Situational Balance in Game Design
There's a reason your favorite game has patch notes every two weeks and your least favorite one felt "solved" by month three. The gap between them is almost always situational balance - the dark art of pricing things whose value refuses to sit still.
Situational balance is about designing things that don't have a fixed value. Their strength changes depending on the situation, which makes them much harder to balance than simple mechanics. A sword that does 50 damage is easy to price. A sword that does 50 damage but only against undead enemies is a completely different problem - and if undead show up in 5% of the game, the "correct" price looks nothing like the one in a game where they show up 80% of the time.
Value Changes With Context
A good example is area damage versus single-target damage. If you're fighting one enemy, both are the same. If you're fighting ten enemies, area damage becomes massively more valuable. The problem is that the game still needs to assign a single cost to both - a single mana cost, a single cooldown, a single gold price. You can't charge players different amounts based on how many enemies happen to be on screen when they cast.
This applies everywhere. Healing is useless at full health but game-saving when you're about to die. Counter spells are dead cards against a deck with no spells and backbreaking against a deck built around them. Anti-armor ammo is wasted on unarmored targets. Stealth is meaningless in an empty room. So their value isn't fixed - it depends on when and how often those situations happen, and more importantly, who controls when they happen.
This is the quiet question underneath every balance discussion: can the player force the situation where their tool is strong? If yes, the tool is worth way more than its average case suggests. If no, the tool is worth way less.
Expected Value Is Only a Starting Point
The usual way to handle this is through expected value. You look at how often a situation occurs and how strong the effect is in that situation, then average it out. A card that does 3 damage 30% of the time and 9 damage 70% of the time has an expected value of 7.2 damage. Simple.
Except it isn't. Expected value assumes situations are independent, random, and out of the player's control - and almost nothing in a good game is any of those things. Play testing matters more because players don't behave like dice. They build decks around the 9-damage case, they maneuver to set it up, they avoid the matchups where the 3-damage case triggers. By the time a skilled player is done, the "average" effect is 8.5 damage and your spreadsheet is wrong.
Magic: The Gathering's R&D team has talked publicly about this exact trap - they price cards assuming average play, then competitive players find a combo that forces the best-case scenario 100% of the time, and the card gets banned six months later. Expected value is a floor, not a ceiling, and your best players will always find the ceiling.
Offense and Defense Don't Scale the Same Way
One interesting insight is that offense and defense don't have equal value. If you're outnumbered, defense becomes more valuable because you're being attacked more often. If you outnumber the enemy, offense becomes stronger. So balance depends on context, not just raw numbers.
This asymmetry shows up everywhere once you start looking for it. In StarCraft, turtling is strong when you're behind and weak when you're ahead - so defensive units are priced to punish overuse. In Slay the Spire, Block cards are incredible against big hits and worthless against bleed effects. In any PvP game, healing is strongest when your team is losing and nearly pointless when your team is winning, which is why healers in MOBAs are often the most frustrating characters to balance - they swing the game disproportionately when the game is close, and disappear from the picture when it isn't.
The rule that falls out of this: defensive tools should cost more than their raw numbers suggest, because they compound. A shield that blocks 10 damage is worth more than a sword that deals 10 damage, because the shield stays useful across every incoming hit while the sword requires you to already be winning.
Versatility Changes Everything
Versatility also plays a big role. If players can switch tools freely, specialized options become stronger because they can always pick the best one for the situation. If switching is limited or costly, then general-purpose tools become more valuable. So even something like weapon switching speed can change balance.
Monster Hunter figured this out early: you pick your weapon before the hunt and you're locked in. This makes generalist weapons incredibly strong because you don't know exactly what the monster will do. Meanwhile, Destiny lets you swap weapons mid-firefight, which is why hyper-specialized weapons dominate the meta - if you have a room full of shielded enemies, you swap to the anti-shield weapon, deal with them, and swap back.
The same logic explains why Hearthstone's "tech cards" (cards designed to counter specific strategies) are priced low - they're dead weight unless the exact matchup shows up. And why Magic's sideboard system, where players swap cards between games, makes tech cards dramatically stronger and their prices dramatically higher. The switching cost is the balancing mechanism.
Shadow Costs Matter
There are also hidden costs, often called shadow costs. These include things like setup requirements or trade-offs. For example, unlocking a powerful unit might require building expensive structures first. That upfront investment changes the real value of the unit. Another example is opportunity cost, where choosing one option prevents you from choosing another, reducing flexibility.
Shadow costs are where a lot of bad balance lives, because they're invisible on the stat sheet. A Magic card that costs 3 mana but requires two other specific cards in play looks cheap, but its real cost is the entire deck you had to build around it. A Dota hero with one devastating ultimate has a shadow cost of "does nothing for the first 15 minutes of the game." A crafted weapon in an MMO costs not just the materials but the hours spent leveling the profession.
Civilization is a masterclass in shadow costs - a single powerful unit might require researching a specific technology, which requires specific resources, which requires specific terrain, which requires you to have founded your city in a specific spot. The unit's "real" cost is all of that put together, and when balance breaks, it's usually because one of those upstream requirements got cheaper than it should have.
Opportunity cost is the subtlest shadow cost of all. Every ability you equip is an ability slot you can't use for something else. Every stat point in Strength is a point missing from Intelligence. The cost isn't the thing you took - it's the thing you didn't.
Systems Create Their Own Balance Problems
In games where players can control situations, like building decks or combining abilities, balance becomes even more complex. Some things are weak on their own but extremely strong when combined. In those cases, you don't just balance the item itself, you balance the entire system around it.
Every deck-building game has had this problem. Hearthstone's Patches the Pirate was priced as if it were a niche card, but once players combined it with cheap pirates that summoned it for free, it dominated the meta for a year. Slay the Spire's "infinite" combos aren't broken individually - they're broken because three mediocre cards happen to form a loop that does infinite damage. Path of Exile has shipped entire leagues where a skill that looked fine in isolation ended the game in one button because the community found three interactions the designers didn't.
This is the scariest part of situational balance: the system isn't the sum of its parts. Two 5/10 tools can combine into a 20/10 tool, and no amount of balancing the individual pieces will fix that. At some point you stop balancing items and start balancing interactions - and the number of possible interactions grows exponentially with every new piece you ship. Game designers call this the "Nth card problem": the 1,000th card in your game is vastly harder to balance than the 100th, not because it's more powerful, but because it has 10x more things to combine with.
Why Some Games Embrace the Chaos
The cleanest response to situational balance is to eliminate it - make every option flatly equivalent and trust players to pick based on flavor. This is what competitive Smash Bros does by banning items and most stages. Pure skill, minimal variance, minimal balance headaches.
But a lot of games go the opposite direction on purpose. Rogue likes lean into situational value because the randomness is the point - a fire ability is amazing this run and useless next run, and discovering that variance is half the fun. Hearthstone's Arena mode, Slay the Spire, Hades, and every deck-builder ever made have embraced the fact that the same card can be a 2/10 pick or a 10/10 pick depending on what else you've drafted.
The takeaway for designers: situational balance isn't a bug you're trying to eliminate. It's a dial you're choosing where to set. Flatten it and you get a clean competitive game. Amplify it and you get emergent depth and replayability. The mistake is doing it by accident.
The Takeaway
The main takeaway is that balance isn't about fixed numbers. It's about context, probability, and player behavior. Something can be perfectly balanced on paper and still feel broken depending on how often its best or worst case shows up - and how much control players have over making the best case happen.
That's why situational balance is less about exact math and more about understanding how the game is actually played. Your spreadsheet tells you what something is theoretically worth. Your play testers tell you what it's actually worth. And six months after launch, your community will tell you what it's really, truly, actually worth - usually in the form of a Reddit thread demanding a nerf.
The designers who handle this well aren't the ones with the best formulas. They're the ones who ship, watch, and iterate - because the only reliable way to balance a situational effect is to see it survive contact with real players, then tune from there.