This may come as a surprise to some based on my recent opinions of Nintendo, but growing up, Metroid was one of my favourite franchises. I have a soft spot for sci-fi, but what really made it special was the game’s structure. Up until Metroid, every game I played was linear, and sent me directly to the next level after finishing the one I was on. By contrast, Metroid oriented its levels into a complex maze. This ultimately meant players would occasionally need to double back through previously explored areas to move forward in the game. Conceptually this sounds terribly annoying, but the lack of linearity gave the game world a real sense of place that other titles lacked.

While Metroid was my introduction to non-linearity, truthfully, it wasn’t the best example of it. In the majority of cases, Metroid games are designed with an intended order of events. That is to say, while the levels weren’t oriented sequentially, they were meant to be played in one specific order. Unfortunately, it can be extremely aggravating to search a seemingly endless labyrinth for hours when you just want to find the start of the next level. While I don’t mind this feeling, I’m firmly in the minority. It’s understandable that most players would be frustrated if they go too long without making any tangible progress before ultimately quitting from boredom.

What if there was a way around this though? The thing that often makes Metroid games annoying is not knowing where to go next. If there was a way to increase the odds of players finding their way, then this irritation could be reduced, or mitigated entirely. Well, recently I played a handful of games that do just that, so I wanted to look at how these games avoid the pitfalls of non-linear structure through use of smart design.

At its core, I think the problem of getting lost is one of probability. If you have 5 potential paths, but only one is correct, then you’d have a 20% chance of guessing right with the likelihood rising as you eliminate incorrect paths. Thus, a simple way to help the player would be to increase their odds by making several of the paths correct. For example, let’s look at the infamous trek down to Blighttown in Dark Souls, for which there are three potential routes. Players can either slog through the Depths, enter the Valley of Drakes from Darkroot Garden, or take a shortcut from New Londo Ruins provided they started the game with the Master Key. While most first timers are likely to go with option one, having alternatives helps to increase the odds of a truly lost player finding their way.

The strategy of upping the player’s odds by increasing the volume of valid pathways can also be observed in Hollow Knight. This game takes a lot of cues from older Metroid games, and features the same complex labyrinthine design that has become synonymous with games of its ilk. However, Hollow Knight’s various areas are much more connected to one another. While some fringe locations only have a single gateway, the majority have three or more. Said areas with multiple entrances are also where all of the game’s mission critical events take place, while fringe areas are relegated to secrets and side content. This greatly increases the odds of players stumbling into any given area, by providing them with more opportunities to organically find their way.

While simply increasing the number of pathways into a given location is great, I’ve always thought it was smarter to embrace non-linear progression in this style of game. As I stated before, most Metroid titles have an intended order that the player is meant to collect each of the game’s different power-ups. What if this restriction was removed, and instead, there were several branching points where players could collect a subset of power-ups in any order? Well then your game’s progression would mirror the non-linear structure of the world, making it a much better fit overall.

To illustrate non-linear progression, I’m going to reuse Dark Souls as an example because there are two separate points where it commits to this idea. The first follows the tutorial, when players are transported to the land of Lordran. Upon arriving, they’re informed of 2 bells that must be rung. Most beginner scrungaloids won’t know it, but they’re being presented with a choice here. They can either choose to adventure into the Undead Burg, and scale the parish to ring the bell that sits atop the church’s spire, or slither through Blighttown, and find the bell hidden within the bowels of the poisonous swamp. Both choices have their merits, but the player ultimately gets to decide in which direction their adventure progresses.

A similar path split happens later, after players finish the infamous Snorlax and Pikachu fight. They’ll now need to collect the 4 Lord Souls, which necessitates exploring the farthest reaches of Lordran. As with ringing the 2 bells, the 4 Lord Souls can be collected in any order. While these zones are somewhat hidden, having 4 correct places to explore next greatly increases the odds of a hopelessly lost player stumbling into an area where they can make some progress. As with increasing the number of paths into any area, by providing multiple points of interest, Dark Souls ensures that the majority of players will happen upon something in relatively short order.

This sort of branching progression may seem complex, but it is also possible on an indie budget. Ender Lilies is one such indie game that employs the exact same method as Dark Souls. After the tutorial, players hit a crossroad where they can choose to trudge through the ruins of a town, or trawl through the waterlogged mage’s library. Either area is a valid choice, as you’ll need the new abilities found in both to progress through to the game’s next zone: the cathedral. Afterward, Ender Lilies opens up once more, providing 2 new branching paths. Structuring the game this way also helps to mitigate potential pacing problems, while keeping things focused on Ender Lilies best bits: player driven exploration into new locales, and experimenting with new abilities.

It’d be remiss of me to talk about player driven exploration and non-linear progression without mentioning my favourite example of both: Hollow Knight. Unlike the aforementioned titles, Hollow Knight keeps players on a fairly narrow path for a much longer period of time. The first 3 power-ups you’ll collect – the dash, wall climbing, and a projectile attack – are always distributed in the same order. However, after collecting these, players will suddenly have a variety of different options for where they go next. There are six different areas that can now be accessed, some of which provide new abilities, while others contain mandatory events that help to unlock the final boss. As with our previous examples, there is a much greater chance of players making meaningful progress when there are so many viable options for where to go next.

As someone who enjoys non-linear games, I’d be lying if I said I didn’t see the problem inherent with their design. By shuffling up the game’s levels, designers run the risk of players becoming lost and not knowing how to progress forward. No one likes that. However, I think I’ve identified two ways that smartly designed games avoid this common pitfall. Adding extra entryways, and providing players with more freedom while exploring greatly increases the odds that they’ll find a path forward. Players should spend the majority of their time excitedly choosing where to explore, instead of asking “where do I go next?” in frustration. Hopefully we’ll see more games adopt this style of design in support of nonlinearity going forward.