I want my game to feature deep strategy--for players to want to sit there considering their next move as in a game of chess.

Things like elemental resistances reward players who research their opponent, but the ultimate strategy of using fire attacks against ice monsters is still relatively shallow. Increasing the size of the elemental table, as in Pokemon, doesn't seem to make combat much deeper.

Things I've considered to create depth include using a multi-axis elemental system (say, there are hot-wet, and hot-dry elementals), and for elemental techniques to buff and nerf stats and elemental resistances so that players have to consider a more complex state as a fight evolves. What other mechanics might I consider to make my elemental system feel deeper? Or are there more fundamental avenues I should first consider in creating depth?

  • 7
    \$\begingroup\$ You may also be interested in How many choices for attacks should a player be given? I wrote an answer there describing how to use contingency or situational decision-making as a razor to judge which combat mechanics are adding strategic depth vs. which ones are more obvious calculations (like "use fire vs ice"). \$\endgroup\$
    – DMGregory
    Commented Dec 14, 2022 at 15:51
  • \$\begingroup\$ See also Balancing stats for combat difficulty in RPG Maker MZ. \$\endgroup\$
    – Theraot
    Commented Dec 14, 2022 at 16:19
  • 4
    \$\begingroup\$ Do you want/need to deepen combat strategy specifically by the use of elements? or do you want to deepen combat strategy AND the use of elements was your first thought? \$\endgroup\$
    – Josh Part
    Commented Dec 15, 2022 at 18:44

8 Answers 8


Elemental resistances usually create strategic depth but not tactical depth. Strategic depth means that they affect which characters are good in combat and which are bad.

Let's take Pokemon, for example. The elemental resistances / vulnerabilities and the different moves available to each mon mean that not every mon is a good matchup against every other mon. That means the player who only trains a single mon will fail as soon as they face an opponent that hard-counters it. So the player needs to train multiple mons with different elemental types and moves to make sure they have the right mon for every occasion. This forces the player to strategize while deciding which mons to invest time into.

But this only works when the player can actually decide what kind of characters they bring into battle. Many JRPGs don't allow that. The player has to fight with whatever characters are in their party at that point of the story and the options to customize the abilities of these characters are limited. So the player can't strategize based on elemental resistances and vulnerabilities. They can only make tactical decisions regarding which abilities to use. However, this gives you as the designer the option to make characters appear weaker or stronger during parts of the story where that feels appropriate for the narrative without actually changing their stats. You can make an individual character shine simply by confronting the player with lots of enemies that are vulnerable to that character's attacks.

Another way to make elemental resistances more interesting is by not making them static but dynamic. That means that enemies might change their resistances during combat. Either automatically, or triggered by the player's actions. That forces the player to decide what to do when. For example, an enemy that changes their resistances based on the element of the last attack that hit them. Or one that transforms through different forms, with each form having different resistances. Or you could have opponents that change their strategy when hit by elemental attacks (a common gimmick in 16bit Squaresoft JRPGs were opponents who would retaliate against each attack with an attack using the same element).

  • \$\begingroup\$ Technically, there is a tactical element to the Pokemon elements. It's just that the "tactic" in question is extremely boring (i.e. both players repeatedly switch Pokemon and occasionally attack, until someone runs out of PP and loses). \$\endgroup\$
    – Kevin
    Commented Dec 15, 2022 at 2:12
  • \$\begingroup\$ Not quite correct because it is the move that the attacking pokemon uses that determines the elemental type specified not the attacking pokemon's type so given a pokemon with wide types of moves then they can cover 12 of the 16 types so not many types of enemies can't be covered. \$\endgroup\$ Commented Dec 15, 2022 at 12:54
  • 1
    \$\begingroup\$ @user2617804 You do also have to consider your pokemon's defenses, though. That's usually a little less flexible. \$\endgroup\$
    – Arthur
    Commented Dec 15, 2022 at 13:29

Have you considered a drenching system?

Have elemental attacks buff the damage of other later elemental attacks from a different type

An example that makes sense is water attacks (and root of the name), make affected creatures vulnerable to lightning damage but grants resistance to fire damage (because they are wet). Fire attacks dry creatures out making them soak up more water when hit by a water attack, and do more damage to.

I don't know what other effects that you could do that would make sense out of game. But you don't necessarily need real world logic to work... This works best of the elements can be used in a spider wave of chains rather then a boring cycle of fire makes water better makes lightning better...


Typically, the assumption is that it effective attacks deal more damage. But as you've discovered, allowing the player to optimize for damage isn't necessarily an interesting system. Instead of relying only on damage, I would strongly consider pairing additional, non-symmetric mechanics or traits with your elemental system.

Here's a simplified example:

  • Paper
    • Bonus damage against rock
    • Smothering: hits an additional random target
  • Rock
    • Bonus damage against scissors
    • Smashing: bonus damage against full health enemies
  • Scissors
    • Bonus damage against paper
    • Bleeding cut: deals an additional damage over time

With the additional riders, in some situations it becomes worth considering attacks other the the basic rock-paper-scissors triangle. Rock has an incentive to be used as as an opener. Scissors has an incentive for battles that are likely to drag out & also makes a good opener. Paper adds some randomness as it may steer the player toward eliminating an injured target rather than focusing just on biggest threat. And if there's just one enemy left, paper effectively gets bonus damage.

Since the riders are asymmetric, each element has its own characteristic feel. Ideally, they provide mechanics that support your game's theme &/or story. It's okay to bend that rule in places. For example, if paper has lots of area of effect attacks & scissors has only one, then that one area of effect scissors attack will stand out. But I would advise against balancing to the point where every rider is equally available to every element as that effectively reduces the meaningful differences between the options.

  • \$\begingroup\$ While not a perfect execution, but that's how it works in SteamWorld Quest: three basic attacks for Copernica are: physical damage with no side effects; fire damage with a chance to apply DoT; and lightning damage that has a big variance in amount of inflicted damage and causes an extra card draw next turn. \$\endgroup\$
    – Dragomok
    Commented Dec 20, 2022 at 17:42

This is what I've been able to come up with:

  • Consumable items (or equipment) that buff elements. Of course.
  • You can have elemental chains/combos. In this case an element would get a bonus if used after another element. So it might be advantageous to use a weak element for the situation in order to set up a bigger follow up.
  • If you have the means to allow mixing skills (e.g. combined techniques by two party members, or by a single character by picking from its own techniques) then you could also elemental combination generate a different elements. You can take this further and say that there are only elements, and it is the combination of elements (performed during battle) which creates different powers/spells/magic.
  • Your enemies - in particular bosses - could change their own elemental resistances during battle. For example, a common trope is a boss that is weak to the last element is used. Another is a boss that has different weaknesses on different phases.
  • You could add another component into the mix. Such as the terrain, the time of day, or the weather. Which would also make some elements more or less effective. And you might even have magic that changes said component during battle.
  • Speaking of another component: give a random element have a buff each turn. To take it one step further, pull it from a deck built by the player, so the game is also a deck building game (and the deck could have other kinds of cards too).

Depth comes from the optimal choice not being straightforward to determine. As you note, simply increasing the number of element types makes it "harder" to determine the correct choice, but only in a trivial sense; it is still straightforward, it just takes longer.

You can make it harder to determine the optimal choice by making the outcome depend on more variables - so it's not just a matter of what the enemy's resistance or weakness is, but also e.g. their location in the battle arena, their distance from the player, what protective spells they have used, what attack you used last turn, or so on. This is sort of what puzzle games do - there is still one definitely optimal choice that does the most damage, but it's harder to figure out what that choice is, and if the puzzle is designed well then the player enjoys figuring it out.

You can also make it harder to determine the optimal choice by adding more variables to the outcome. Suppose the spell you cast doesn't just damage the enemy you target, it might apply status effects, it could damage enemies nearby, or allies nearby, or perhaps also damage yourself; it might affect the battle arena in some way, e.g. melting ice, moving an obstacle, blocking visibility. Or so on. The point here is that the outcome is not a one-dimensional space where doing more damage is better. This is more what strategy games are like - there isn't an obvious metric by which you can judge short-term actions, rather it's up to you as the player to make a judgement about the relative values of each possible outcome.


The thing that gives Chess its strategic depth is that the moves each side can make (and especially the moves that each side can safely make) depend strongly on every move that has been made so far.

Each side doesn't just have a simple current state (e.g. hitpoints and mana remaining) where it doesn't matter how they got there, the "fight state" state is a complex arrangement of all the pieces, essentially very little less information is relevant than the entire move history of the match. (At least until the end game where there aren't many pieces left; at that point lots of different move histories could have reached this state)

Combined with this, it's very difficult to gain a lot in a single move that cannot be countered (although novices such as myself may not reliably see the threats to counter them).

The net result is that every time you make a move you need to consider what state you're working towards several moves ahead, trying to gradually manoeuvrer your opponent into a weaker position over many moves. (And also considering what state your opponent is simultaneously trying to manoeuvrer you towards!)

Typical jRPG combat has you deciding what attack to use against what enemy (or what non-attack ability to use instead, e.g. heals, buffs, debuffs etc), trying to maximise the damage you're able to do to the other side and minimise the damage they'll be able to do back. A more complex system of resistances and vulnerabilities makes these choices more interesting. For example, if to win I need to both heal my party and kill some monsters in the combat so that I'm taking less damage, and if all attacks are equally good against all target, then I'll just use my strongest healer to heal and my strongest attacker to attack (targetting either the enemy with the lowest HP or the strongest attack). But if each of my party attacks with a different element, then which one is the "strongest attacker" depends on which enemy I decide I need to take down first; there can be complex trade offs where I'm sometimes better off attacking with my strongest healer, and therefore someone else needs to heal instead, etc.

But this kind of complexity still doesn't ultimately create that "chess like" flow of both sides angling for position over multiple rounds. The clashes still essentially "add up" in the HP damage exchanged (and resources depleted) over time, rather than forming a combinatorial explosion of possibilities. You can make the best move take more knowledge of the game systems and the enemies and more complex calculation to determine.

So if you're after a "chess-like" experience where players carefully consider their moves based on thinking through a long term strategy, rather than just repeatedly calculating the best move for the next round or two, maybe you need to think rather differently than a typical elemental resistance & vulnerability system. Maybe you can build up very powerful effects via sequences of elemental effects applied over multiple rounds, with the enemies also choosing between building up sequences of effects on your characters vs disrupting the ones you're trying to do to them. Or something along those lines.


Top of the list would be making the system dynamic in some way. IOW, instead of having a system where a given character only has one elemental attribute to their attacks, something allows the players to change the attributes, either outside of combat (for example, by modifying their weapons, or learning specific attacks), or in combat (perhaps you can swap characters in-combat like in Pokémon, or maybe they are able to spend an action in combat to just reconfigure their weapon).

This gives players more direct agency in choosing how they fight, in addition to allowing them to more efficiently utilize elemental effects. For example, if the game lets the player swap out elemental attributes outside of combat, but does not limit where they can do this, then players who have trouble with a boss can use this to make the fight easier.

The second thing I would recommend is to make the choice of element more than just a simple matter of looking up numbers in a table. Each element should have things it’s good at and things it’s bad at independent of damage, and ideally should also have specific buffs and debuffs. For example, maybe fire excels at applying a damage-over-time debuff, and specific fire attacks add additional effects to burning enemies. Maybe ice is good at reducing enemy movement or swinging the action economy in favor of the players.

There’s literally a million things you can do here, but the important thing is that this aspect helps incentivize playing differently for reasons other than just raw numbers, which in turn lends depth to the system because you have to consider more than just numbers.

My third recommendation is to have interactions between the elements that are not pure numbers. For example, it makes logical sense that using a fire attack on an enemy that’s frozen solid would thaw them out (this is actually a thing in Pokémon games for example). Similarly, maybe a water attack will put out an enemy that’s on fire. And the interactions don’t all need to be negations like this. Maybe ice attacks do bonus damage to enemies that just got hit by water attacks (the water freezes) or fire attacks (emulating thermal shock).

This further builds upon the second point above, making it even less of a case of simply looking up an element in a table to decide what to use, and incentivizes experimentation if combined with the first suggestion above.


Elemental Attributes are Bad for Strategic Depth

We've see it done so many times, that we just assume it is good for making a game more strategic... but it actually does the exact opposite. It makes the relationship between any two given adversaries more predictable which makes any other rules in your game less meaningful.

Let's take a common elemental chart like this:

enter image description here

Now lets say you have to choose between attacking with a Fire Hero who drops a single enemy's defense by 30% for 5 turns, and an Earth Hero who can AoE silence spells and poisons enemies for 90 damage over 3 turns, and you are going up against a Water hero who can reduce accuracy by 40% for 4 turns... who do you bring?

Knowing which power is best is really hard to guess and may change a lot based on the battlefield and what other heroes there are; so, without elemental factors to worry about it takes a lot of critical thinking skills to decide which one is better... but once you put in elemental bonuses, it becomes clear that you should attack with the Earth hero, because all other powers aside, you know he will do better than your Fire hero.

What normally happens in JRPGs is that your heroes have an element, a level, and unique powers. If your level is high/low enough, there is no strategy. The battle is a lost cause for one side or the other. Then you match up colors to offset some of that level to punch a bit above your weight class... while this is a satisfying experience for many players, it is just as deterministic as not having elemental bonuses. Now lets imaging you have 15 well leveled heroes to pick from and can only bring 3. Each of those heroes has some unique power that may or may not help based on the unique powers of your enemies. There are many matchups to consider, but if the enemy has 3 fire heroes, you will always want to bring your 3 water heroes no matter what. 12 possible strategies are immediately removed from consideration based on a counting and color matching strategy that a 4 year old can solve. If you remove the elemental bonuses and simply drop the matchup level of enemies a bit, then the game becomes more strategic because now all 15 heroes are valid picks. You can still hit above your weight class a bit, but only with well though out combinations of powers that takes a much more mature though process (and a fair amount of trial and error) to figure out.

Why Are Elements so Popular?

They are popular because they reduce the difficulty of a seemingly complex game. Game designers often have to choose between complexity and playability. If your game is too complex, then it will only appeal to niche audiences, but simple games appeal to people of all intelligences. Games like Pokemon are appealing to children because before you learn all the complex rules, you can still do well understanding only the very basic rules of levels and elements.

They can also be seen as helpful for making games where the point is to collect a lot of different heroes. If you have 100+ heroes in your game, seeing a lot of thier powers and roles overlap in inevitable. Elemental bonuses can basically be used to copy many very similar powers/roles across many heroes without making any one of them completely useless because having an Earth Healer might not be very helpful when fighting a Fire Army, so also owning a nearly identical Water Healer becomes important. This does not really make the game more strategic though, just makes it easier to stuff in lots of extra content and grinding to increase play-through time.

If you want a game to be enjoyed by millions of players of all ages and demographics and keep people entertained for a long time, then elements is a smart way to go, but if you want to make a very tactical game that focuses on thinking things through... it's actually better not to have them.

The deepest strategy comes from the vaguest METAs

Part of what makes Chess so strategic is that both players have characters with a wide range of powers, but no one piece can be uniformly described as superior. Pawns are numerous making them good sacrificial pieces for baiting traps and blocking movement. Rooks, Bishops, and Knights can all cover a lot of ground but have significant blind spots that can be exploited. The queen is a bit OP, but you only get 1 so you have to plan heavily around what the queens are doing... but at the end of the day, a well played pawn is every bit as lethal as the queen. Bishops don't beat Rooks because anyone playing a Bishop can always beet a Rook with one, but they beat them WHEN one player does a better job of maneuvering the bishop, than the Rook player does of maneuvering the Rook.

If you add elemental abilities, make them represent abilities and not paper-rock-scissors relationships. Maybe all Fire heroes have a chance to cause burn damage, all dark heroes have a chance to drop enemy accuracy, etc. That way, it becomes more about bringing the right hero for the job, and not any one of the right 30 heroes for the color matchup. Another way elements works is to tie them to skill trees: so, you could make it such that all Fire heroes get a common set of skills and bonuses they can get, (probably in addition to individual hero or class skills and abilities.)

  • \$\begingroup\$ "It makes the relationship between any two given adversaries more predictable which makes any other rules in your game less meaningful." - it makes the relationship between two given adversaries more predictable if they would otherwise be equally powerful. But that is just a balancing issue; put a level 5 character with fire magic up against a level 8 enemy who is weak to fire magic, and it's not so obvious. Coming at this from another angle, having more predictable outcomes isn't bad for strategy; strategy is about thinking several moves ahead, which requires some predictability. \$\endgroup\$
    – kaya3
    Commented Dec 17, 2022 at 23:19
  • \$\begingroup\$ @kaya3 Levels also make tactics less meaningful. If you have an enemy who is significantly weaker than the hero, then your personal choices may no longer matter at all. And when they do, it becomes a predictable formula you can work out in your head. You learn your hero can punch up X levels with an elemental advantage without having to think about it. So, the only time you need to plan is when the enemy is your level +/- elemental_pairing +/- a_couple_of_levels. But take away levels and elemental bonuses, and you have a game more like chess where every move matters. \$\endgroup\$
    – Nosajimiki
    Commented Dec 19, 2022 at 14:43
  • \$\begingroup\$ Then imagine my same comment but with "level 5" and "level 8" replaced with some other difference in power. Even in chess there are pieces which are obviously more powerful than others; nobody doubts that a King + Queen beats a King + Knight in the endgame, for example. \$\endgroup\$
    – kaya3
    Commented Dec 19, 2022 at 15:53
  • \$\begingroup\$ @kaya3 You can't look at last 4 pieces on a chess board as a representation of strategy. That's like saying strategy in JRPGs fall apart when both players are down to thier last heroes and 10hp left. Yes, a point comes when the outcome becomes certain, but you only get there as a consequence of all the strategic choices made up to that point. My point is that in chess there are countless strategies, and its hard to predict which will be good/bad. When you add elements to a JRPG, you reduce the number of viable strategies, and make the best ones really obvious. \$\endgroup\$
    – Nosajimiki
    Commented Dec 19, 2022 at 18:01
  • \$\begingroup\$ Then imagine my same comment without the "endgame" part. A queen is unquestionably more powerful than a knight. Chess has strategic depth even though some pieces have clear advantages over others; the fact that some pieces are more powerful and hence more valuable adds depth, it does not take depth away. So the analogy to chess cannot imply that a combat system with elemental advantages will have less depth than one without them. The fact that a pawn cannot take on a queen one-on-one does not reduce the number of viable strategies when you take the game as a whole. \$\endgroup\$
    – kaya3
    Commented Dec 19, 2022 at 18:34

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .