in little details. for example, a bear trap you need to jump over will almost always be the same challenge- it’s varying the environment around it that creates alterations. On the other hand, a hole to jump over can be scaled up or down as much as you like. A lot of portal’s puzzles feature elements with this kind of property- goo that sprays everywhere, light bridges that can be big or small etc. Of course it has limited elements like turrets as well, but even those are presented in such a way that you can drop twenty into a level at random and POOF new puzzle.

  3. Halefall

    March 16, 2014 at 10:27 pm

    I totally agree with you. That’s why I play the workshop levels and eventually create some. But I suppose you already do that, or at least heard about it.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: