Formulaic approach to enemy stats?

[WilsonE7]

So, I'm a bit of a math nerd, so as I'm designing enemies for my first game, I'm looking for a formulaic approach to minimize trial-and-error. I've tested RPG Maker MZ's default enemies (specifically, the Goblins) with the default starting party (Reid et al.) and they're pretty easy.

I wanted to see the mathematical relationship between the default actors' and enemies' stats, so I tried taking the average of the 8 classes' HP parameter and comparing that with each enemy's HP. The default database enemies (excluding the Hi-monster, which I presume is meant as a boss) had between 44% and 55% of the average of the 8 classes' HP. I did this because I wanted to see if I could just "scale up" the enemies' stats at higher levels relative to the party's stats.

Anyway, all this to say: Was there any actual mathematical consideration when designing the database enemies, or am I reading way too into this? Also, what methods (besides the time-consuming trial and error) have you found for designing and balancing enemy stats? I'm going to use trial-and-error until I find a quicker, more consistent method.

Thanks for your patience and your assistance!

 

[Cyberhawk]

Basic Enemies



Normal Enemy

MHP: Double the player's health.
MMP: Depends on the skills the enemy has. If it's a magic user, like a mage, give the same amount or double the enemies health. If it isn't a magic user, I recommend either none, or half of the enemies health.
ATK: Around 30% less than the player's attack if you are using low health and such as I am. (Using 15 health at level one.) If you are using stuff like 250 health at level one, keep it the same as the player's. If it is a magic user, I recommend half the player's attack.
DEF: Same as the player.
MAT: If it's a magic user, Around 30% less than the player's attack, or magic attack. (Which one is higher.) If it isn't, either none or half of the player's.
MDF: Same as the player.
AGI: I would recommend around one or two points higher than the player's agility.
LUK: Same as players.

Dungeon Enemy

MHP: Triple the player's Health.
MMP: Depends on the skills the enemy has. If it's a magic user, like a mage, give the same amount or double the enemies health. If it isn't a magic user, I recommend either none, or half of the enemies health.
ATK: Same as the player, or one point higher. If it is a magic user, around 30% higher than the player's attack.
DEF: Around 10% higher.
MAT: If it's a magic user, the same as the player's attack, or magic attack. (Which one is higher.) If it isn't, either none or half of the player's.
MDF: Around 10% higher.
AGI: Same as the player's or one point less if you use certain scripts.
LUK: Same or two points higher than the player's.

Bosses



For bosses it is pretty much the same as basic enemies, but with higher stats. For the experience, I give them a lot. Use the same technique as the one for basic enemies, but double the value of the experience. Then for the gold, give half of the amount from the experience.

MHP: 5 times higher than the player's.
For All the Rest: Times it by 1.5 higher (Or 150%) than the player's.

You can read this here. for more
This is an OLD guide from like VX iirc.
but I've been using it for the longest of times now. It still works for me in the long run.
Even in MZ this is how i balance it.

 

[Andar]

the default database isn't balanced in any way, it just gives enemies as examples and to test fights.

the problem with balancing here is that there is no way to know what level the party will have when approaching any enemy, because that depends on how the game's maps are filled - how many enemies there are before the player gets to any specific map will determine how many levels they have gained, and that will make any fight easier or more difficult.

As such balancing can't be done without having the specific game available, including maps, equipment gain and more.

that said, you can use something like excel or calc with formulae to test through different levels and enemy values to get a better starting point for balancing. You just have to get the variable level in somewhere and then decide how many xp which enemy is giving to get to those levels by which time.

 

[Cyberhawk]

I found EXP to be easy, It goes into the fundies of design, I will set a specific level you want to reach by the end of the dungeon.
(tends to be one or two levels)
If your a level above the boss is going to use a stronger skill to keep you on your toes.

As for counting how much per enemy
Normal (trash) mobs give out about 1/32nd of the player's level after the previous dungeon.
Dungeon enemies get 1/8th to 1/16th of the xp to the next level.
Mini Bosses:
I usually give them 1/2 or 1/8th of the xp to next level (Depending on the enemy(ies) used.)
Bosses
Guaranteed level up

FOr gold
I made it half of the Xp enemies give out.

 

[WilsonE7]

@Cyberhawk Thanks for the quick response! Thing is, I was hoping to use as many premade weapons, armors, skills, classes, etc. as possible, including using the default EXP curve. Not a big deal to change up the numbers. My game is meant to be pretty simple and linear. No shops, the player gets a new set of equipment at certain points in the story. That's why I was hoping to use the premade equipment. I'll give your method a try, maybe combine it with the defaults into something new that works.

 

[Frostorm]

Anyway, all this to say: Was there any actual mathematical consideration when designing the database enemies, or am I reading way too into this?

Honestly, I dont think ur overthinking things at all. However, you need to standardize your damage formulas and determine exactly what each stat does in your game. Only then should u start looking into mathematical proportions/relationships.

But to answer your question, yes there is a methodology (or at least I do) to determine unit stats. I can go into detail once I get home to my desktop cuz I'm on my cell rn...

Edit: As far as enemy units go, 1st I assign an arbitrary value for the unit's level, which will also be a general indication of the unit's difficulty. So let's say we have a lv10 Goblin Fighter, it'll have 100 points to distribute among the 6 primary stats (10pts per level). Base HP & MP are 100 & 60 respectively (level x 10 & MP is weighted 60% of HP in my project). Then I adjust & fine tune those values. For example, I will increase HP by 5 for every 3 MP reduced. Since this is a melee unit, we'll give this guy extra HP at the cost of MP. So we end up w/ 120HP & 48MP. Next, we'll distribute 100 pts among the 6 stats according to the role this unit will fulfill.
E.g.
ATK 22
DEF 20
MAT 10
MDF 14
AGI 18
LUK 16

 

[WilsonE7]

You can read this here. for more
This is an OLD guide from like VX iirc.
but I've been using it for the longest of times now. It still works for me in the long run.
Even in MZ this is how i balance it.

Hey, Cyberhawk, I was looking this guide over, and I was wondering: when making equipment, do you use the average of the party's relevant stats, or do you use the individual class's stat for the weapon. Like, a bow is meant for the archer class, so do you use the archer's stats at Lv. x or the party's avg. stats at Lv. x? I could see getting specific for mages, since their physical stats are usually much lower than their magical stats. This could be relevant because using the avg. vs. the individual affects the strength of the weapon to an extent.

 

[Cyberhawk]

Hey, Cyberhawk, I was looking this guide over, and I was wondering: when making equipment, do you use the average of the party's relevant stats, or do you use the individual class's stat for the weapon. Like, a bow is meant for the archer class, so do you use the archer's stats at Lv. x or the party's avg. stats at Lv. x? I could see getting specific for mages, since their physical stats are usually much lower than their magical stats. This could be relevant because using the avg. vs. the individual affects the strength of the weapon to an extent.

I actually tend to go up in increments of numbers for equipment.
So let's say my Short Sword has 7 ATK.
The next sword: the Long Sword has 14 ATK and so on to the final tier of that weapon archetype. I typically use six to eight tiers total.
I do this before adding in things like Equipment Slots, and the tiers available to the player at that point in the game.
Like for armors
The highest ratio variance or difference between two stats i've used.
I might give this type benefits and the specifics.

General Armor i leave this alone and that has the least total number of tiers.
(They tend to cap at 14 Defense)

Magic Armor provides the least DEF while giving the most MDF (I call it RES).
I think the highest ratio i've done for this is a 70/30 MDF/DEF difference
The defense given is half or less of the MDF.
Late game tiers give boosts to AGI or LUK


Light Armor gives a medium amount of Defense and a half of that is the amount of MDF given.
60/40 DEF/MDF difference
Late game tiers give boosts to AGI

Medium Armor functions like Light armor but has gives more Defense than MDF.
65/35 DEF/MDF
I haven't used this armor type yet but this is what i would do if I were to.

Heavy Armor gives the Most DEF and the least MDF.
The highest Ratio I used would be 80/20 DEF/MDF.
Mid Tier armors start to get lighter and will begin to lessen the impact on the unit's AGI.
Late game tiers don't reduce AGI at all and tend to increase LUK.

 

[WilsonE7]

I actually tend to go up in increments of numbers for equipment.
So let's say my Short Sword has 7 ATK.
The next sword: the Long Sword has 14 ATK and so on to the final tier of that weapon archetype. I typically use six to eight tiers total.
I do this before adding in things like Equipment Slots, and the tiers available to the player at that point in the game.
Like for armors
The highest ratio variance or difference between two stats i've used.
I might give this type benefits and the specifics.

General Armor i leave this alone and that has the least total number of tiers.
(They tend to cap at 14 Defense)

Magic Armor provides the least DEF while giving the most MDF (I call it RES).
I think the highest ratio i've done for this is a 70/30 MDF/DEF difference
The defense given is half or less of the MDF.
Late game tiers give boosts to AGI or LUK


Light Armor gives a medium amount of Defense and a half of that is the amount of MDF given.
60/40 DEF/MDF difference
Late game tiers give boosts to AGI

Medium Armor functions like Light armor but has gives more Defense than MDF.
65/35 DEF/MDF
I haven't used this armor type yet but this is what i would do if I were to.

Heavy Armor gives the Most DEF and the least MDF.
The highest Ratio I used would be 80/20 DEF/MDF.
Mid Tier armors start to get lighter and will begin to lessen the impact on the unit's AGI.
Late game tiers don't reduce AGI at all and tend to increase LUK.

So you don't actually stick to that guide. That only confused me... In my game, each tier of equipment is meant to be used at a certain level, so I'll plug the party's average stat at that level into the formulas given by that guide. For example, the guide says that weapons should be 50% of "the player's ATK," which I interpret as the party's avg. ATK. If my weapon is meant to be obtained around Lv. 20, I plug in the party's avg. ATK @ Lv. 20. That guide seems to work okay for me. I may make some minor adjustments to enemy stats.

 

[RachelTheSeeker]

There's a much better guide than the one provided here, especially as it's on this forum: Godlike Rodents. tl;dr is comparing stats between your character and the enemy, then asking "how many hits will it take to kill this character?" This applies to both players-vs-enemies, and vise-versa.

 

[Cyberhawk]

Oh the guide i totally went off a point on that part. I don't stick to it 100%.
I think it might be referring to individual classes.
Well it depends on how your setting it up in the long run.

 

[duty]

The way the default attack formula is written, every point of attack is essentially 4 points of damage and every point of defense is the equivalent of 2 extra HP.

When you're looking at enemy stats compared to player stats, you're really measuring how long you want a battle to last, and how many resources you want the player to spend to win.

If you take a level 1 swordsman class and goblin as an example: the swordsman damages the goblin for 25%-40% of its max HP with each attack and the goblin deals 10% of the swordsman's max HP per attack.

In a 1-on-1 matchup, the player will defeat a goblin in 3-4 actions and lose 30%-40% of its max HP.

If you had a dungeon with only 2 goblins that you fight one at a time, then this would be pretty well balanced. The player should be able to complete the dungeon with 20% or so of the lone swordsman's health if everyone's dealing average damage. Throw in a few critical hits or missed attacks one way or the other, and the player is finishing the dungeon just below 50% health - or not finishing the dungeon at all.

Although balanced, that dungeon would also be rather short and anticlimactic. So if you want the player to fight more goblins or even a boss battle at the end, you have to add extra party member, or provide healing items, or plan for the player to get enough XP to level up after the fits two goblins, or have the player find better equipment, or any combination of these solutions to increase survivability over multiple battles.

Ideally, non-boss encounters should last 2-3 turns. This gives the player the opportunity to be on offense and defense at least once every battle, but without eating up too much of the player's time and attention. And this is what you'll consistently get with any combination of 4 of the default DB actors at level 1 against the default 2xGoblin troop.

Personally, the default damage formula seems unnecessarily complicated. The defense stat is essentially hit points with extra steps and seems to obfuscate the actual benefit of armor. You can accomplish the same thing by having the attack formula just be the actor's ATK stat with a 20% variance, and leave all the default HP values the same.

This frees up the DEF and MDF attributes for different uses, like if you want a DEX stat to be used for ranged weapon damage, or a WIS stat that grants a bonus to healing.

You can also focus on making your equipment slots more interesting than just different opportunities for a defense bonus.

 

[Frostorm]

There's a much better guide than the one provided here, especially as it's on this forum: Godlike Rodents. tl;dr is comparing stats between your character and the enemy, then asking "how many hits will it take to kill this character?" This applies to both players-vs-enemies, and vise-versa.

I just read that article and noticed my numbers already align proportionally to the ones given in the article. For example, it uses a lv1 rat HP of 8 and lv1 player dmg of 3-5 (3 for weak hit, 5 for strong hit). In my project, a lv1 basic enemy has 38HP and my lv1 char does: (after armor calculation)

• ~22dmg using a 2H weapon
• ~15dmg using a 1H weapon & empty offhand
• ~15dmg x2 using a 1H in each hand (DW) <- suffers from -15% hit penalty for DW

I've also modified the way I approach enemy stats since my previous post in this thread. I now use Yanfly's enemy levels plugin. I give each enemy 50 stat points to spread among the 6 stats. I set base HP/MP to 15/10 respectively (since HP is valued at 1.5x of MP in my game). 15+10=25, which is half of the 50 I gave for stats, so that worked out nicely. In my project, DEF & MDF also increase HP/MP by 3/2 per point respectively, so I usually leave the 15/10 base HP/MP figure alone unless the enemy is a specialized unit (i.e. a pure mage or pure warrior type). But more often than not, I will adjust DEF/MDF 1st before touching HP/MP. On the plugin side, I use the following:

• HP/MP formula: base * level * rate + flat
  • Rate Growth: 0.50
  • Flat Growth: 0.00
• ATK/DEF/MAT/MDF/AGI/LUK formula: base * (1 + level * rate)
  • Rate Growth: 0.25
  • Flat Growth:0.00

So a basic lv1 enemy's stat block would look like this:


If I want to create an enemy that's strong for its level, I simply allocate more stat points for them. I haven't gotten around to it yet, but I'm thinking something like 75 for an above average monster and 100 for an elite monster. I also make it so that a "basic" monster has a -2 (negative) level fluctuation, while "elite" monsters get a +2 (positive) level fluctuation. The "above avg" monsters get ±1 level fluctuation. Levels are also map/dungeon based instead of player/party based.

Now I just gotta come up w/ a methodology for bosses...

 

[Tai_MT]

Honestly...

You're going to be using "Test Play" a lot regardless of how you decide enemies need to be balanced. At least, if you're serious about your enemies being balanced at all. You need to do this to ensure each encounter is exactly what you need it to be.

Personally though, the way I've balanced:
1. Low stats and simple formulas. I avoid using multiplication and division at all so that I can more easily control damage output. My enemies are meant to take 4 actions to defeat when you first encounter them and 1 action to defeat once you're experienced in dealing with them and know how to take them out. Player Characters are meant to be taken out in 5-6 actions at minimum.

2. Enemies aren't balanced using stats. They are balanced with skills. This involves balance through "economy of actions" rather than some raw numerical stat that most any gamer will "brute force" with a bit of purposeful overleveling. Enemies will inflict your party with powerful states. They will use your own stats against you (I have bandits who use your attack power to hit your defensive stat and this action is given 9999 priority in combat so it always hits first. They use this skill if your attack stat is high enough.). I have enemies that can also nullify your damage to them and generally make your life miserable if you aren't utilizing every action appropriately. The player is meant to use the "Attack" command as little as possible and be using "Items" and "Skills" as often as they can.

3. I've divorced stats from level ups. I did this for a few reasons. The first is that I don't want players grinding levels in order to finish the game quickly. The game is story driven, so to get stats, you do quests and engage with the story. The second is that it's too easy to break a well-oiled combat system that relies on gimmicks and priority targeting through brute force by just leveling up a bunch. That's not to say my levels do nothing, because they do have a purpose (often opening up new paths in the world, new shops, new quests, and things of that nature), but that purpose is "indirect power".

4. Skills each are potent and fill a purpose. I do not have elemental spells whose only difference is in the element they use. I do not have Fire 1, Fire 2, Fire 3, Fire 4 either. No skill is useful in every situation. Buffs are powerful, but not always useful. Skills fill their particular niche and every character has a "tool kit" they use with every skill emphasizing usage of that tool kit to provide synergy. Skills are also VERY RARELY "multi-target". Almost every skill in the game is single-target. Well, unless an enemy has it. Their "ultimate skills" often hit the whole party.

5. There are 3 attack stats and 3 defense stats. If targeting one of those stats with your attack seems useless, you can try one of the other two. Attack, Magic, Speed are the attack stats used. Defense, Magic Defense, and Reflex are the defense stats used. If an armored enemy has high defense, you can try a Magic based skill, or a Rogue based skill. If the player just hits "optimize" on their screen to have the highest Attack and Defense stats, then their other two attack stats and defense stats will be weak and they will be prone to being one-shot by enemies that use those other attack stats.

6. Equipment is limited and scales. Equipment is broken up into "Tiers" as well as "Sidegrades". Tier 1 may provide 1 Defense point, while Tier 2 provides 2 Defense points. A "Sidegrade" mixes the stats around to provide specific advantages towards builds. Instead of 2 defense points from Tier 2, you may get 1 Defense point and 1 Reflex point. Or, maybe you get 2 Magic Defense points. Or, maybe it offers no stat change and instead offers a passive 5% buff to Defense.

7. There is no Dedicated Healer. Magic Healing is too cost effective and renders almost every RPG "too easy". The only games that are difficult with Dedicated Healers are ones in which the healing had to be nerfed or severely restricted in the first place. Which, should tell you everything you need to know about how easy any RPG is as long as healing is quick, easy, and cheap. So, no Dedicated Healers. Buy your consumables, spend a turn to use them. Avoid damage through smart gameplay, or use consumables and turns to fix mistakes.

8. Regular encounters are meant to teach players mechanics and strategies. Bosses are meant to test players on the application of those. I spend time deliberately teaching my player how to break my combat just so I can ramp up difficulty and annoyance I can drop on the player later. I want the player to break my game, exploit mechanics, and feel powerful. That way, I don't have to hold back in what I can do to the player to hinder them.

 

[Pootscooter]

You're going to be using "Test Play" a lot regardless of how you decide enemies need to be balanced. At least, if you're serious about your enemies being balanced at all. You need to do this to ensure each encounter is exactly what you need it to be.

Too bad the "Test Play" button simply throws an error if you're using a TBS of any kind. Or if it doesn't then at the very least it's not gonna be representative of actual combat since a bunch of mechanics are missing. But I get your point: testing is key, TEST TEST TEST.

1. Low stats and simple formulas.

I like low numbers too, but not single-digit low. When it's that low, % stuff doesn't work well due to rounding.

2. Enemies aren't balanced using stats. They are balanced with skills.

They don't have to be mutually exclusive though. A game/combat system can be balanced w/ both stats and skills at the same time. It's just up to the dev to design it well.

 

[Frostorm]

Too bad the "Test Play" button simply throws an error if you're using a TBS of any kind.

Rofl, I feel ya on that (the struggles of an LTBS user). That's why I have a test map solely for battle testing purposes.

4. Skills each are potent and fill a purpose. I do not have elemental spells whose only difference is in the element they use. I do not have Fire 1, Fire 2, Fire 3, Fire 4 either. No skill is useful in every situation. Buffs are powerful, but not always useful. Skills fill their particular niche and every character has a "tool kit" they use with every skill emphasizing usage of that tool kit to provide synergy. Skills are also VERY RARELY "multi-target". Almost every skill in the game is single-target. Well, unless an enemy has it. Their "ultimate skills" often hit the whole party.

I totally agree w/ this. I hate when an RPG has Fire, Fira, Firaga, etc... Or even just different elemental spells in which the only difference is the elemental damage as you mentioned. That's why not only do I avoid using tiered spells like the Fire, Fira, Firaga example, but I also design spells w/ synergy as the main goal/focus. Basically, spamming a single skill over and over will yield very mediocre results. The player must combine the various skills/spells they have available, not only w/ themselves but also with their party members and the environment. For example, casting an Ice spell on a puddle of water or a unit w/ the Damp state will Freeze them immediately, whereas simply casting Ice Blast over and over requires 4 turns to fully freeze the target (Chilled->Frosty->Frigid->Frozen). Basically, there are "mini-rotations" and combos the player could (and should) do. I try to make every spell interact w/ as much stuff as possible.

Like your project, Buffs (or States rather) are powerful in my game as well. I noticed in many RPGs, skills that increase stats are almost never used by the player. Usually, it's because the buff's effect isn't large enough or the player simply doesn't want to spend that turn on a skill that doesn't deal damage. I addressed this by making various skills, including buff spells, a "Bonus Action". Bonus Actions don't consume your turn, but do cost resources to cast, whether it's MP or TP or w/e. Other examples of Bonus Actions include "Stances" and various Movement-type skills (e.g. Leap). Buffs also make use of utility scaling, as does pretty much every skill/spell. This removes the need for tiered spells like Fire, Fira, & Firaga.

7. There is no Dedicated Healer. Magic Healing is too cost effective and renders almost every RPG "too easy". The only games that are difficult with Dedicated Healers are ones in which the healing had to be nerfed or severely restricted in the first place. Which, should tell you everything you need to know about how easy any RPG is as long as healing is quick, easy, and cheap. So, no Dedicated Healers. Buy your consumables, spend a turn to use them. Avoid damage through smart gameplay, or use consumables and turns to fix mistakes.

I'm curious to see how that'll work out for you. I have the opposite opinion regarding healing. In fact, it is a critical aspect of combat in my game. Many enemies will have healing as well (not every enemy, but usually at least 1 unit in the troop can heal). The "healing is OP" issue you mentioned isn't an issue since Mana is extremely limited in general. For the 1st few levels, a player character would only have enough MP to cast 2-3 spells. If they want to be a more efficient caster/healer, they'll have to build for it. That means equipping a Staff or other caster-type weapon and putting points into Hematurgy (mana management & caster utility skills/passives), Hieromancy (Holy-themed skills/passives), and/or Hydromancy (Water-themed skills/passives).

There is literally only 1 direct healing spell available to players (found in the Hieromancy tree). This is intentional since I've designed healing to be more than just simple direct healing. For example, Hieromancy is all about damage prevention, mostly via Absorption Shields, but also through other means like Spirit Link (shares/equalizes HP pool w/ an ally) or a dmg reduction buff. Many of the passives in that tree affect/improve the Shield spell (this is the skill that grants an Absorption Shield). This can range from granting the Absorption Shields a dmg reflect component to making overhealing (from using the direct Heal spell) grant an Absorption Shield for the value overhealed. This tree also has the only Revive spell available (at the bottom of the tree though).

Hydromancy, on the other hand, is all about HoTs (healing over time). This tree also contains the only AoE heal available to players (Healing Rain). However, even that is a HoT spell. Many of the spells in this tree can also be used offensively, not just for healing. For example, the spell Drench has 2 components: 1st it attempts to remove Burn, Disease, or Poison on the target, then it also inflicts the Damp state, which increases both Healing received (from Water-themed heals) as well as Frost damage received by 30%. I've also mentioned that Damp makes the affected target susceptible to Freezing. So it's like a double-edged sword, but in most situations, it's not a problem.

In short, I've tuned the encounters to be pretty difficult if the player doesn't have access to some form of healing (including Absorption Shields & HoTs as mentioned above). They'll have to be smart w/ their MP management. There are many ways of regenerating MP (not just via the +4% MP regen from Staffs). Most of the methods of gaining MP involve a combo w/ certain spells or a certain setup/rotation.

In addition, I went the opposite route in terms of item usage. I discourage the use of items as much as possible. I want players to use their skills/spells instead of relying on consumables. Hence, HP/MP Potions are in the form of HoTs as well, which also simulates "metabolism".

Edit: Btw, what's your definition of a "dedicated healer"? Is it a character that can ONLY heal? Or does a caster-type character that has healing spells in addition to non-healing spells count? If it's the former, then I agree with you regarding the exclusion of a "dedicated healer". If it's the latter, well I've already typed enough on that matter...

Anyway, there's no right and wrong way of doing things. The various mechanics simply have to culminate into a cohesive and balanced battle system. So it's hard to say whether a certain specific mechanic is balanced or a good/bad idea. It's all about the context.

 

[Pootscooter]

@Tai_MT @Frostorm Jeeze what's with the insanely long posts?!

Alrighty, back to the topic at hand... No wait actually I want to have a word about consumables/items...
I just want to say that I never use consumables or if I have to use them to beat the encounter, then I'm just gonna stop playing. I simply HATE consumables, they're unrealistic and disrupt the flow of combat. Imo, it's worse than the "Magic Healing makes the game too easy" thing.

Ok, now really back to the topic of "Formulaic Approach to Enemy Stats"... I simply imagine enemies as another player character so that the stats balance on both sides. I even give them skills the player has access to, using the same formulas. Oh, regarding formulas...I used to use something REALLY simple (like @Tai_MT mentioned). I'm talking extremely simple, like: a.atk for Normal Attack (I use armor scaling plugin so the DEF part isn't needed in the formula box). For spells, I used to use a.mat * x, where x is a coefficient of my choosing. But I got bored with such simple formulae, so I ninja'd one of the formulas @Frostorm posted in the formula thread:

Physical Skills: a.atk ** (1 + a.agi / 1000)
Magical Spells: a.mat ** (1 + a.mdf / 1000)

It's basically the same as a.atk or a.mat alone, except it gains a small bonus from another stat (you can use any stat btw, it just happens to be AGI & MDF respectively). I like the idea of using the 2nd stat to nudge the dmg a bit.

