My Unlikely Roguelike Dislike

On paper, roguelike games seem like they were tailor-made for my personal tastes. They are jam-packed with things I love: procedural generation and the exploration it offers, complex mechanical depth that emerges from systemic/simulationist design, diverse character progression options and build choices that allow you to overcome challenges in many different ways, turn-based combat with tactical depth, and much, much more.

In practice though, they never truly caught on with me, no matter how much I wanted to like them. I can’t quite put my finger on why though. At first I thought it was the permadeath, but I do love permadeath in other games, like for example the Ironman mode in XCOM or the Hardcore mode in Diablo II. I think permadeath is a great fit for roguelikes and makes the design coherent as a whole, so why wouldn’t I like it here as well?

Why then, do I find myself unable to truly get into roguelikes, despite wanting to? That question has honestly puzzled me for quite a while now.

First Encounters

My first encounter with roguelike games happened many, many years ago. It was with NetHack or ADOM or both - somehow these memories have melded together.

With each run, I learned something new and got better at playing the game as my understanding of its mechanics increased. I started dying less often, managed to descend further into the dungeon more reliably, and kept discovering new things along the way. At first, the whole experience was very enjoyable.

But the further I progressed, the more often I ended up having to replay the first few levels. And that endless repetition started to wear on me. Sure, the level’s layout was different each time, and sometimes the level generator would throw in some special feature. But after some time I felt like I already knew the entire list of possible features it was picking from.

Despite knowing these levels so well, survival still wasn’t guaranteed. That early on I might just not have the means yet, to fend off death through starvation or poison. So no matter how much I learned - just getting through those initial levels still always might take multiple attempts. Over time I started to wish I could just skip that part.

But I soldiered on, continued to play “just one more run”, and my persistence eventually paid off. During one particularly amazing run I made it deeper into the dungeon than ever before - by quite some margin. Every enemy I encountered was unknown to me, every challenge new. Of course I had no idea how to deal with any of that - but partially I got lucky, partially I had the resources to mitigate the consequences of bad choices. That sequence of gameplay was great fun and felt exactly like what I had been waiting for.

But eventually I ran out of luck and resources, and met my inevitable demise. That’s how you learn in these games: through failure. On the next run, I’d see these challenges for the second time, and already know more about how to solve them. But that next run never happened.

I knew that just getting back to where I died would take me many, many more runs. Beating that new personal best would take a multiple of that. Only to die again and then start it all over again. Every time I progressed further into the dungeon, the number of times I had to repeat the opening levels just to get back there, increased near exponentially.

So I did what felt like the only sensible thing to do in that situation: I quit.

With no character to feel attached to and no progress to lose, there really was nothing holding me back.

I’m just not a gambling man. And at that point, it felt like an increasingly lopsided gamble to hope that my next run would be anything other than another repeat of dozens of runs I’ve already experienced before. The ratio of repetition required before you could learn something new, grew ever worse. The grind was becoming too much for me.

That memory stuck with me, and in every roguelike I tried afterwards, I would quit sooner. Often as soon as I recognized that it was following the same pattern. I never again invested the same amount of time and effort into another roguelike.

But I always kept trying new games, because of all those other, genuinely tasty ingredients.

Other Experiences

Here are just a few of the more memorable examples.

FTL is one of the very few roguelikes I’ve actually beaten - possibly even the only one - and I’ve beaten it more than once. I love its amazing mechanics and build-diversity. But I never explored all the possibilities I wanted to, because that would have required too many repetitions. I knew the builds I wanted to see would require many tries to get them to work, because of how dependent they were on getting the right loot items, and the right shop offers.

I’ve also played a bunch of roguelites featuring some form of permanent progression. Unlocks and options carried over between runs instead of forcing you to start entirely from scratch. That sort of permanent progression is one of the main distinctions I draw between roguelikes a roguelites.

Neon Chrome allowed you to skip the early levels and start at later checkpoints with an appropriately levelled and equipped character, once you’d unlocked them. That greatly reduced the need to repeat the same old levels over and over again.

Deep Sky Derelicts and Darkest Dungeon both gave you some overarching squad and story progression while retaining permadeath for individual units.

There are plenty more examples, and there’s no point in listing them all. The point is that while I quite enjoyed a few of these games, at least for a while, ultimately not a single one really caught on with me in the long term.

But it’s not the run-based gameplay structure that I dislike. I actually really like that, in games where it’s combined with an overarching campaign. And then I also don’t mind loosing individual runs. Instead I actually like it when the game throws some real consequences at you, and challenges you to overcome these negative outcomes. That’s exactly why I love XCOM’s Ironman mode.

Weirdly enough, though, when playing Fire Emblem - another turn-based tactics series - I always turn permadeath off.

Why do I love permadeath in one tactics game but hate it in another? That might be worth digging deeper into, even if these aren’t roguelikes at all.

Permadeath for Squad-Units

I have finished XCOM on Classic difficulty in Ironman mode. That’s a pretty punishing combination, and you’ll have to deal with unit losses if you play that way. Funnily enough, I had to restart that game many times, because it’s near impossible to continue, if you loose any soldier that’s no longer a rookie, during the first few starting missions.

I really didn’t like that, and felt those initial missions were not well-balanced for Ironman mode. It’s only after you get your first few base and research upgrades, that you can start to maintain at least a few moderately trained backup soldiers as a contingency to deal with losses. But doing so adds a lot to the strategic gameplay and makes the game more interesting.

To me, Ironman mode is the best way to play the game. If Classic Ironman is too hard, lower the difficulty - but don’t turn off Ironman mode.

Without ironman mode, you’re playing an almost entirely different game. If a mission goes south, or a soldier dies, you can restart the mission or load an earlier save and try again. Whenever something goes wrong, you basically just turn back time and undo it. The timeline that remains, the one that becomes your game’s canon, is the one where everything went perfectly, from start to finish.

In the end, no choice ever has any real consequences that way. You can dodge all the bad outcomes by reloading a save and retrying as often as necessary until you get everything right. You don’t even have to think about it - just randomly try one thing, see if it works, and if not reload. You’re not really committing to any choices at all.

But with Ironman mode turned on, you can’t do any of that. You can’t restart a mission or load an earlier save. The game is saved every single turn - and it always overwrites the previous save, so you only have one save reflecting the current state of the game.

As a sidenote: I was very happy about Diablo II doing away with the save games of it’s predecessor and introducing constantly autosaving character saves instead. I don’t understand why that hasn’t caught on more widely. Do people love save-game juggling that much?

With such a constantly autosaving system, you can’t undo your decisions. You have to deal with their consequences instead.

And in XCOM, that’s great and adds an entirely new layer to the gameplay. But in Fire Emblem, things are different. Or at least in those titles of the series I played.

Fire Emblem hands you a limited pool of predefined units who are also characters with their own prewritten stories and personalities. You have enough of them to survive a few losses, but ultimately, the idea is to keep everyone alive so you can experience all their stories.

Permadeath doesn’t add an extra gameplay layer, or lead to any exploration of what consequences that has. I only removes that character and their story arc, while everything else stays unchanged. With permadeath turned on, you’re supposed to just replay the mission as often as needed to keep everyone alive.

And that’s the difference: in XCOM permadeath stops you from replaying missions, in Fire Emblem it makes you replay missions.

That’s why I play Fire Emblem with permadeath turned off, so I can continue on without restarting. There isn’t much exploration of the consequences of someone’s death either way. In XCOM though, the loss of a soldier adds to your personal story, and alters the way you continue through the game. That’s the kind of meaningful loss that I am looking for.

Telling a Story

I’m sure I’ve told this story before, but the first time I finished XCOM on Classic Ironman, it was a soldier called Maria “Hardcore” Garcia, who killed the final boss.

I didn’t give her that name. The game generated it for her, and it couldn’t have picked anything more fitting than “Hardcore”, because that’s exactly who she was. She was a fearless Assault, always first into the fray, leading the kill count throughout the game, and the only surviving member of the original team I started with.

A rookie had to drag out her unconscious body, after the entire team got hit by two missile salvos from two different Sectopods. With her wounded and everyone else dead, I had to play it safe for a while, until I had a new team trained up, and Maria finally returned to active service. That cost me a few points on the doomsday counter, but I made it through and eventually won the game.

That is by no means an amazing story. This is a very truncated version - but even the longer, full story isn’t really all that interesting. As far as literary sophistication goes, this is pretty much bottom tier. But it still feels meaningful, because it was not prewritten, scripted or designed to play out that way. It’s the story that organically emerged from the game mechanics. And it’s uniquely mine.

I love these kinds of emergent stories, and that’s also the reason why I adore games such as Dwarf Fortress and RimWorld.

I still absolutely love Fire Emblem with it’s pre-written stories and memorable characters, but for entirely different reasons. Those stories could just as well be told in books or movies. And they are definitely not mine.

And players want these stories to feel like their own. That’s a big part of what made the branching story system of the early Mass Effect games so popular. It’s also part of the reason why people got so upset, when ME3’s original ending broke that illusion.

Interesting stories, like the one about Maria “Hardcore” Garcia, can’t emerge if I keep rewinding time and reloading a save every time something goes wrong. All those memorable moments get erased that way. You need for bad things to happen, and for the protagonists to overcome negative consequences, in order to create an interesting story.

I can tell Maria’s story over and over again, talking about why it fascinates me. I’m never going to tell the story of Fire Emblem or Mass Effect in the same way, because that would just be spoilers.

My Rogue-Dislike

Roguelikes absolutely can tell stories. In fact, they have all the components needed to create massive potential for emergent storytelling. They’re uniquely well-equipped to do so.

And that’s probably a big part of why I like their procedural generation, mechanical depth, randomness, systemic/simulationist design, diverse character build options, challenges that can be overcome in diverse ways, and more, in the first place.

But looking at the actual stories created by the roguelikes I played, I can’t help but feel disappointed. Hundreds of different variations of: “something happens - protagonist dies” that all don’t feel very interesting to me.

I’m sure there’s a chance these games could eventually create a more interesting story with more ups and downs, if you just repeated enough runs. But I don’t have the stamina to do that. Maybe I just haven’t found the right game yet. That’s why I keep looking.

But from what I’ve seen so far, all of that great potential seem to mostly go to waste. No great comebacks, no turnarounds, no big twists, very little ups and downs. No stories that feel uniquely mine either. Characters are disposable, and so are their stories - and when I’ve seen the basically same story unfold dozens of times already - so has anyone else playing the game. I wouldn’t want to retell them, like I do with Maria’s story.

From what I’ve seen, they keep endlessly retelling narrow variations of the same story: the utterly pointless struggle against inevitable demise.

Over and over and over again.

How uplifting.