Submitted this because Balatro’s one of my favourite games of all time; this an overview of what it’s like to develop and publish an indie game (on the less technical side), and the timeframes involved.
Thanks for sharing it! I wouldn’t have known LocalThunk had a blog otherwise. He seems like a really grounded and thoughtful person, and there’s a ton of behind the scenes stuff in this blog post.
The success of Balatro is absolutely incredible, and very well deserved. It’s a pity that it seemed to have such a physical toll on the developer.
Balatro’s probably the most accessible roguelike deckbuilder I have ever played, and it really is a ton of fun. In a lot of ways it feels like ‘outsider art’ to the genre - the blog post mentioned that the developer didn’t really play any similar games during development and it really shows. Balatro really is a breath of fresh air, and brings a lot of innovation to the table.
I genuinely wonder what do you consider to be Balatro innovations? Joker mechanics is not different from “all orcs get +1” effect cards I think. (I only have unlocked all decks)
The biggest innovation is the core game mechanic of playing poker hands and receiving a score based on a multiplier.
Most roguelike deck builders have you starting off with a fairly small, fairly weak deck that you increase in power by adding cards to. Balatro innovates here by starting the player off with a standard 52-card deck!
Another innovation is the way scaling works over time - the most powerful ways to scale are all “meta” - scaling that is preserved between different rounds.
The different types of rewards, such as the Tarot cards are also very interesting.
Not sure it counts as innovative, but Balatro is a game for the sake of it. It doesn’t have a story, doesn’t have a message, doesn’t have lore. It’s just a game.
An extremely well designed and fun game that you can play whenever you want for how long you want, which some mechanics you likely already know (poker hands) and some you don’t (jokers). It nailed the infinite replayability with Endless Mode without making it (too) frustrating, it’s just bonus challenge.
Besides that, it has all the ingredients of a rogue like, nothing really innovative here in my opinion, and that’s ok.
For me, a number of the “innovations” are simple details that make the game more compelling and fun, and the fact that these design choices aren’t oriented around getting the poor victim to open one more loot crate while draining their wallet.
Like the way the multiplier graphs burst into flames, the sound effects and the like.
It’s interesting to me that they never played Slay the Spire until 2023! It’s my favorite game of all time and I consider it largely foundational to the genre of modern deck-building roguelikes.
Thank goodness I avoided playing it until now because I surely would have just copied their incredible design (intentionally or subconsciously).
Really interesting perspective and way to approach game development! Ignore the herd and do your own thing, then go back and explore the genre. Kinda the inverse of what I expected.
That’s what leapt out to me too. If he played other games from the start, Balatro wouldn’t be nearly so unique. If he hadn’t played other games at the end, it wouldn’t have had the stake system, which makes the game so much more replayable.
Research makes sense for exploring the design space for solving a specific design problem (i.e. “I know I want a mechanic/dynamic for X. What mechanics have other designers used for this?”), which should really only happen after you’ve gotten through the first steps of, “What experience am I trying to make? What mechanics would I need to generate that experience?” It’s how I prefer to do my game design as well, and I’m much happier with the systems I’ve made as a result. You get more novel ideas that way.
Doing things in reverse is how you get a preponderance of clones (Hollow Knight, for instance, spawning a zillion Metroidvanias that share certain key traits with HK) where people want to make a game that’s Just Like The Thing I Like. They take a package of mechanics based on something they like and flex one or two of them. It makes something slightly new, but not wholly new, and ultimately it feels very same-y to the source material.
[…] because making games is my hobby, releasing them and making money from them is not, so naively exploring roguelike design (and especially deckbuilder design, since I had never played one before) was part of the fun for me. I wanted to make mistakes, I wanted to reinvent the wheel, I didn’t want to borrow tried-and-true designs from existing games. That likely would have resulted in a more tight game but it would have defeated the purpose of what I love about making games.
This really resonated with me. I have a lot of project ideas; some of them fall into the category of ‘not fun but might make some money’ and others fall into ‘fun but definitely will not make money’. These ideas constantly battle within me for my attention and my contradicting desires of having fun while programming and wanting to make money to become a solo developer.
“How can you not be romantic about [indie game development]?”
Really heartwarming read. Glad the author got to experience this, because it brought a smile to my face for the joy of the craft:
I start properly playing the game myself about a week before launch - and it’s actually fun. I have a pretty emotional moment where I feel like I did the thing I set out to do. Finally. I made the fun game I wanted to make.
It’s the kind of game you can recommend without any qualms because the price is very reasonable and there are no inducements to get the player to spend more and more money which is so common in many other modern games.
I’m not trying to litigate the terminology, but calling Balatro — which is brilliant — a roguelike feels just so weird from the POV of someone who hasn’t really followed the discourse. In my mind roguelike meant a game where a @ symbol walked around a randomly generated dungeon slaying one-letter monsters. Now suddenly a poker solitaire is a roguelike?
That second point about unlocking stuff is pretty common nowadays but not a part of e.g. Rogue or Nethack, so sounds strange to me to use as a criterion.
I would have replaced that point with having randomly generated maps to some degree (which Balatro has by randomising the shops and blinds).
Words change. Rogue would maybe not be considered a rogue like with the current definition, and that’s ok. TBH I don’t believe a lot of today’s rogue like players know much about Rogue. I think Rogue would qualify as a dungeon crawler, which is orthogonal to rogue like: rogue likes can be dungeon crawlers, and dungeon crawlers can be rogue likes.
I know the Wikipedia definition is all about ascii characters and dungeons, but that’s just not how the word is actually used today. There’s a blurb about roguelike-likes but no one say that.
It doesn’t matter what Wikipedia or dictionaries say. How words are used is what they mean. Wikipedia and dictionaries are just a reflection of real life usage, not the other way around. People believing it’s the other way around are the reason why Wikipedia is wrong in this case.
You can keep the definition of roguelike to mean a dungeon crawler with ascii characters and no item unlocks, but no one will understand you.
In fact I challenge you to find real world usage of “roguelike-like” on the internet except on Wikipedia, and on forums where people say the “roguelike-like” term is stupid.
Currents that came to the surface, perhaps, around 2011? There’s a review of The Binding of Isaac from 2011 where multiple paragraphs, in multiple places, discuss whether the journos should go along with people calling it roguelike. Whether roguelike is the right word for a game that is and isn’t.
Absolutely loved reading this. Balatro is really super fun and genuinely viral: I found out about it from a friend who got it from some obscure spanish website. It’s incredibly well made, and the story (this story) how it got made is an absolute dream for many.
The game source code is in Lua and totally readable. It’s one of the files in the Steam release and is easy to extract. IIRC it even includes the author’s comments.
If you look for next similar game - check Drop Duchy - demo is available on Steam - I loved Balatro and I really look forward to Drop Duchy game - it will be released in 05/05 this year.
Similar in a way that a new concept will be exploited in all new possible ways - Drop Duchy is not a card game.
Thanks for sharing this! Balatro is one of my favorite indie games recently, and the fact that it’s written in LOVE makes it all the more amazing given the incredibly high quality of the game.
Balatro is extremely well made but in the end for me there wasn’t much more there than try to stack your deck and see the number go up in a bunch of relatively predictable variants. Fun to play through once or twice, but nothing to be gained from continuing to do it other than time wasted.
Submitted this because Balatro’s one of my favourite games of all time; this an overview of what it’s like to develop and publish an indie game (on the less technical side), and the timeframes involved.
Thanks for sharing it! I wouldn’t have known LocalThunk had a blog otherwise. He seems like a really grounded and thoughtful person, and there’s a ton of behind the scenes stuff in this blog post.
The success of Balatro is absolutely incredible, and very well deserved. It’s a pity that it seemed to have such a physical toll on the developer.
Balatro’s probably the most accessible roguelike deckbuilder I have ever played, and it really is a ton of fun. In a lot of ways it feels like ‘outsider art’ to the genre - the blog post mentioned that the developer didn’t really play any similar games during development and it really shows. Balatro really is a breath of fresh air, and brings a lot of innovation to the table.
I genuinely wonder what do you consider to be Balatro innovations? Joker mechanics is not different from “all orcs get +1” effect cards I think. (I only have unlocked all decks)
The biggest innovation is the core game mechanic of playing poker hands and receiving a score based on a multiplier.
Most roguelike deck builders have you starting off with a fairly small, fairly weak deck that you increase in power by adding cards to. Balatro innovates here by starting the player off with a standard 52-card deck!
Another innovation is the way scaling works over time - the most powerful ways to scale are all “meta” - scaling that is preserved between different rounds.
The different types of rewards, such as the Tarot cards are also very interesting.
Not sure it counts as innovative, but Balatro is a game for the sake of it. It doesn’t have a story, doesn’t have a message, doesn’t have lore. It’s just a game.
An extremely well designed and fun game that you can play whenever you want for how long you want, which some mechanics you likely already know (poker hands) and some you don’t (jokers). It nailed the infinite replayability with Endless Mode without making it (too) frustrating, it’s just bonus challenge.
Besides that, it has all the ingredients of a rogue like, nothing really innovative here in my opinion, and that’s ok.
For me, a number of the “innovations” are simple details that make the game more compelling and fun, and the fact that these design choices aren’t oriented around getting the poor victim to open one more loot crate while draining their wallet.
Like the way the multiplier graphs burst into flames, the sound effects and the like.
It’s interesting to me that they never played Slay the Spire until 2023! It’s my favorite game of all time and I consider it largely foundational to the genre of modern deck-building roguelikes.
Really interesting perspective and way to approach game development! Ignore the herd and do your own thing, then go back and explore the genre. Kinda the inverse of what I expected.
That’s what leapt out to me too. If he played other games from the start, Balatro wouldn’t be nearly so unique. If he hadn’t played other games at the end, it wouldn’t have had the stake system, which makes the game so much more replayable.
Research makes sense for exploring the design space for solving a specific design problem (i.e. “I know I want a mechanic/dynamic for X. What mechanics have other designers used for this?”), which should really only happen after you’ve gotten through the first steps of, “What experience am I trying to make? What mechanics would I need to generate that experience?” It’s how I prefer to do my game design as well, and I’m much happier with the systems I’ve made as a result. You get more novel ideas that way.
Doing things in reverse is how you get a preponderance of clones (Hollow Knight, for instance, spawning a zillion Metroidvanias that share certain key traits with HK) where people want to make a game that’s Just Like The Thing I Like. They take a package of mechanics based on something they like and flex one or two of them. It makes something slightly new, but not wholly new, and ultimately it feels very same-y to the source material.
This really resonated with me. I have a lot of project ideas; some of them fall into the category of ‘not fun but might make some money’ and others fall into ‘fun but definitely will not make money’. These ideas constantly battle within me for my attention and my contradicting desires of having fun while programming and wanting to make money to become a solo developer.
“How can you not be romantic about [indie game development]?”
Really heartwarming read. Glad the author got to experience this, because it brought a smile to my face for the joy of the craft:
What an incredible journey — I immediately bought the game for iOS and spent the rest of the evening playing.
It’s the kind of game you can recommend without any qualms because the price is very reasonable and there are no inducements to get the player to spend more and more money which is so common in many other modern games.
I’m not trying to litigate the terminology, but calling Balatro — which is brilliant — a roguelike feels just so weird from the POV of someone who hasn’t really followed the discourse. In my mind roguelike meant a game where a
@symbol walked around a randomly generated dungeon slaying one-letter monsters. Now suddenly a poker solitaire is a roguelike?To many, a rogue like is a game:
where you die at some point, and restart from the beginning
where you unlock stuff that you can get in new runs
with infinite replayability
A rogue like doesn’t have to be in a dungeon, at least in the current definition of the genre.
That second point about unlocking stuff is pretty common nowadays but not a part of e.g. Rogue or Nethack, so sounds strange to me to use as a criterion.
I would have replaced that point with having randomly generated maps to some degree (which Balatro has by randomising the shops and blinds).
Words change. Rogue would maybe not be considered a rogue like with the current definition, and that’s ok. TBH I don’t believe a lot of today’s rogue like players know much about Rogue. I think Rogue would qualify as a dungeon crawler, which is orthogonal to rogue like: rogue likes can be dungeon crawlers, and dungeon crawlers can be rogue likes.
I know the Wikipedia definition is all about ascii characters and dungeons, but that’s just not how the word is actually used today. There’s a blurb about roguelike-likes but no one say that.
[Comment removed by author]
It doesn’t matter what Wikipedia or dictionaries say. How words are used is what they mean. Wikipedia and dictionaries are just a reflection of real life usage, not the other way around. People believing it’s the other way around are the reason why Wikipedia is wrong in this case.
You can keep the definition of roguelike to mean a dungeon crawler with ascii characters and no item unlocks, but no one will understand you.
In fact I challenge you to find real world usage of “roguelike-like” on the internet except on Wikipedia, and on forums where people say the “roguelike-like” term is stupid.
[Comment removed by author]
If you’d like to get a feel for the discursive currents tugging at a word you thought you knew, Alice O’Connor at Rock Paper Shotgun had cause in 2016 to ponder the diffused and drifting meaning of roguelike, roguelikelike, and roguelite.
Currents that came to the surface, perhaps, around 2011? There’s a review of The Binding of Isaac from 2011 where multiple paragraphs, in multiple places, discuss whether the journos should go along with people calling it roguelike. Whether roguelike is the right word for a game that is and isn’t.
Excellently written! The seamless weaving of personal life updates with game updates made for a compelling read.
For another indie roguelike game that took off in a similar way, see also the SNKRX post-release log.
Absolutely loved reading this. Balatro is really super fun and genuinely viral: I found out about it from a friend who got it from some obscure spanish website. It’s incredibly well made, and the story (this story) how it got made is an absolute dream for many.
There was a great interview with LocalThunk on the Eggplant podcast last year: https://eggplant.show/136-going-in-blind-with-balatro
What a success this game is, a real inspiration.
The game source code is in Lua and totally readable. It’s one of the files in the Steam release and is easy to extract. IIRC it even includes the author’s comments.
If you look for next similar game - check Drop Duchy - demo is available on Steam - I loved Balatro and I really look forward to Drop Duchy game - it will be released in 05/05 this year.
Similar in a way that a new concept will be exploited in all new possible ways - Drop Duchy is not a card game.
Thanks for sharing this! Balatro is one of my favorite indie games recently, and the fact that it’s written in LOVE makes it all the more amazing given the incredibly high quality of the game.
Balatro is extremely well made but in the end for me there wasn’t much more there than try to stack your deck and see the number go up in a bunch of relatively predictable variants. Fun to play through once or twice, but nothing to be gained from continuing to do it other than time wasted.