 

[Tai_MT]

@Pootscooter

I'm a "writer" essentially. I paint with words. This is my canvas. You sort of need to expect long posts if you are having a conversation with me. This is the best and most effective means I have of expressing myself at all.

As for the low stats comment:

Most of my stats are single digits to start with. HP and MP being the exceptions (they don't really have much more than 10-20 at game start for any given character). Toward the end of the game, I do expect the player to have somewhere around a couple hundred into each stat (or more, as the rewards tier up for each "phase" of the game). For example: Early in the game, players will receive 1 or 2 points per stat for each quest completed. At the end game, they're looking at a boost of about 20 points into an individual stat for each quest completed.

The system itself is more "the stats only matter if they're the correct stats" rather than, "all my stats are really high, so I can steamroll everything!". It does little good to have all three attack stats at over 500 when the enemy has "Physical Attack Resist" slapped on it at 80%. In such a case a skill would have to be used that uses "Magic" instead of "Physical".

My system is mostly about gimmicks and strategy and the stats don't matter all that much in terms of creating/preserving challenge. You can still go back to early areas with your buffed stats and do thousands of damage to the early game enemies, but it's a method that scales well with stronger enemies.

@Frostorm

My skills will "Tier Up" as the game goes on and the player is allowed to choose how they want their toolkit to work for each character. As an example, the Mage type has two overall playstyles:
1. DPS. Lots and lots and lots and tons and tons and tons of elemental damage.
2. Controller. Access to almost every state in the game and the ability to inflict them (almost) at will.

I have no need for "Fire 2". Fire 1 turns into Fire 2 and the player just decides what that will look like. The skills can even be a "middle ground" between Controller and DPS. Though, honestly, the skill name never actually changes. If it was "Fire" before, it is still "Fire" after the upgrade.

I made my buffs powerful for the same reason you did. Players rarely use them as there's often no need to. No need, or it wastes a turn for not much benefit. My buffs START at a 25% gain. Some can achieve a 100% gain. Early on, the buffs are meant to be used at boss fights or with a full party. The buffs are fairly useless with low stats or without a lot of party members to capitalize upon the spent action of casting the buff (my buffs tend to be "Full party" unless they'd be too powerful). All buffs last 3 turns, so players shouldn't cast them unless they plan to DPS the enemies. Later on, the buffs can shave turns off of large groups of enemies the player will fight, or play into some of the more powerful skills and multiply damage by a significant amount.

Also, by "Dedicated Healer", I mean a character/build specifically designed to spend most of their time healing the party. You bring them along because they heal.

I have two "healing" skills in the entire game. The first is a "self heal" and only heals a percentage of the user's max HP (it starts at 10%, which is 2 HP healed at the beginning of the game. It can go up to 40%). This exists as part of the "Adrenaline" toolkit. It is an MP activated skill that works in tandem with the other 5 TP activated skills... most of which are highly self-destructive to the character using them (this character has a skill that does more damage based on missing HP... and another skill that synergizes with it to make them immune to death for a few turns while also eating up a percentage of their HP. This can be further synergized with stat buffs the character can employ like "Berserk" which makes the character uncontrollable for a while, but with the added benefit of multiplying TP gain from every source for a few turns and having a percentage increase to the Attack stat... and then there's the "Battle Cry" skill which buffs the entire party with a percentage increase to one or several of the attack stats). The purpose of this healing skill is essentially to drop it on the character who has hit 0 HP in order to keep them from actual death when the "death immunity" wears off. But, it could also be used in a pinch or something too. The character who gets this skill starts with only two casts of it because of a very small MP pool. The player would have to invest in the MP pool of this character to get more uses... but it's the only MP based skill the character has and it only heals himself.

The second healing skill is just a "Revive party member" skill. It's a TP usage skill. Requires all 100 points. It can be turned into a "revive everyone" skill if the player wants. It has a side effect of also applying a state to anyone revived that increases their consumable effectiveness by at least 50%. this can also be useful because I have some consumables in the game that can only be used in combat... and they have combat buffs on them as well as regular healing properties.

Otherwise, if the player wants to heal at all, they're stuck going to an Inn or buying consumables to use inside or outside of combat.

I honestly haven't had a lot of other people test my combat (because it's designed in such a way that testing an individual fight is near worthless... a player needs to interact with it as a whole for any data to be useful to me), but what little I've run myself has forced me to exploit enemy weaknesses more often, counter enemy skills/mechanics almost all the time, and keep a good supply of healing items on hand in case I muck up or the enemy lands a critical hit or something.

We'll see how that holds up later once I've got a good chunk of the game and monsters all scripted out to allow for a much wider scope in data collection.

I'm actually kind of curious if you've test played for a version of your game in which players build for the healing like you suggest they can. How effective is it? How much difficulty does it remove? As a player, if it looks like I can build a "Dedicated Healer" for my party, I generally do that just because it's the most efficient thing to do for your party and MP healing is insanely efficient even at x3 costs and reduced MP pools.

 

[Wavelength]

There was clearly no thought put into the stats of the default MZ database enemies aside from "make each one's HP somewhat bigger than the last one's". Same goes for everything else in the MZ database - "make each weapon's ATK somewhat bigger than the last one's in the same class".

XP, VX, and Ace had a little more thought put into theirs, with the idea that you could create a decent-length game from the sample data and have it at least be playable, but in most cases it still wouldn't be balanced or fun if you did that.

So, the better question might be should you take a strict formulaic approach (each stat scales), a loose formulaic approach (the total of stats roughly scales), or a trial-and-error approach (go with your heart and then adjust based on playtesting) to enemy stats when you're designing your game?

I would say avoid a strict formulaic approach unless you are absolutely awful at balancing and have no one to help you. While it can be easy to have both the player and the enemies scale by 20% in each stat per "level", for example (with the enemies' level being what you'd expect the player's level to be in the area), this makes for very boring enemies that all take the same number of hits to take out and all present similar levels of threat. I've seen this in some mobile games and it dampens the fun a little.

Whether you take a loose formulaic approach, by using a similar method to the above and then adjusting some stats up and some stats down for each enemy depending on what they specialize in (e.g. a Harpy might have higher AGI and lower DEF; an Earth Golem might have higher DEF and lower ATK) - or whether you take a trial-and-error approach where you take a guess at fair stats and then playtest it against a party with reasonable levels and gear for that part of the game - is up to your style. Personally, I'm of the persuasion that playtesting experience is a more valuable guide than any kind of formula is (because there are so many underlying factors in balance like skill selection, resource fatigue across a dungeon, and party composition), so I always opt for the trial-and-error approach myself.

All of these balancing methods are contingent on getting the EXP curves right, since you need to know (or be able to guess) approximately what Level the characters will be when they reach a certain stage in your game. Watching another player playtest is the best way to do this (they will inevitably get into more battles than you would because they don't know the layout of your maps), but if you don't have a playtester yet or you haven't created your maps before working on your database (that's fine!), you want to use the formula Average EXP per troop * Expected Number of Encounters fought to get the total EXP awarded for the map, then add that up for each previous map and compare to the "Total" (not "Next") tab on a Character's class to determine what Level you think they'll be when they first enter an area in your game.

 

[duty]

Adding onto @Wavelength - the stats you give to each enemy is going to set the optimal strategy to take out that foe during battle. And you want a lot of variation on those strategies.

Think of each enemy like a little puzzle to solve - with getting out of the battle with the least amount of lost resources as the reward.

You can have an enemy that deals a lot of damage with an attack, has lots of agility, but pretty much dies in one attack. If you don't want these baddies all acting first in combat and putting the hurt on the party, the player can use "always act first" or put actors in the party with equal to or greater AGI. All you've got to do is hit those mobs before they hit you and you sail through that encounter.

You can have an enemy that has a large amount of HP and splits into multiple enemies after a certain number of turns (easily done with a troop event). Focusing attacks to destroy this foe before it can divide means an easier battle.

You can have enemies that are not at all stellar in any way, but reflect magic or counter attack. If there are several in the troop, you probably want to take them out one at a time rather than use hit-all spells and skills.

You can make that one enemy that doesn't really do anything, but will eventually cast sleep on the party and leave you at the mercy of whatever else is opposing you. Eliminate drowsy mcsleep face first and you'll spare yourself a turn or two of helplessly taking hits.

In short, just try to make every encounter a bit memorable.