I am making a 2D platformer where the player has to reach the end of the level as fast as possible. I would like the player to be able to find new ways to reach the end of the level faster.

Making multiple path is not enough, in my opinion, because the shorter one will be easily findable and the others will become useless. I thought that maybe I should add abilities to the player to increase the number of possibilites.

I would like the player to be able to find strategies I have not thought about. For example players who are doing speedruns find new path and strategies to beat a game faster. I would like to help the research of new strategies. I would like to make new strategies possible even if I have not specificaly thought about them.

My question is, how can I make it possible for the player to make up new strategies ?

  • 2
    \$\begingroup\$ The concept you want is called emergence. \$\endgroup\$
    – rlms
    Commented Jun 30, 2014 at 19:50

3 Answers 3


This is almost paradoxical, because if you focus too much on "planning" for alternative routes you ultimately tend towards constricting the number of available routes to exactly those you've planned for, achieving a pale semblance of your goal instead.

There is some level of specific design work you may need to do to facilitate this (for example, rooms in Rogue Legacy must have explicit extra paths added for heroes with dwarfism traits, levels in Fallout 3 often have lock-picking and computer-hacking paths to achieve some goal); you just don't want to take this so far as to end up always having "the path for this ability," and "the path for that ability."

The abstract techniques you should employ to allow for multiple paths and strategies in a game align closely with those you'd probably want to employ to foster replayability in general. This includes things like designing for composability:

  • give a player (eventual) access to many abilities, each of which do a relatively small thing on their own, and possibly have a small downside

  • approach the design of your gameplay space in terms of verbs and nouns: simple things that can be done, and simple things that can be operated on

  • don't artificially restrict those combinations, and let your playtesters explore, and listen to their feedback when they say "hey, I tried to use X to do Y here and I thought it would work, but it didn't."

  • if possible, design levels and areas to be true to their real-life analogues; players can relate to real spaces and their own internal fantasies of being (for example) a crazy parkour acrobat traversing an actual urban landscape, or a daring Lara Croft or Indiana Jones-like spelunker. Forming gameplay spaces around familiar structures allows players to more easily project those fantasies into the game and see options for exploration.

For example, grenade-hopping and the like is a direct consequence of two simple factors:

  • explosions push things around in relatively physically-accurate way
  • the player is tougher than most other things

These combine to make grenade-hopping a viable option, even if it wasn't designed into the game (by contrast, some sequence-breaking-enabling things, like the foam gun in Shadow Complex, were specifically designed into games for sequence breakers).

This sort of thing is a far more organic then premeditated process. Keep building simple composable gameplay elements and then trying them together yourself or through playtesters. See what they come up with; that may inspire you to add other abilities to further compliment a strategy you see emerging ("emergent gameplay" was the 90s buzzword for this sort of thing, incidentally). Or it may necessitate you adding downsides to abilities to complicate the exploiting of some mechanic you feel trivializes the game too much, or something.


The kind of gameplay experience you are talking of is a result of glitches and chaos in an environment that is too complex to allow for complete testing of every possibility. The more complex the world and set of abilities you create, the more likely it is there are paths you haven't thought of. Even if there is an infinite number of ways, there are always some ways that are superior to others. A game's speedrun can only be improved up to a point and from there it's very hard to make a tiny improvement.

Consider leaving in bugs like an enemy attack that knocks the player character far away allowing you to traverse over chasms you wouldn't be able to without getting hit. Another example would be that getting hit leaves you invulnerable for a second or two and you are then able to go through things like spikes that would normally instakill you.

The way to create gameplay that is unpredictable is by leaving in buggy behavior, giving a set of powers that you haven't fully tested and nerfed to limit the player and letting the player / monster /environment interaction result in cool things like very strong knock back or temporary invulnerability. If all that fails you can leave in bugs like players being able to phase through walls at certain speeds.


One way is the Quake 3 way. Based on a bunch of what were originally bugs in the code, you get all sorts of weird interactions.

  • Plasma climbing.
  • Grenade jumping.
  • Rocket jumping.
  • Strafe jumping.
  • Grenade/rocket jump (youtube video of me doing one of these: https://www.youtube.com/watch?v=i6rqumD-9Hk)
  • Overbounce (never learned to do this one).

If your player has access to a load of different ways to move, they will learn to combine them in weird ways to get through the level in new ways. Check out stuff like Quake Done Quick to see this in action.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .