This is also a famous problem in board games, typically referred to as "hidden but trackable information." The classic example is Catan, where player's hands are considered secret, even though every player knows exactly what is gained by each player and what cards each player spends, so a player with a pad of paper or a good memory can remember all of it. Despite massive player push back, the designer of Catan has held firm on this rule. He said that the game just gets too slow and dull without it, as always knowing exactly what every other player has in hand means players can math out the game too precisely. Interestingly, instead of allowing players to get around this rule, it is actually enforced in competitive Catan play (and players are not allowed to use paper to track). In effect, it adds a memory or deduction element to the game. According to top Catan players, logical deduction typically gets them close enough that very few players actually bother with trying to track every card draw and discard. Just an interesting note that despite massive push back from the part of the player base that wants to "optimize" this rule away, it has remained. And that even the competitive community has ended up embracing it.

Amusingly, mods have already done what you've described: there's a mod out there that adds a bunch of new jokers, and one of them is Scouter - and all it does is show you how many points you're going to score. Making it a Joker is also really interesting because it means you're not using that Joker slot for something that actually provides a mechanical benefit, so there's an inherent trade-off to having that information available.

To me, not being able to preview the score is one of the game's greatest strengths, but the ability to calculate it yourself as a last resort is exactly what makes Balatro an all timer. Sure, you can mostly rely on rough estimation without thinking too hard about everything, but there's also an undeniable magic in the adrenaline of having only 1 hand to cover a ton of points and having to desperately pull out your calculator to figure out what combination of jokers, card order or hand type can get you through. Plus, even when I do so, there's been so many times when, in the heat of the moment, I fail to take card interactions, debuffs, effects, or boss abilities into consideration in my calculations and end up failing miserably when I was so confident, or crushing the blind when I thought all hope was lost, and it is extremely fun no matter the outcome. To me, the only thing that would suck the fun out of it is using the dedicated hand calculator, but even then, if someone chooses to play like that and still gets a fun and rewarding experience out of it, then it's still a net positive.

The game does not give you your score, but it gives you enough info that you can reliably estimate pretty quick. The scores in Balatro are so bonkers anyway that you rarely need to calculate them precisely, but with a little experience you can ballpark them in a bracket narrow enough that you don't need to worry. And anyways this is not the kind of game where you have a lot of different ways to play the cards in your hand, so even if a hand is not enough there's probably nothing you can do better with what you have. To me the most fun about Balatro is to set this little formula in your head and tweak it to maximise your score, it's a really fun game for people that like mental mathematics and a score display would take that away.

The thing with Binding of Isaac is, there's literally over 300 different items you can collect. And some of their effects can be pretty niche. I've been using the "item description mod" for as long as I can remember. Makes playing the game so much more fun when you can actually understand what you're getting. Maybe when BoI first came out, not having items described made sense. But now? After like 4 expansions and a ton of free updates? There's too much for any player to possibly remember. And it's just not that fun. ESPECIALLY because so many power-ups can actually be detrimental to your playthrough. I have had a few runs end because I picked up a power-up that really fucked my guy over. Imagine in Balatro if your Joker cards didn't tell you what they did until you used them. It would make the game unplayable.

I think Isaac's "figure it out" approach worked at the start because there were less 200 of them and most of them had a a pretty immediate effect. Nowadays there's over 700 of them, and some of them have really weird effects and/or only activate under specific circumstances. Especially with trinkets, I have over 1000 hours in the game and I really just do not remember what half of them do. Without Item Descriptions, a lot of trinkets just become "thing I pick up and then replace with something that I actually know about without ever seeing their effect activate". I like the method used in The Void Rains Upon Her Heart. Item pick ups are initially kept hidden to the player, but after picking one up, you can unlock their description. Something like that in Isaac would be a nice compromise.

Cool mechanics and fun gameplay aside, can we talk about the art for the joker cards for a minute? Look at the subtle smirk on their faces, the tasteful sheen of their foil. Oh my god, they even follow your cursor when you mouse over them.

When Tunic nerfed an overpowered item combination (increasing its mana cost from 1 to 4), they also added a code that reverts the nerf for 80 seconds, specifically because the speedrun depended on the cheaper mana cost. I can't remember if this code is alluded to in-game.

One possible solution in a game like Binding of Isaac would be to implement an in-game knowledge database you could fill up as you learn mechanics. The game would reflect this in the HUD by only providing you the info that you've already learned.

For Isaac, I always thought the easiest thing to do was to add full item description after you tried said item at least once. That way, I feel like players would be more willing to experiment.

I wonder if McMillen only rented Zelda as a kid? The game came with an instruction booklet that explained all the items. This does bring up something I think a lot of young people playing retro games don't always understand, which is that the instruction booklet was part of the experience back then. Thanks to memory and technical limitations, there were rarely in-game tutorials, so you were expected to reference the book. Of course, many games were also intentionally obtuse beyond that. Every time I got a new game, I was always super excited to read the story section of the booklet. I kinda miss that feeling.

If people knew they would win before gambling, then it would be significantly less popular. The adrenaline rush of not knowing a result until the ball drops is what keeps people playing. Seeing that before you even hit "Go" would definitely make Balatro far less popular than it currently is.

Fun fact: Balatro is actually an old word for jester or joker, which was used frequently in Ancient Rome

On the subject of issac, I think one of the biggest issues with not giving info was the fact you have a fair amount of items with tradeoffs, negative effects, and anti-synergies. So times looking stuff up felt less like "which of these options is optimal" and "will taking this item ruin my run." Ironically, Issac Pills feel like a better implementation of "hidden information" - each run they're randomized, so you can't just wiki it. (and I think once you checked a pill, you'd know what it'd do for the rest of a run)

It's not quite the same vibe, but as a designer myself who likes thinking about combinatorics and the action economy of a game? I absolutely calculated my Balatro turns in advance starting from my very first session. The DIFFERENCE is that I don't pull out a calculator, I just kinda roughly eyeball it and make sure I'm landing where I want it to. It keeps a lot of the thrill of not knowing just HOW big it will be, while still giving me the assurance I'm not making a tactically stupid decision.

I think the specific issue that high level very optimization focused players have in balatro is that there are some strategies that don't rely on beating every blind in as few hands as possible. While that is the default play pattern due to safety, convenience and the gold you get for doing it, sometimes it's better to play hands rather than just finish the blind. This makes your calculation requirements much much harder. If all you need to do is win is as few hands as possible, it's fairly simple to just figure out what your biggest scoring hand will be, even if you don't calculate exactly how big it'll be, and just play it out. If you have an ambiguous situation where you need multiple hands it's a bit trickier to work out whether your biggest scoring hand is worth playing versus maybe playing something that scores less and saves important cards, but doable. If you are trying to squeeze all four or five of your hands out of every round without accidentally busting or spending too few hands it becomes very tedious to calculate your score (and much more important, because you're a lot more likely to make mistakes that lose the run than if you are score maximizing).

Of course Mark makes a video entirely based on the game he's been addicted to throughout developing his game

I love that Balatro doesn't have the score preview for all the reasons you/localthunk have stated, I think an additional advantage is that it allows you to get a different feeling/intuition about each run, instead of having to do a calculation most of the time you can draw on your past experience in the run and just make an approximation of if the hand you're about to play is good enough based on your recent history.

I'm (mostly) on team "leave it the way it is". I've had a handful of Balatro moments where I've cleared the blind by the skin of my teeth, came just shy of the point total, or completely miscalculated and either blew the blind out of the water or fumbled the bag. Trackable information isn't necessarily a design flaw, as it gives players an analog choice: Some will ignore the idea entirely, some will brute force it with calculators or mods, and some like myself will do something in the middle. Face down cards (and I think stone cards as well?) can have their information sussed out by sorting by rank or suit and comparing their position to neighboring cards. Usually I don't go to great lengths to identify anything beyond suit (flush gang where you at), but sometimes you gotta do some heavy tracking and logic puzzling to nail that one straight so you don't die to a stray boss blind. I'll even slow the game down back to 1x just to track face down jokers and reshuffle them back into their proper order. The key thing here is letting players discover and manipulate the game the way they see fit. Adding the score calculator or a hand scoring preview could be an assist feature for players who aren't fans of math, but I don't think it has any place in affecting the core game.

If your game has a "ruin fun" button (using external tools to calculate everything and remove suspense) but you manage to design the game so that only a tiny percentage of players actually bother pressing it, then that's good design right? I get that from a purely objective lens the game has something you could describe as a game design flaw, but it's also something that most people won't encounter. Effort and time are resources too even if they aren't in the code, and this time the game has been balanced in a way where the time/effort to calculate your hand just isn't worth it to most players so in my eyes the problem has been solved.

In magic the gathering, the information in between "hidden" and "public" is "derived information". "Public information" is everything a player is required to share to his opponent about his card on demand, things like amount of cards in hand, life total and others. "Hidden" is every information unkown to one or both player, like what cards are in your hands, what is the order of cards in your library. "Derived" Is everything you can deduce from public information, if a player returns a card from play to their hand, they are in no obligation to reveal that this card is now in their hand because it's a "hidden" zone, but a skilled opponent will remember this "derived" information. I think it's a pretty good naming convention for this "in between" for information technically public but not easily available.

I think Binding of Isaac is kind of a different case. The game has like 1000 items and not knowing what they do is frustrating. Half of them have effects that you would NEVER figure out unless you looked it up, and others will instantly destroy your nice build that you spent an entire run making. The item description mod has in my opinion been the best thing to ever happen to the game.

Something very very similar happened with Rimworld. The game creator, Tynan Silvester, created the game as a story simulator with drama, loss and struggle. But that's not how most people play it. In fact it's most often played as a base building strategy game in which loss is seen as something to be avoided. The game designer decided to push back hard against that kind of playstyle. People made kill boxes for raiders so he nerfed turrets, people responded with traps so he added tunneling enemies which ignored walls so people made the wall into a killbox, so he made enemies which drop down from the sky right into your base and.... you get the idea. The constant arms race has, in my opinion, made the game worse for both types of players so I think it's best to just give the players the freedom to choose how to play is by far the best, possibly with a message explaining the pros and cons like you suggested.

I learned this lesson a couple of years ago when my ex taught me mahjong. Poker and shithead can be calculated, but things like mahjong require you to maintain a risk for much longer, to pick a hand early and commit, to risk changing, to reveal your plans early. The most important thing for a player of mahjong is not, and Balatro too, giving or hiding information, but teaching what information can or should be ignored. The most crucial tip that I ever received was "try to play fast. Trust your gut. Overthinking and over analysing actually reveals you more than it furthers your interests." Turned my results around completely. Balatro works on the same premise. Riskier play is more fun and more rewarding, because you learn what risks you can and can't take.

3:38 "Stern spreadsheet style strategy" is my new favorite vocal warmup

Ngl, i only calculate my score on the boss bind where you are allowed one hand.

Some ASCII roguelikes have solved this problem by hiding what an item can do until you use (or identify) it. In Nethack for example, a "Pink Potion" or "A Scroll that says Abhfbshdjfj" could heal you in one playthru, but explode in your face the next. This leads to fantastic moments of tension where the only way to identify something is to use it and hope it's what you need. There's a finite number of things a potion or scroll can do, so it's still possible to know what an unidentified item does by process of elimination. But it gets you 90% of the way there, while still rewarding players with expert knowledge.

So its not a cursed design problem, it is a deliberate design choice.....

I'd say Balatro nailed that mechanic when it comes to my style of gameplay - I'm not crazy or sweaty enough to calculate exact score, but available info is still useful to calculate approximate result in your head. Then it's just a matter of tens of tokens, that's the uncertainty, the "slot machine" feeling. I love this game for that.

Zelda was that way exactly because they wanted people to reach for information outside of the game. But because the internet was too new, it implied asking friends.

It's a classic cost of convenience scenario. The harder technically available information is to gather, the fewer people will actually seek that information out except for those people that take joy in the process of optimization itself. Strictly speaking, gaining access to information is like getting through a lock, even the best attempts at foiling someone still only slows them down and doesn't prevent a truly determined person from eventually doing so. It just has to be hard enough to deter the average person. Including a score preview in the game would make it a part of the game, but when installing a mod or using an external tool like a browser, you're making a choice to seek resources outside the game itself to breach that barrier. The option remains non-default, clearly defining the desired play experience and putting it on the user to decide whether they want the intended experience or to seek out ways to modify the game to their preferences.

I think another aspect of the message that wasn't mentioned in the video is the seeds in Balatro. If you've played Minecraft, you most likely already know how seeds work there, and that knowledge can be applied to Balatro as well. For those who haven't already seen, if you've played a game of Balatro (regardless of success) you'll be met with an end screen showing your seed for that game. Seeds give the game that "new journey" feeling since each one is a new experience with a different set of Jokers to be bought throughout. There are several reasons why seeds make the game better, and one of those things is allowing the community to find and share unique combinations or possible strategies they found in specific seeds, which can then be recreated by others using those same seeds. This has gone to such an extreme level, that there is a program out now that can search Balatro's entire seed set (if you let it search long enough) for very specific conditions set by the user. For example, the program can filter seeds based on whether they have good potential for a certain hand build, like flushes or high cards. it then starts listing every seed it can find using the fact that each seed has already been predetermined by the game and cannot change between identical playthroughs. So, after all that, even the experience you have in the game can be as random or specific as you want it to be. I personally think that Balatro is fun no matter which way you go about it (obviously the latter will start to get more repetitive) and even speedrunning is an interesting option some people try just to have a new experience. This is what I love about Balatro - there is always something new to try, and you have to be careful about your choices.

I think a big difference with Isaac's pickups is that pickups aren't always free. There are plenty of times in Isaac runs where I picked up a random boss drop and was excited to experiment and figure it out. However, there are just as many times when I encounter an item in a shop and have to decide whether to spend my hard-earned coins on it, which is hard to do if I can't figure out what it does. Same goes with the items you get by sacrificing your health to the devil statue. Because I tend to be conservative with my resources, my choices basically become "look it up on the wiki to see if it's worth it" or "save my money for something whose consequences I know how to predict."

I usually calculate or estimate my score in the first two/three rounds where it's easy to do it by head, without a calculator, because I feel like the beginning of a run is really defining, and I don't want external tools. I can't say I'm surprised some people actually use external tools though, I indeed did that with Isaac as well (only when I felt the need to understand an item)

I always love your videos, man. Genuinely grateful for all the insight and entertainment.

I always find it weird when people refer to Legend of Zelda as a game you had to figure everything out for yourself because it comes with a manual that explains every item and enemies. Certainly there was an element of trying to see an item being used in a practical setting, but the manual basically explains the use of it. I think this mindset comes from people who just ignored the manual. To be fair, I think most people don't want to read them. Very often nowadays it's hard to ignore the manual since it's built right into the game with tutorials. Anyway, I think a solution to this issue in Balatro could just be an unlockable option awarded from winning a certain way. That way new players get the experience of figuring out what things do, and experienced players get their calculator.

Something I've noticed when I play Balatro is that I usually don't try to calculate, but when calculating seems like it will be easy enough to fudge in my head, I'll do it. The information being there makes it easier to spice things up by shifting between the working out the ideal move and the pick n hope for the best playstyles without leaving the game

2:48 quick note The score burns when it is equal or bigger than the amount of score you need to win the round. It's not luck based and doesn't get hotter.

Fascinating, ty. I loved hearing about this in Binding of Isaac. What's most interesting to me about this, is how it shares the issue of losing being unfun, which stealth games have. That's what drives people to optimize the fun out of their games.

i think a score counter is always a bad idea. as someone who follows the “high level” community, balatro is an easy game, especially for optimization. adding a score counter doesn’t help high level players at all, it’s more a crutch for newer players than anything

0:07 egg mentioned

I just got back home from a day at college and the first thing I did was open balatro and right then the notification came. I think the universe is telling me I have a problem with this game.

So while the Isaac comparison I think is the best option that comes to mind, there’s one big issue with it, and that’s that Isaac has INNUMERABLE variables that are both changed by items as well as introduced by them too. Maybe when Isaac first launched and there was just short of 200 items and pickups it was much more feasible that people could pick up a new item and decipher what the pickup did. But there is over 700 now, it is way too much information now. Besides just the process of learning what something is, you may not SEE that item again until you’ve picked up 300 others and you’ve tried to learn what they do so the original item has completely fallen out of the players head. I imagine when the game launched it was a much more feasible thing to expect people to figure out what an item did but as the number of items has gone up, interactions have gone up exponentially as a result, it’s a feature that’s gone too far and I don’t blame new players for using mods to find out what pickups do. As for Balatro, I don’t think it’ll have the same issue. It’s just one feature and piece of information missing. Yes it’s a very big and important piece of information, but you can still ballpark it based on your current run and what other hands played have gotten you up until that point. It’s more that the specific number isn’t known, but you’ll only really be lost on the score of a potential hand if you haven’t been paying attention at all up until that point. But I do think that the option should be there in some capacity. I do fully agree the game should be open to modding, but I personally like the idea akin to using cheats where if you have the score display enabled then you can’t “unlock” anything (like how for a time when cheats were still a thing, usually enabling any cheat codes in a game suspended the ability to unlock achievements). Any new poker cards you buy with the display active will still remain undiscovered, you can’t unlock any new decks, winning a run doesn’t really count as winning a run that’s tracked, etc. It would make it so having the score display enabled would make the next run more of a test run to experiment and get a better understanding of how systems and score multipliers work. That I think will let people play the game how they want with training wheels so they can learn, but will force people who want to progress to play the game how the developer wants them to.

Actually a cheat code does sound like a really good solution for this. Generally only the hardcore people are going to be looking up something like a score preview and the people who just hop on will just play the game as intended.

The Binding of Issac situation actually reminds me a lot about how the Risk of Rain series (1, 2 and Returns) has a similar issue, albeit to a smaller extent, where the effects of each item are described but not the exact numbers (for example, Soldier's Syringe tells you it gives attack speed but doesn't directly specify via. popup that's it's +15% per stack). Then again, Risk of Rain does have tools to see those exact numbers via. mods, wikis, and the logbook outside of active play.

I played through binding of Isaac for hours upon hours before finally caving to adding an item description mod. The reason was simply this: there was too many items for me to remember and it was a hassle to stop mid run to figure out what something did despite me using it already. That is the flaw imo in Isaac. It had become a hindrance. Balatro is different because you will already have a decent idea of your hands capabilities beforehand. You have enough accessible information from your jokers to make the right decisions. It impacts so little people in comparison to something like Issac where the vast majority faces that issue. Change Isaac’s systems to tell you descriptions after discovery still results in a vision Edmund desired, or at least 90% of the way there.

I think the issues are quite different between Balatro's and Isaac. With Isaac you have to remember what 100s of icons mean for your build, some you've seen only a handful of times, some that look a lot alike others. I think a better solution for Isaac would be to slowly unlock rough descriptions / synergies as you pick up the item. This would solve most issues I believe

I love both Balatro and Isaac, but the difference is that Balatro is based on a rule set most people know with way less options. It's basically just figuring out multipliers and you have all the info you need, it's just if you want to get really specific and optimize then you'd need auxiliary resources. Isaac imo extremely frustrating without a item guide or mod as there is no way of knowing what does what outside of experimentation. It can be novel at first but becomes incredibly frustrating quickly, esp with many items causing negative effects.

I believe most players of Balatro could guess whether their hand is good or not based on their previous rounds and current hand. I think the calculator is clearly the antithesis of that and not what most players would use. For Isaac, the more items are added then the more a person requires a wiki to remember what they do and there isn't enough information at a glance to understand if the item is good or not unlike Balatro where the context of your run gives you the power to know what is good or not.

Happy to see "streamed by pretty much everyone" = Northenlion

Oddly, this reminds me of an indie game I played a demo of: "Howl". It's a turn-based tactical game where you can chain multiple actions together during a single round, with one caveat -- for every action you take, every enemy gets to take an action too. In other words, each action constitutes a "turn" (your turn, then the enemy's turn). You do not have to perform the maximum number of actions per round -- it's perfectly possible to execute only one action per round so you can concretely observe the results before you plan your next move(s). However, this is NOT how the game is "meant" to be played, so the game designs a few systems to incentivize planning multiple actions per round: 1 - When you clear a mission, you are scored based on the number of _rounds_ elapsed, not the number of _actions_ taken. 2 - Certain events occur on a "once per round" basis --- notably, werewolves turning innocent villagers into more werewolves only occurs at the end of a round, so the more actions you can chain together, the better you can rescue survivors. 3 - Certain powerful spells/abilities can only be executed as part of a chain of actions taken during the same round; you cannot execute them individually.

I kinda disagree with your conclusion here. I think the solution Balatro arrived at is awesome, because it's easy to "guesstimate" your score, which is a major strategy factor...but it's hard to precisely calculate it. Often not quite "impossible" but still...Thus, for me at least, the score preview provides guidelines, and I usually find it more interesting and efficient to experiment with different hands at various points in order to test or confirm what actually works best for my current deck and joker loadout.

Maybe I'm off the mark, but I think the ambiguous state of the score expands player expression. Making it so the score is technically calculatable makes it so people who prefer not to min-max get to do that without the score looming over them, while those that like to optimize can attempt to do so within a fundamentally chance based game, similar to how some poker players calculate the odds of winning a hand and some don't. I think this is one of the reasons the game has such broad appeal.

In the case of Isaac, there are 2 “solutions” i can think of. First, you can have items be not communicated the FIRST TIME that they’re found. My biggest problem isn’t that items don’t have descriptions (I’m totally with Edmund that finding a new power up and not knowing what it does is very exciting), but the problem is that the game has HUNDREDS of them! If you play RELIGIOUSLY you would remember them anyway and they wouldn’t be a mystery, but for people that only play so often, they’re kind of punished by that fact. Showing descriptions after the first time sort of fixes that problem. You could also have a “memory” mechanic, where the game only remember x number of item descriptions before forgetting them, meaning you’d have a rotation of rediscovering items and being excited by that. I’ll also note here that items/power ups in Isaac can be very hard to understand on their own too, and that pushes player towards descriptions. AND can be required for certain aspects of the game The second “solution” is having item icons only HINT at what the item does. So for example, the cat paw could have 5 possible effects. Which one did you get? Well try it to find out!

excellent video, as always. keep up the good work!

Great video, as always! For binding of Isaac, I think a progressive item description could be nice. So when you first pick up an item, you don't have a description, but the more you use it, the more of its ability description is unlocked for you. It would make players more curious about using said item.

An easy fix and good compromise for The Binding of Isaac seems so obvious to me. It would keep the original design intention while also granting much more accessibility for players - without the need of referring to wikis and mods. Just keep the original design with item descriptions and effects hidden until the player gets and uses(!) them for the first time. And do the same for synergies, interactions and combinations between items. Therefore players feel the excitement of getting and new item and finding secret combinations and synergies, but also gain an in-game wiki of all their previous findings. "Backpack Battles" actually does a great job with an item database and secret combinations exactly like that. Binding of Isaac is just nearly unplayable without any wiki or mod with its 400+ items after all those updates. I'm pretty sure that most players wouldn't use a mod or wiki with that design compromise. And there'll always be a small playerbase of people who want to know everything in advance and get precise information down to the decimal point. Those people will always look up information in a wiki anyway.

God looking at gameplay at normal speed after getting used to play at max speed feels so slow now ! Not a problem per se, just an interesting thing I noticed.

I love these analytical think-pieces you do!

True Solitaire for our time. Never going back. I didn't even think about this thing before, I'm just trying to run these calculations in my head and it's a good mathematical test, even if I'm average, keeps your mind calculating all the way. Then you realize that some common and uncommon cards can be broken if you get them from the start, and even later gotten rare cards won't necessarily run over those if you built them well. I have 4k hours in Isaac, goin my second full run now, balatro is a nice change for that grind, even if I'm only on the easier levels.

11:51 Heat Signature mention! Honestly adore this game and I wish more people talked about it

I think there was a brief golden age of social play for single-player games, essentially before the internet, where a huge part of games was talking about them at school (or wherever, but let’s be honest-it was mostly kids then). I was watching someone recently do a blind play of SMB3 and complaining about one level that “you couldn’t get through without buying Nintendo Power” and it just drove me crazy because there was always *someone* at school who’d figured it out.

I really felt the problem you are describing in a game called Poker Quest. It’s a very hard poker themed, rogue-like, and so knowing the odds was really important to do well. They weren’t officially in the game but their discord had a calculator and they also included a spreadsheet of the odds in the game in the game download. So I found myself constantly querying their calculator on discord to help give me a necessary edge which did bring down the overall experience of the game.

In my opinion, the way Balatro's dev did it, i.e simply hiding the score was a perfectly valid way of doing it. Yeah the game isn't exclusive to "casual" enjoyers but I think it's less about the "right" way to play it and more about the way YOU want to play it. I think a late game unlock could work but I also like the idea of the developer keeping his game the way he wanted it to be to begin with. Mainly I think if the way you're trying to play the game is boring to you, then you're not playing it right. You're not "cheating" but it's like making up rules to guarantee you can win, which is never fun. At least not in my opinion.

Yeah this isn’t a design flaw. I’ve only tried computing a hand once or twice primarily because I was on the last round and wanted to know which hand was the better choice. It was really nice to have that info and I could make the choice how to play. The real flaw with Balatro is the overwhelming importance of early game economy. It makes for a very rote early game in the first couple antes.

Calculating or estimating the score yourself is an expression of a skill. It's part of the tactical gameplay. That feels very different than opaque hidden mechanincs. If each Joker had no description, that would be more like Binding of Isaac.

You can also design your game in a wiki-hostile way to preserve the necessity of exploration and experimentation. For example, say you have a roguelike like Isaac with mysterious items. You could randomize what sprite is shown (each could have an overlapping pool of fitting sprites)

The challenge mode in Balatro is really fun, adding little difficulty spikes without the grind of playing on really high stakes. But there's no unlocks for it, exept more challenges! Coming off of Isaac I would have expected to be able to unlock more jokers from completing these challenges. But having a score preview unlockable from beating a good chunk of these challenges would be really fun. Maybe as a toggle, so that people can only turn it on when they really need to see if their run's about to end or not.

This is super interesting to me. I'm a player that tries to avoid looking up anything on a game at least for the first full playthrough as it completely ruins the fun for me knowing where I "should" be going or what I "should" be using. Obviously I know there's a completely different sect of people that enjoy min/maxing everything in a game, so it's an interesting dilemma.

We're going through this right now with Brailliance. People asked for a "wordle mode" that makes our puzzle game less like its unique self, and more like the popular option. So, we're A/B testing exactly that. As of this writing, the variant without "wordle rules" is winning decisively with day-one retention.

I think an interesting example is Slay The Spire, it also has a lot of information that is attainable but hard to calculate, mainly enemy patterns are predictable and calculating optimal card, relic and potion combinations and order, and accounting for them makes a huge difference. There is a enemy behavior graph mod but generally the community isn't as obsessed with calculators and the like. I think part of the reason for it is that it's abstracted away as health and effects rather than being a single pure number, so it feels different. The way I play both of them is usually with a friend, without calculators but when we feel like we need to we meticulously calculate in our heads rather than use some calculator, especially in Slay The Spire coming up with a hail merry solution in a completely doomed scenario, getting just the card you need, and killing the enemy with exactly enough damage like you calculated is some of the most fun and memorable moments I've had with that game, and is the thing that made me fall in love with deckbuilder roguelikes. This whole situation is also very similar to the use of calculators in minecraft speedrunning. I think people using calculators (especially Balatro specific ones) aren't spoiling the experience for themselves because they are calculating, it's because they are not actually doing it themselves.

I know its unrelated to the actual video, but you talking about cursed reminded me that i really want to see a video of you talking about risk vs. reward in games and how to do it correctly, and different ways to approach the 'risk' part.

11:25 A better example might be Hollow Knight, where the completion % isn't viewable until you've reached the last bench before the final boss.

If he were to add the option to enable score previews I'm curious how he could handle cases where some of your cards are flipped over or where the odds are randomized. Most likely you would have to either leave a "?" or present a range to the user. Very interesting design problem indeed

Similar thing playing Pacific Drive right now. When you scan a new anomaly it adds it to your logbook. But the description isn't "this does that if you touch it", it's an in-universe journal entry or report describing an encounter with the anomaly. It hints at its nature without being explicit because after all, it is a mysterious anomaly. And it is a lot more enjoyable to not know for sure, to hesitantly approach them, or to avoid them in case and then one day when frantically driving bump into one and be like "oh no". But it's a somewhat universal issue with games. RPGs don't tell you "this option will give you a good ending and this one will be bad", but we can look it up, follow walkthroughs, and miss the enjoyment of immersing ourselves. I've started being very conscious of my bad gaming habits, and it's helping me to rediscover joy in a lot of different types of games

Counterpoint to Isaac. As someon who played since the Beta was on Kongregate and yes supported every Isaac kickstarter including the boradgame and Bumbo (yes I am basically a dinosaur). The first Isaac had only a 114 or so items, and none of them that I remember were exceptionally obscure, you would see the effects immediately. About half of these were unlocks for things that, for the most part were fair, achievable challenges. Then came The binding of Isaac rebirth. And I was exceptionally hyped, it had a LOT of cool fun ideas, updated graphics and fixed one of the biggest problems of the first game: Inconsistent synergies: The first game the problem where if you picked up some items in certain order they would multiply differently, the most egregious example was I believe: Sacred heart + Mutant spider, normally both of these items were ABSURDLY good but if you picked them up in the wrong order they would actually lower your damage, I am talking a 8 times difference in dmg with the wrong order. There were a couple of items like this, and this was the main reason people used an external webpage in vanilla, finding all other items, which you eventually sorta learn, is one thing and discovering effects is fun and I ahd a blast figuring out myself what they did, I even accidentally found out both bible interactions. Having your run ruined because you forgot that you needed to pick up the items A - C - B - D instead of A - B - C - D, can kinda suck. Luckily this was fixed in The binding of Isaac rebirth. But it introduced another need for such webapges, three reasons. 1) This time around the item total went up to over 400 and later even close to 450 (if I remember correctly), far more to remember and arguably too much. 2) Some challenges, like anything to do with the lost, a one-hit character were asanine (I think 75% of my 400 hours were spend on this) meaning that getting that HAHA funny item was suddenly ruining that one good run that you REALLY wanted to complete because it gave a super neat item if you did, so better check the wiki to make sure I don't pick up an item that literally accidentally kills me. 3) And lastly, items stopped being simple. Its one thing to have a player figure out that a certain upgrade bends tears to enemies, or pierces obstacles or provides knock back or poison, or makes you move faster. Its another entirely to expect a player to stumble upon the fact that the black candle removes all curses and grants increased changes at the Angel and devil room, and thats an easy on in the spectrum of obscure items. How about the flat penny trinket, that eveyr time you pick up a coin you have a chance to spam a key? Or how picking up the match stick trinket is the only way to remove the tick trinket (I think). It was a very nice design decision for the first game, but with the second game literally giving you multiple items that hurt you (looking at you spark plug) and combining this with a one-hit char (which imo should not be in the game) it drove even me, one of the least wiki oriented person to use the wiki. I will never forget using spark plug on a Lost run where I had both odd mushroom, Brimstone, Ludovico Technique Tiny planet and Mutant spider. I had the mother of all rolls and item combination, the best run I had ever had, and then imagine my surpise when my use item just killed me. I stopped playing Lost and other challenge character without wiki ever since.

The best feeling in Balatro is having a feeling about the performance of your hand and being proven right by the hard math.

Was waiting for a video on Balatro! Thank you GMT!

