Bad game design is the result of a conflict between a design and the goals of the game, but “bad” is too vague a term for it. The terms good and bad design are thrown around too often, usually from anxious junior designers eager to improve upon their favorite games. I personally ban the use of good and bad from formal discussions around game design. The terms ‘good’ and ‘bad’ make a fatal presumption – that the speaker, the listener and the player have the same goals, values and priorities. This simply isn’t true – replace the words good or bad with real adjectives: a design is engaging, boring, cost-effective, well-tested, intuitive, or frustrating, not good or bad.
A frustrating feature in a cozy game is likely to cause the entire playerbase to quit. In a horror game, it’s often an essential element to make it a memorable experience. Game design is a contextual art and a technical discipline. You need to keep your critical thinking skills engaged the whole way. That said, we’ll look at games which players say have poor designs, examples of frustrating or unintuitive designs, and when using a frustrating or unintuitive design is effective. Read on to learn what most players are saying when they say a design is bad.
What defines bad video game design?
What defines “bad” video game design is a design that doesn’t match the emotions, themes, and core pillars of a game. An alignment between the goals and reality results in (hopefully) a fun experience, and a misalignment doesn’t. The challenge in defining whether a design is effective comes from the fact that fun is subjective, so defining a game as good or bad based on pillars or themes alone is impossible without considering the audience. A more helpful way to define “bad” game design is that it’s the result of either ignoring the target audience or slavishly following the target audience’s criticism.
One player’s frustrating, unengaging, or unfair experience is another player’s favorite thing to do. An activity as mundane as organizing one’s sock drawer is fun for some people, but certainly not for me. In the same way, a design appropriate for one game isn’t appropriate for another. One game evokes sadness, as in Heavy Rain, another evokes fear, as in Outlast, and another evokes a sense of complete simulation and realism, like Dwarf Fortress. The full simulation of Dwarf Fortress transplanted into a horror game is a complete mismatch: the old Dwarf Fortress interface certainly inspires fear, but doesn’t give the sense of disempowerment that’s essential to a horror experience.
A game without a target audience is going to fail, since there’s no way to know whether the gameplay is going to be too easy, too hard, or just not the right fit. The fact that fun is subjective means a large number of people are going to find a game unfun no matter what, and creating something completely unheard of leaves it up to chance whether someone who buys the game is going to love it or hate it. Conker’s Bad Fur Day had disappointing sales on release, and a part of the failure lies in the fact the mature-themed title shipped from a company known for making child-friendly games. The game even matches the cartoonish aesthetic of other games by Rare and in the Nintendo 64 catalogue.
An ineffective design is guaranteed if the designer doesn’t test the game and see how the target audience reacts to it. The challenge to playtesting effectively is parsing player feedback and deciding how, or whether, to address it. Identifying bad game design, in playtesting or after release, often comes with the following vague language.
- The game is unfair or the game is hard
- The game is unoriginal
- The game is unengaging / boring
- The game isn’t like another game
- This tool feels overpowered
Listening to all feedback and designing by committee is a recipe for a “bad” design too. A playtester with little experience playing games isn’t going to have as valuable feedback if the game has an intended hardcore audience, for example. Jaime Griesemer said in his GDC talk on balancing that the question he asks after feedback is: “Is the game in the player’s head the same as the game I’m making?” The feedback has less weight if the answer to the question is no.
A design that matches the target experience isn’t always the best solution either, if the design is beyond the studio’s means. Each design is a tradeoff between the value it adds and the manpower, resources, and time that go into implementing it. Red Dead Redemption 2’s plodding pace works for the simulated world Rockstar was aiming for, but most studios don’t have the resources to hire horse designers or create animations for looting every cupboard and skinning every animal. For any other studio creating a gritty, realistic game, this level of detail is still bad design because the developers won’t ever get the game out the door.
Finding a game that’s clearly low quality and saying it’s bad is an easy approach but not the approach we take here. A more effective way to proceed is to look at games and designs players often say have bad design and discuss what’s going on. Games that have designs clashing with each other or targeting different audiences at the same time are likely to be unsuccessful. Other games that players online identify as having bad design are targeting a different audience or using a normally ineffective design for a reason.
What games have bad design?
Games that have bad design include games like Battletoads, 1000xRESIST, and The Guild 2 because they haven’t quite achieved their design goals or the right level of polish. However, many successful and acclaimed games, including those just listed, have elements that are considered bad intentionally. Difficulty, obscurity, and frustration in games like Fear & Hunger, Dark Souls, and Caves of Qud are perfectly fitting for those experiences.
Older NES-style games tend to be very difficult and attract criticism from modern gamers for this reason. The intended experience of these games is for a different world of gamers, so while some aspects of the difficulty seem like bad design, they work well when considered in their intended context: NES games are meant to be finishable in one sitting. What makes them engaging for the player despite their brevity is the high level of skill required to complete them in that one session. Players must play and replay to become skilled enough to get to the end, at which point they now have a high score to chase and improve.
Battletoads shows what happens when difficulty isn’t implemented well even for its context. Battletoads has been called one of the most difficult games of all time, and the challenges are less effective here because they aren’t well-taught or well-telegraphed. The game starts as a rote beat ‘em up game, but each level introduces new types of movement or a new challenge. The player has no chance at mastery when one level involves driving, another rappelling, and another riding a giant snake.
The player not only has little chance at mastery, but the challenges aren’t dynamic enough to encourage replay. Turbo Tunnel, for example, has the player dodge obstacles on a speeder at increasingly higher speeds. Navigating the tunnel comes down to memorizing the track layout because obstacles at the end hit the player mere moments after they appear. The level certainly challenges the player, but once the player has beat the level, they’ve memorized the layout. Replaying the level doesn’t involve using a newly developed skill but instead showing off how they’ve mastered a single, static track.
Unpredictable attacks and challenges come back to be a problem in Ninja Gaiden. Ninja Gaiden is an action platformer which starts simple enough but ramps up the difficulty with increasingly dangerous platforming challenges. The combination of melee enemies, ranged enemies, and bottomless pits make for a hectic experience. While the mass of enemies create an interesting challenge, the fact they endlessly run from off-screen makes planning jumps between platforms extra-challenging. The inability to predict where enemies come from makes the level feel like endless trial-and-error for a casual player, without truly mastering a new skill.
The opposite problem with difficulty is when a game is too easy or lenient on beginners, a problem skilled players identify in Mario Kart. Mario Kart is a casual series aimed at general audiences, and the designers want a mixed group of experienced gamers and complete novices to be able to have fun together. The issue is that the skilled players always win in a completely fair and skill-based game. That state of affairs isn’t fun for the skilled players who’ll be bored and the unskilled players who’ll always lose. Mario Kart includes catch-up mechanics for players at the back of the pack as a result. The blue shell is a famous example, which is only available to players behind second place and homes in on the player in first place. Random powerups restricted to players at the back ensure new players have hope of getting ahead in the rankings while at the same time keeping skilled players on their toes.
The issue with Mario Kart’s approach is that the catchup mechanics alienate skilled players in service of the broader audience. Locking first place racers out of powerful items has the unintuitive result that getting better at the game makes the player less effective. There isn’t a good solution here. Making the game completely deterministic (that is, predictable) is great for strategic gameplay but creates a terrible experience for novices and casual players who’ll never beat experienced opponents. Adding a dash of randomness is necessary for the target audience but inevitably creates a game that is “unfair” for a skilled player.
Challenges which are difficult, awkward, or impossible to overcome for even skilled players have their place in survival horror games. Battletoads causing frustration, fear, and anger is a problem, but a horror game evoking the same feelings is just the intended experience. Alan Wake 2 is an example of a survival horror game with an intentionally awkward combat system. The player must weaken enemies with flashlights before they are vulnerable to melee or gun fire, which makes the player feel more helpless when in conflict with the game’s threats.
Unfriendly design is effective in tense, story-based games like Heavy Rain. Heavy Rain not only includes difficult choices but allows the player to face permanent consequences if they fail. One of the characters, Ethan, is asked to perform dangerous tasks to learn the location of his son: drive against traffic on the highway, crawl over broken glass, or commit murder. Failing changes the ending but doesn’t result in a game over.
Fear & Hunger similarly takes control away from the player to sell the horror experience. A standard RPG empowers the player to build their character as they see fit and become powerful enough to overcome any challenge. Fear & Hunger’s dungeon full of horrors is going to be disappointing if the horrors are easy to defeat. The RPG combat is punishing as a result, with the player having little control over whether the monsters they face tear off a limb or kill one of their party members in a single attack. The grueling combat encourages flight over confronting the horrors of the Fear & Hunger dungeons directly.
FromSoftware’s Souls games, like Dark Souls, Bloodborne, and Elden Ring, are often criticized for their punishing gameplay. The advantage of the games’ challenge is the satisfaction they bring players. Souls games choose to withhold information from the player so they find solutions organically. Finding solutions on their own makes players feel much more satisfied, if they’re able to overcome the challenge. However, the danger of leaving solutions up to players is that many players never find a solution, which leads to frustration over time and money wasted. This obscurity is one thing for optional content, but the fact that the path forward in Dark Souls’ Anor Londo leads up along a buttress is especially egregious, as there’s nothing but a distant broken window to suggest it’s the path forward and not a piece of environmental art.
Souls games are intended for an audience of hardcore gamers, which changes the way we look at difficulty in those games. Players who aren’t able to find a way forward are likely not the intended audience for the game in the first place, as the “Prepare to Die” marketing for the original Dark Souls made clear. An absurd boss fight like the Capra Demon, where the player is boxed into a small arena, chased by dogs, and stunlocked to oblivion is clearly designed in favor of the boss, but this makes overcoming the challenge feel meaningful.
Souls games include ways of adjusting the difficulty as well, which turn unfair challenges into the motivation to improve and surpass them. The ability to leave messages on the ground for other players makes it clear that getting help from other players isn’t a cheat but an intended way to proceed. Summoning other human players for help trivializes virtually any boss in the Soulsbourne catalogue, Sekiro excluded. Considering the game as a singleplayer experience, players have the option to upgrade almost any stat: their attributes, the amount of healing flasks they get, the effectiveness of healing flasks, their shields, armor, and weapons in the case of the original Dark Souls.
Games like Caves of Qud and Noita show the risks of designing for a hardcore playerbase. Both games require a lot of investment from the player to learn. The balance between obscuring information from the player and giving it out freely is difficult to get right, as the patience and tolerance of players is going to vary even within the target audience. Adult players have busy lives and aren’t going to have extended time to devote to the game, meaning some players miss things not through a lack of skill but time. Tunic, Caves of Qud, and Noita ask the player to experiment and give their full attention to overcome challenges.
Caves of Qud has overwhelmingly positive reviews but is difficult to learn. The Steam page’s description shows much of what makes the game obscure and difficult: the game has over 70 factions, over 70 modifications/mutations, and 24 castes to choose from in creating their character. The game world is randomly generated and thoroughly simulated, with each creature in the game having its own stats, inventory, level, and all properties the player character has themselves. The number of ways a player is able to die is immense, and the only way for the player to learn about them is to experience those deaths. The simulation doesn’t reach the complexity of Dwarf Fortress but does alienate players who don’t have time to master the game.
Noita is a simulation-based game whose deep physics and arcane game rules make for a player-driven experience that is too off-rails for some audiences. The selling point of Noita is that “every pixel is simulated.” The game consists of procedurally generated environments that are 100% destructible, and gameplay emerges from the rules of the world: liquids flow, wood burns, smoke rises, oil floats on water, water condenses. The simulated world creates an endless number of interactions by giving the player wands that explode, freeze, burn, and melt. The game comes highly recommended, and the obscure rules are intentional, but the player does need to experiment to understand how to control such a complex world.
1000xRESIST is obscure to players unintentionally, which makes the game feel unpolished and awkward even to its target audience. The game is a visual novel with some light exploration and platforming between the dialogue. The focus on narrative instead of level and gameplay design damages the target experience, though. Every character is a clone of a single person, which means keeping track of the cast is challenging. The player also needs to wander through a labyrinthine hub to find the next part of the story, which feels like something players need to get through to get to the good stuff instead of part of the game.
A game that doesn’t teach the target audience well is frustrating, but a game like Psychonauts that spends too long in the tutorial is likely to lose players. Onboarding in games often lasts much longer than is clear in the game. Games let players go from the tutorial as soon as possible while hiding other lessons behind gameplay so players are able to get into the action ASAP. Psychonauts’ intro cutscene, exploration of the camp, and first level lasts between 30 minutes to an hour, which is quite a while before getting free access to the psychic cadet camp, the core promise of the game.
Users who identify bad designs in games like The Guild 2, the Majora’s Mask Switch 2 port, and Sword Art Online: Alicization Lycoris (SAO:AL) are identifying issues with polish. Polishing game mechanics is an aspect of game design, but a lack of polish has more to do with budget and time constraints than the design itself. Much criticism of SAO:AL and The Guild 2 has to do with polish issues: the camera struggling to track when switching characters, assets popping in close to the player, and poor performance. Poor optimization takes players out of the game, making what good is there difficult to appreciate. Majora’s Mask is another great game, but the port to the Switch 2 wasn’t well-considered: the camera controls weren’t modernized and it’s incredibly difficult to play without the N64 hardware.
A game that takes away control from the player in service of story runs the risk of giving the wrong impression, as in Hardcore Mecha, or taking the players out of the gameplay, as in Metal Gear Solid 4 and Kingdom Hearts 3. Metal Gear Solid 4 has an 80 minute final cutscene, which is still only a part of the nearly 8 hours of cutscenes over the course of the game. The optional codec dialogue adds another several hours. Not only is the excessive exposition less effective than experiencing it (i.e. show, don’t tell), players come to games for interactivity. Metal Gear Solid’s tactical espionage action isn’t to be found in a movie.
What are examples of bad game design?
Blind trial-and-error, grindy gameplay, and poorly managed output randomness are examples of bad game design according to players. A player usually is saying a design “hurt” them when they say it’s “bad,” but their feedback isn’t to be taken as universal or infallible. A designer must consider when that experience is an appropriate part of the game or not in alignment with the game’s audience and core principles. Each of these three “bad” designs have their own use cases and even positive contributions.
Trial-and-error in games is ineffective when it leads to frustration, which has a different threshold for different players. The long boss runbacks in Silksong and the Soulsbourne games are examples of the problem, where the player must run back through other enemies and face the same challenges over and over again which they’ve already mastered. Platformers like Celeste let players reset on each screen so they’re able to focus on the next challenge alone.
Effective trial-and-error comes about when the player has choices, given that different players have different thresholds for it. The issue mentioned above with Battletoads was that the player has no choice but to learn the single skill they require to beat that level, after which point that skill becomes worthless. In contrast, systemic games let the player experiment with trying multiple approaches. Dishonored and Metal Gear Solid V give the player a large toolset: lethal weapons, non-lethal options, interrogation, and other methods of subterfuge. The objective is often open-ended: an objective to “eliminate an adversary” leaves room for players to kill, threaten, blackmail, or even recruit the target. Players switch tack when a strategy doesn’t work, overcome the encounter, and learn from the mistake to take on the next challenge better. Rather than feel like they are aiming at a single target the developers intend, they feel like they’re discovering their own solution.
Grinding is considered bad design by the community but doesn’t have a lot of value as a term for a designer. Grinding is the need to perform a mundane activity repetitively to progress. The definition of “mundane” is subjective, and “repetitively” isn’t only subjective but an essential component of a game. All games consist of a series of repeating loops: Jaime Griesemer of Bungie succinctly captured the studio’s philosophy by saying game design is about tricking the player into re-experiencing the same 30 seconds of fun over and over. The trickery comes about with small variations. In the case of Halo, the player fights the same enemies over and over, but new weapons, environments, and vehicles keep gameplay fresh.
A game that feels grindy doesn’t have a problem with repetition then but the quality of the repetition. The repeated segments need a certain amount of variation to keep players engaged. The variation doesn’t even have to mean new content: an outpost in Metal Gear Solid V in the early game with normal enemies is fairly predictable, and the player’s able to rely on their binoculars, tranq gun, and CQC to take out most enemies. The game introduces increasingly tough enemies: enemies with helmets that block tranq shots, enemies with night vision to catch the player when they think they’re safe, and shields to block shots when all hell breaks loose. The same outposts with a new set of enemies force players to take a different approach each time they revisit them.
A game without challenges and special objectives runs the risk of feeling grindy even if the gameplay ought to be a hit. As Soren Johnson said, “given the opportunity, players will optimize the fun out of the game.” A single incentive or objective encourages players to aim at that objective with maximum efficiency, even if it’s less fun. The development of Diner Dash is an instructive example in how a game goes from grind to great: the designers said their game only became fun after giving a point incentive to chaining multiple of the same action in a row. The optimal way to play is to serve the closest table as soon as possible, but that method requires minimal thought from the player. They mindlessly hit a button until they’ve reached the end. Chaining actions (that is, bussing several tables in a row or taking several orders in a row) requires slightly more thought to positioning and timing, making for a more interesting challenge. No players went for this less optimal but more fun strategy until the point incentive came in.
Players tend to criticize random number generation (RNG) as bad game design, although the issue tends to stem from handling the two types of randomness incorrectly. A game that is completely predictable, after all, isn’t interesting. RNG is a crucial ingredient when it creates fresh experiences for the player and gives them maximum agency. The two types of RNG are input randomness and output randomness. Understanding the difference is the first step to understanding why RNG is ineffective for some games.
Input randomness occurs before the player takes an action. It’s most easily envisioned in a deck building game, where the cards you receive are dealt randomly, but the player then has the agency to choose which card to use from their hand. Randomness, in this case, hasn’t diminished their control of the situation. Input randomness as a term applies to situations outside deck builders too: a random placement of enemies in a level, random resource placement in Offworld Trading Company, or a completely random arrangement of cells as in Spelunky.
Output randomness occurs after the player takes an action and has a greater potential for frustration. Examples of output randomness include a chance-to-hit applying to an attack like in an RPG or rolling dice to see how far the player’s able to go in a board game. While input randomness gives players the chance to respond to random events, output randomness leaves no such room to strategize, as the player’s already made their move.
Games like Slay the Spire realized the advantages of effectively switching from output to input randomness during development. The game originally gave the player no preview of the enemy’s next move at all, but players had to select their card and hope that was the right move. The game is much more strategic when players know exactly what the effects of playing a card are, and adding the ability to see the enemy’s next move during development showed that it made the game more engaging. The final game retains this change.
Output randomness isn’t a problem when it serves the goals of the game, much like difficulty and obscurity in Caves of Qud or Dark Souls. RPGs make extensive use of output randomness in the form of dice rolls to create storytelling moments and mimic the complexities of a character’s skill level. A player fulfills the fantasy of building a character, but it’s up to that character to execute the actions in game. Output randomness creates that distance between the player’s skill and their character’s execution. Fear & Hunger, as previously mentioned, uses output randomness to make the player feel even more afraid and small. At a fundamental level, though, output randomness takes away the player’s agency, so including it must be the result of careful consideration.
RNG and what the player isn’t able to predict are the same thing, in essence. What makes output randomness irritating, unclear, and less responsive to skill also makes one-shots and deadly sniper enemies problematic. A one-shot leaves no room for the player to adjust their plan and respond; the player doesn’t have the opportunity to make the “interesting decisions” that make up games. One solution with no margin of error has a few issues: there’s no room for different skill levels, the optimal method of play is the only method of play, and mastery means there’s nothing left to experience in the game.
The correct way to handle one-shots and snipers is to make sure the attack is telegraphed in some way. A slow and well-telegraphed attack that is easy to dodge feels much less frustrating because the player knows what they need to do next time to avoid it. The player feels like it’s their fault and becomes motivated to overcome the challenge. A one-shot that is quick or tricky to dodge is only effective if a designer’s willing to sacrifice a larger audience for a smaller group of hardcore fans. The number of players who play games like Elden Ring at level 1 on New Game+ shows how a subset of the audience is perfectly willing to learn how to play without making any mistakes, but few players have that drive.
What is an example of a bad game character design?
Examples of bad game character design in multiplayer include old versions of Sion, Urgot, Yorick, Xerath, and Teemo. Giving one example of a character design isn’t helpful here because a bad character isn’t simply bad but has a specific problem: one is thematically unappealing, another ineffective in combat, one other is unfair to play against in PvP, and another has little agency. Here is a deeper dive into why some heroes from League of Legends missed the mark.
Multiplayer characters are challenging to balance for, as a designer isn’t just balancing for how fun a character is to play but how fun they are to play against. Taking sudden damage without the chance to counter is frustrating and removes a lot of interesting possibilities. Taking an attack, countering, having the enemy counter back, and then recovering in some way creates a memorable battle, and it tells a story in the player’s head. Teemo is a hero whose abilities weren’t fun to play against, as they were a hard counter to most top lane heroes and didn’t offer much room for countering. He laid down mushroom traps which were an instant and unavoidable punishment, and his ability to blind heroes wasn’t a tactic many heroes had methods to deal with.
On the other side, a hero that doesn’t have strong enough counterplay is going to be boring or unpopular. Classic Urgot was (and is) a ranged tank, which is interesting but also somewhat counter-intuitive. A hero without a clear niche is difficult to play, as his old ultimate was. The ability allowed Urgot to swap places with an enemy, which made him very difficult to siege against, but it wasn’t often the best move for a ranged character. He had a niche usage, but the new rework that gives his four knees canons better fits his role as a durable juggernaut.
Characters in a game like League of Legends compete with many others for attention, so a character must justify itself to the player. Each character has a distinct aesthetic, fantasy, effective use case, and balance of risk and reward. To the last point, a harder to play hero ought to come with higher benefits than an easier hero, otherwise no-one is going to opt for that playstyle. The old Xerath kit was a problem because the abilities were difficult to use. Xerath’s beam on Q was easy to miss, and Locus of Power prevented him from moving for the duration of the buff, a significant drawback. The lack of corresponding tradeoffs made him a less worthy option for players.
A character that goes after a unique aesthetic and is effective but doesn’t deliver on unique mechanics is going to have trouble drawing players too. Old Sion is an example of a character that follows a trope but doesn’t quite deliver. He’s a zombie, but most of his old abilities involved gaining health through attacks and getting bonus melee damage. His old passive ability, for example, gave a chance to take reduced damage from attacks, which isn’t something the player feels or notices as strongly. The new passive lets players come back and berserk on taking a fatal blow, taking advantage of the fact he is undead to give him a unique ability on death. Sion’s old ultimate boosted attack speed and added lifesteal for 20 seconds, which was useful but not unique or interesting. The new ultimate, Unstoppable Onslaught, puts the player in a barely-controlled charging mode that also makes the player immune to crowd control, which are both distinct advantages and tradeoffs.
The fix in each case is different per character and changes over time as the metagame develops. A live service experience always makes old characters obsolete as new characters and new strategies come into play. To take just one mechanic, Teemo’s ultimate ability, Noxious Trap, didn’t undergo just a single fix since day one but has received attention as recently as two months before the writing of this article. Designers have tweaked time for recharge, mana cost, range, Ability Power scaling, number of charges, and arming time. The largest update reduced the effect duration from 10 minutes to 5 and added the effect of revealing enemies who’ve stepped on the trap, making the attack less potent but have more utility. Faintly showing the collision radius when an ability reveals the hidden trap is another small change which made the feature less problematic. Constant small tweaks have gone into effect to address the issues since launch.
What is an example of a bad UI in game design?
Final Fantasy XVI (FFXVI), Modern Warfare 2 Remastered, and For Honor are examples of bad UI in game design, among others. While gameplay tends to be more subjective and more complex, UI design is more black and white. An effective UI keeps the player as immersed in the gameplay as possible while being clear and responsive.
A UI that adds friction and slows down the player’s connection to the game is likely to lose them. Extended periods of time spent in menus or waiting for a response aren’t ideal for any game. Some have criticized the CoD MW2 remake for its confusing and lengthy menus. Switching between campaign and multiplayer closes and re-opens the app, while getting to game modes within either menu requires navigating through several submenus to get to the right location.
A lack of immediate feedback is going to leave the player frustrated with the user experience. Hitting a button and having nothing leaves the player to wonder whether they did something wrong or the game did, and that moment takes them out of the gameplay. Another issue with slow menu buttons is that a player who accidentally went the wrong direction or accidentally opened a menu needs to wait for it to open just to close it. The smoke effects and loaded 3D models in Final Fantasy XVI’s menus, beautiful as they are, cause these issues by diverting resources to loading the effects in.
RPGs and adventure games have a large number of items for players to deal with, and the process becomes tedious without proper tools for sorting, filtering, and organizing those items. Tears of the Kingdom is a great recent title that unfortunately misses the mark with its inventory. The inventory screen itself works, but item selection in game takes place through a linear menu. This means the player scrolls through the entire list looking for the item they want, so an item in the middle of the list is going to take some scrolling.
Unskippable full screen overlays take away the ability for the player to decide how much time they spend with it, even if they contain important info. A second playthrough is more irritating when the player has no option for skipping lengthy cutscenes or information that they’ve already seen. Final Fantasy XVI answers the question of how much unskippable content is too much: each post-battle summary, level up, story chapter, and special target comes with an unskippable screen.
A HUD full of information at all times becomes overwhelming, and the game ought to provide options for modifying complex interfaces. A HUD with a lot of moving parts is helpful for beginners at least: having an icon with a button on it helps onboard players before muscle memory takes over. However, a cluttered UI is likely to block the player’s view and become completely unnecessary after the early stages of the game. The HUD’s colors sometimes clash with the in-game art or are difficult for people with certain types of color blindness to parse. Customization lets skilled players or players who have trouble seeing the overlays change the colors, where HUD elements appear, or turn some off entirely.
A lack of customization or effective customization options is a problem that’s rightfully being increasingly recognized these days. Implementation of many accessibility features is challenging, but UI is an area where some simpler steps are helpful. The ability to customize text size, font, and color, for example, lets players take charge and decide what’s right for them, rather than hoping the options provided in the box will serve most of the audience. For example, Deathloop’s hand-written style menu font is such a small component of the game, but the small size and scrawling text makes the game reach just a slightly smaller audience than its potential.
What distinguishes good game design from bad game design?
Experience distinguishes good from bad game design, since it teaches designers to understand player behavior and test their designs effectively. Common mistakes rookie designers run into are a lack of playtesting and not having a clear, specific vision for what goes into the game. Good game design is the scope of our entire website, so we’ll look at level design here to narrow down a few examples of effective and ineffective designs.
Novices tend to be overconfident in their designs, underdocument and underplan, and implement features before they know what they’re going to add. A consistent problem here is a lack of planning and a lack of knowledge about how designs will come out. A design idea is mediated by a lot of factors. To name a few: how it fits with other systems and mechanics, the bandwidth of the programmers to implement an idea, whether it fits the expectation of the target audience for the feature, and whether it’s worth the cost of implementing it. A successful design relies on more than the rockstar skills you think you have.
Getting past overconfidence and playtesting are the most important pieces for knowing whether designs are going to work. Getting a game in front of players as early as possible is the reason why designers start with ugly prototypes. Spending a lot of time figuring out the art and fancy features is going to become time wasted when players identify the fundamental flaws AFTER all that work.
Having a dream game or big project in mind from the get-go is a recipe for failure, especially for a solo indie dev who hasn’t made a project before. Designers who had successful first forays into the business make it look like no experience is required but they never went into the process cold; they knew how to create experiences for an audience before they took to the storefronts. LocalThunk, creator of Balatro, had never designed a commercial game before but spent years designing card games for his friends to play. He already knew how to improve designs based on feedback from his target audience.
The hallmarks of good game design have been discussed already, as they boil down to giving the player the most agency and the greatest ability to respond to challenges. Exceptions to this rule aren’t only possible but common in genres where disempowering the player or distancing them from their avatar is the goal. One area we haven’t discussed so far is level design, where the rules still vary from game to game but there are a number of consistent principles for “good” (read: effective) design. Level design isn’t my domain, but I’ve asked for help from another regular contributor, Brandon Dolinski, in discussing the topic. This section wouldn’t have been possible without Brandon’s help!
Effective design comes about through the contributions of many individuals, so whether a level comes out “good” or “bad” (and what we mean by that) depends on a lot more than just design. Often there were reasons for the choices that have nothing to do with the design of the game but with the execution and realities of the project (time, money, manpower, etc.). If you need broad strokes, good level design is always in service of the player fantasy and emotional experience. Bad level design is anything that breaks immersion or delivers the wrong emotion to the player.
Good level design (and game design for that matter) ought to have the following elements, generally speaking.
- Clear navigation: Nothing breaks immersion faster than not knowing both where to go and clearly where you aren’t supposed to go. Poor navigation comes both from design/art (it looks like I should be able to go there) and clipping, invisible walls, camera collision, and other tech issues.
- Clear objectives: Both technically—where do I go, how do I do it—and narratively, well-designed objectives align with the player’s internal fantasy and the story inside the game.
- Good sightlines/guides: Players pick up quickly on the design language we place to guide them to their objectives or even just to create a pleasant visual flow through a space
- Good layouts: A level ought to offer choices and have engaging flow (not complicated without a reason; good clearance around walls and objects; continuous navigation challenge, etc.)
- Pacing: Emotional narrative beats, challenges to the player’s skill, the controller feel, variety of the gameplay elements, intensity, and rewards all affect pacing.
- A sense of place and belonging: The level itself makes sense in a design/narrative sense of fantasy, and the architecture, art, and layout support that.
An effective design is about matching the intended emotions to the player’s experience. A long, boring walk down a corridor when the player ought to feel tense is an emotional mismatch and a “bad” design. For an example from a classic game, The Water Temple in Ocarina of Time is a situation where the emotions don’t match the action-adventure experience. Navigation is confusing; the maze-like multi-floor structure, misleading paths, and changing water levels make spatial understanding difficult (i.e., frustrating). This game isn’t about being scared or frustrated; it’s about adventuring and exploration! There’s also the fact that players must repeatedly equip and unequip the Iron Boots, a process many find tedious and interruptive.
There isn’t always a difference between a “good” and “bad” design, though, as in some cases the intended response requires unclear and frustrating experiences. An experience like Blighttown in Dark Souls is a great example. Dark Souls is a game that wants the player to feel frustration to make overcoming difficult situations feel more valuable. The Blighttown scaffolds break the pacing with frequent unfair deaths, the twisting paths give the player no clarity, and the whole experience causes frustration. Despite all that, though, once the player makes it through, the emotional impact is that the player feels they made it through hell.
Duke Nukem Forever shows what poor design looks like when it doesn’t serve a greater purpose. The game breaks the rules of pacing, engaging layouts, and serving the player fantasy throughout the game. The intended experience of a Duke Nukem game is an empowering experience putting players in the shoes of an action movie hero who uses imaginative and wacky weapons like the shrink ray to battle aliens and find secrets. Players expect interesting level layouts with lots of hidden areas, multiple paths, and engaging combat. Duke Nukem Forever, as a result of its troubled development cycle, misses the mark in this fantasy to say the least.
A franchise about engaging levels devolves into on-rails turret sections and corridors full of enemies to shoot. Boss battles and big set pieces that ought to change up the pacing in an otherwise linear experience instead keep the intensity low, as boss battles involve little more than center and fire, even glued to a turret in the case of the Mothership. The lack of options and engaging moves from the bosses makes for an experience that continues to be dull, and that’s without considering the story, visuals, or gunplay.
Is intentionally bad game design effective?
Intentionally “bad” game design is effective when it matches the intended experience emotionally and thematically, like any other game design. Trial-and-error gameplay, excessive difficulty, and both types of randomness have already been discussed as viable in their own ways. Three other characteristics of a game, (obscurity, cheap deaths, and weak PvP characters), are almost always ineffective but come with their own advantages, as the next three games show. Inscryption, Dark Souls, and Yorick in League of Legends are examples of experiences that use unfriendly, frustrating designs to their advantage.
Inscryption’s obscure and obtuse gameplay fit its horror theme, but the game doesn’t sacrifice the player’s ability to learn and master it. The game takes place in a cabin where a mysterious figure teaches the player how to play a roguelite style card game. The game throws a few curveballs at players as the game goes: a player must permanently sacrifice a card, which they only learn afterwards adds sigils to another card, or they must place an animal next to the fire, which a group of survivors quickly takes and eats. The strangeness and obscurity surrounds an identifiable core, though. The premise of a roguelite card game is familiar from titles like Slay the Spire, and the basic card game combat still feels strategic and engaging.
A frustrating experience that is still learnable presents a much more effective challenge than an experience that doesn’t consider the player’s abilities at all. An example is a skeleton from the original Dark Souls, a quick and deadly enemy that unfortunately appears near a lot of bottomless pits. The first time a player goes down into the Catacombs, they’re likely to get kicked off a bottomless pit. The game gives the player a number of ways to deal with skeletons after their first deadly encounter. Their frustration turns into motivation to find a solution: wear heavy armor for poise, be careful near edges, or better yet, use the camera to peek around corners and kick them off first.
Making a character intentionally “bad”, in this case weak, in a multiplayer experience is worthwhile if it serves a larger goal. Yorick is a character who received a buff to make time for the final fix. Yorick, in years past, was a pressing problem but one without a simple or quick solution. He was no fun to play against, blocking lanes with few options for countering. His passive reduction to incoming damage and boost to outgoing damage made for a frustrating experience, where an enemy tries harassing him, sizes him up, and then suddenly finds his defenses much higher on closing in. The choice to nerf Yorick was a lesser of two evils: let him remain a problem until finishing a rework years later, or make him less effective so at least he wasn’t as bad until then.