Whenever this kind of thing comes up, I'm reminded of a very different sort of game - Darkest Dungeon. There's no question that the brutal difficulty, permadeath, and roguelike 'roll with the punches' design was the whole point. You're MEANT to sometimes lose heroes you've invested heavily in building up, meant to sometimes lose entire EXPEDITIONS and have to claw your way back up from a fresh batch of recruits. Meant to learn from it, and strategize around it by always making sure you keep some veterans in reserve who can lead some fresh-faced new heroes into the dungeon if necessary, as well as realizing that investing in the village is the only thing you CAN'T lose, making it all the more important... But some players, like myself, just bounce off that philosophy hard. Preferring to see the heroes as unique characters rather than disposable minions, unwilling to face the immense amount of lost progress that a failed expedition can land you with, or just too impatient to proceed with the amount of paranoia-riddled patience the game demands... and we, rather quickly, figured out how to copy the game's save-directory to another location. Things went sideways? Exit the game, overwrite the save-directory with the copy you made earlier, open it again, and you're back where you were. It's more laborious than a traditional save-load system, sure, but it works - and how much you wanna use it is up to you. Any time you lose a hero? Only if you lose the entire team? Are you still going to play cautiously to avoid the hassle, or taking advantage of this safety-net to proceed more brashly? And, of course, on top of that there's a bunch of 'rebalancing' mods, many of which were very popular. Increasing your party inventory lets you carry more supplies into a dungeon, and more treasure back out. Enabling characters to equip an extra Trinket allows for more powerful combos, and makes them all significantly more potent overall. You could even grab a mod that just let you revive fallen heroes every time you returned from an expedition, though that one had drawbacks. For those with the vampire-focused DLC, you could mod the 'Crimson Thirst' to make it less virulent, and thus less likely to infect your crew or spread among them. All clearly counter to the developer's intentions. But they allowed people like me, who would otherwise simply have abandoned the game after the first major setback, to play and enjoy it for dozens and dozens of hours... ...shame about the sequel. =_=

I deeply revere the fact that Thunk did NOT show the score. A few times I’ve really done some mental math, and either been way off or just barely missed it. It feels like such a core part of the game experience. I think your suggestions for how to include this are really clever, especially as a late-game unlock. Great video!

Popular StS creator Jorbs created an hour long presentation to explain his feelings on Balatro and the matter at hand, which he designated under Delayed Variable Rewards. (Bear in mind I am paraphrasing) In his experience, he isn't a fan of DVRs, and is largely convinced that they shouldn't form the backbone of entertainment - for exactly the same reason as people are against slot machines. It was an interesting video on the topic that I would encourage other people to seek out and form their own opinion on.

At first, Binding of Issac is small. Card, item, power up is quite a few to remember. Then DLCs happen and too many things to remember. I have no choice but to search for help online. because who love losing?

There is a small, but dedicated, modding circle. Balatro is written in Lua and if you use 7zip to open the exe, you can extract the source files, edit them (they're not obfuscated, though the code is a mess), and insert them back in. Note that Window's built in zip application will tell you that the archive is corrupt, but 7zip has been confirmed to work. No one's done a score calculator, as far as I know, but there've been a few other notable additions and tweaks, such as a motion-sickness tweak that removes the swirling background. I did one that put the "won with this joker" sticker (visible in the collection) onto jokers during regular play, so you could track the Completionist achievements without having to constantly reference the collection via the pause menu.

Reading through the comments, I think I might be the only one in the target audience for the way the Binding of Isaac does item discovery. Maybe it’s because I’ve never been “super serious” about getting optimal runs (despite having over 300 hours in the game), but rather enjoy going through and discovering different items. At this point I know what a lot of them do, and so when I do manage to combine a good synergy it feels a lot more like I managed to do it by my own knowledge of the game and repeated trying, rather than by having someone guide me through it. I understand why some people may not like it, but I definitely get the air of mystery of old-fashioned rogue-likes and I really enjoy it.

This reminds me of my experience with Darkest Dungeon, where, at first, I played the game normally without outside information and then, after going through a few particularly frustrating events, looked up what is recommended for each boss/location which drastically altered the feel of the game

Isaac is still pretty different, because the game itself changed. Not having description for a 100 items was still manageable, and each one was clear enough. With rebirth and extensions, they added so many items, some of them arbitrary symbols, that it was just impossible to remember most of them unless you played it regularly Balatro won't get to that point unless it asks specifically for optimal plays

The deck peek is a vital part of the game for me. I don't bother to calculate my hands, but sometimes I need to know what to discard if I want to keep going on that run. It's a clutch mechanic, to know whether I should use my last discards going for a Full House with cards I'm trying to draw but that I may or may not have unwittingly discarded or used them previously.

I’m very happy the points are hidden, and there isn’t an option to show them. I’m definitely someone who would turn that on automatically, and that’s been one of my favourite parts of the game so far! I like the idea of making it a joker, or locking it behind an achievement that you get after playing for a bit, so you can experience playing without it.

I sometimes watch my friend play Isaac and up untill now I've never realised that the UI with all stats and item list and descriptions hovering on the top left weren't integral part of the game but a mod.

I personally always found the lack of item descriptions in _Binding of Isaac_ (and the mechanically similar, though mildly less opaque _Enter the Gungeon_ ) to be a hindrance to the gameplay experience. In my view, it slows a player's sense of Progression in a way that feels unfair. If you sink time into the game, your skill level with the base mechanics - which is central to the Roguelike gameplay loop - will increase faster than you can deduce what every random-drop item does, which sandbags your ability to strategize. You could end up with a combination of items that would help you finally get that sweet sweet Winning Run, but not know that you have items that synergize in that way, making you feel in hindsight (i.e. the next time that RNGesus smiles upon you with that combination of items) like you were cheated out of a Victory.

Risk of Rain 2 tries a similar thing with the lack of item information accessible while in a run, which is frustrating for making builds to get to Mithrix etc. the first mod my whole group downloaded for RoR2 was a UI rework mod that gives item descriptions. not only does it make item decisions easier, it makes the game more beginner friendly. (the next mods we grabbed were item beacons, all-ready check, and bigger group lobbies lol)

I'm pretty sure the exploration of game mechanics vs. community-made knowledge bases problem was mentioned in the GDC talk about cursed problems. If you want to look into this video's topic more in-depth, I recommend checking it out.

Wait, people calculate their score??? I just cross my fingers and get disappointed if I fucked up and didnt play a good enough hand against the needle.

The thing with The Binding of Isaac is that everything revolves around the random items and how they synergize with each other. If an item is unknown, you can't know what it does, and how it'll enhance (or screw up) something else. Balatro, on the other hand, gives you the information, but leaves you to figure out how to best synergize the jokers and your deck. Yes, this isn't a score preview, but at least Balatro leaves you the information to figure things out, rather than TBOI, which left you with hardly any useful clues. Given also a TBOI run is more involved and a player may not see said item again in a long while, the relation of the information is also a bit different.

The thing that changed my mind about this was trying to do the Golden Needle challenge, a challenge where you have only one hand and discarding costs you money. It's an extremely precise challenge where you need to be extremely sure about what you're doing. For one thing, I don't think the underlying mechanics of the game really support that kind of intense challenge in the first place (higher difficulties like Orange and Gold Stake have similar problems). But mainly, when you only have one hand and discards to try and better craft your hand cost resources, not having a score preview really, really, REALLY hurts the experience. It's one of a few things that makes me feel like Balatro is a pretty good game on the lowest difficulty through the first 8 (maybe 10) Antes, but the second you start straying outside those parameters, the cracks start showing REAL fast.

Another element from Binding of Isaac that this video reminded me of are tinted rocks, which drop rewards when blown up. Past the starting floors they become barely distinguishable from other props. While not technically hidden information, they are easy to miss when not paying a lot of attention to finding them, and and for me they become a sort of surprise treat I sometimes notice in the corner of my eye. And yet there's a whole collection of various mods that exist solely to make it impossible to miss the hidden rocks by making them stand out in various ways. It kinda felt like another way of sanding down the intended experience in order to optimise the gameplay in some aspects.

This entire video is bizarre to me because I thought literally the entire point of Balatro was putting together this series of modifiers and multipliers and then the challenge was actually understanding how they worked together and knowing which cards to pick for them. Either everything worked the way you'd figured it would and you win and it's satisfying, or you're surprised by the effects or order that things happen in, good or bad, which is exciting and can lead to fun moments. Either a score preview OR fully precalculating seems like it robs the fun to me, and the idea that the intended experience is apparently that you just guess and act almost at random sounds really strange. It sounds like I got something from the game that neither Mark nor the developer did.

Hey Mark, I'm a fan of your channel and really appreciate how you think and discuss videogame design...but I very much disagree with your take on Balatro. I'm one of those players that calculates scores. Not in all hands mind you, but I do it sistematically enough that I do have a spreadsheet for it. And the thing is: I'd HATE to have a score preview on the game. Because calculating in Balatro is PART of the fun! It's an aspect of exploration, like tracing you own map of an interesting region or piecing together lore from found books. It's not about seeing the resultant score, is about using math, statistics and probability to deepen my understanding of the game, discovering how mechanics work, becoming better at playing, learning and applying those learnings into the next run. I understand not everyone will play like I do, but that's the genius of games as an interactive media: there are multiple ways of having fun, and they don't necessarily need to compete. What you (and LocalThunk) are calling a flaw to me is a wonderful strength: by not having a score preview but showing all needed information to derive it, there's wide open space for player expression and for more people to find their own fun in the game. It's weird for me that you of all people would imply that there's a "wrong way" to have fun while playing games, instead of recognizing that the dev's vision is only one aspect of what a game can be when it meets the real world and it's players. Being open to emergent behaviour and learning from how people interact with your creation, and find their own fun in it, is essential to making games (or really, making anything meant to be used/seen/heard/experienced by other people).

In Isaac, there's actually a much more fitting microcosm of the problem described than with item descriptions, and that's Tainted Cain. For those that are unaware, he has a crafting mechanic where he can convert a combination of pickups into items based on fixed recipes. When Repentance first came out, the optimal playstyle for him was to leave every pickup on the ground while making a note of them externally, then going to a recipe browser online and *meticulously* trying to figure out the most profitable combination of pickups-to-items before moving on to the next floor. It was not uncommon to have T. Cain runs go for upwards of several hours just squeezing out every little bit of value you could flipping between recipes and pickups (not to mention how easy it was to break the game when you can essentially pick the item you want) It was such a problem that a patch came out that randomized the recipes by seed so as to discourage this rigorous playstyle, not unlike how Balatro chooses to purposefully hide the hand score to allow the player to discover it naturally!

I'd love to check this out but only recently finally got Slay the Spire (and tons of other games). Glad to see you cover it here though.