# Trouble with class stats

#### KeroTani

##### Veteran
I've done a fair bit of writer for my project and I've started the point where I need to start filling in stats for the classes. Some background from my project and the system I came up with. There are 5 races in my game that are the normal fantasy races: Human, Elf, Half Elf, Dwarf, Orc. Each character's base stats will depend on race and their stats on level up will depend on class. There are 2 tiers of classes for each character above the starting class and they can change their class once hitting a certain level. One of my goals with this project is to keep the numbers on the low side as this is my first project.

Any advice on coming up with stat curves?

#### ATT_Turan

##### Forewarner of the Black Wind
1 - Decide what "on the low side" means to you. Do max-level actors deal 100 damage on a good hit? 50?

2 - Decide how you want damage to vary based on level/parameters (do you do 0 damage to something that's one level higher than you? 5 levels? It never gets to 0?) and construct a damage formula.

3 - Use your attack and defense values as variables in that formula to determine what they have to be at any given level to produce the numbers you want.

If you're not sure how to evaluate or decide upon a damage formula, there are a number of threads where people have posted samples and described how they work mathematically.

Honestly, if you're not mathematically inclined, there's no reason not to just use the built-in A/B/C/D/E curves that RPG Maker offers.

#### LordOfPotatos

##### Veteran
I have 3 tips:

first, default maker stats are wack:

HP : 450-5350
MP : 90-1070
attack: 16-138
magic: 16-138
defense : 16-138
M defense : 16-138
agility : 32-277
luck : 32-277

those numbers are just bad to compare and understand as a player. and as a newbie developer, probably.
I use these:

HP : 1000-5000
MP : 100-500
attack: 100-500
magic: 100-500
defense : 100-500
M defense : 100-500
agility : 100-500
luck : 100-500

but you can do anything and it will probably be better than default. be sure to modify enemies and damage formulas if necessary.

second, maker has tools to create early and late stat curves. my advice is don't touch those and keep it linear. if you don't know what you're doing you can break your game balance in half and have difficulty spikes at seemingly random points because your stats took too long to go up or suddenly stopped increasing.

and third, try not to make characters TOO gimped in a defensive stat. having your squishies oneshotted is annoying.

#### KeroTani

##### Veteran
I'll take a look at these, thanks for the link.
1 - Decide what "on the low side" means to you. Do max-level actors deal 100 damage on a good hit? 50?

2 - Decide how you want damage to vary based on level/parameters (do you do 0 damage to something that's one level higher than you? 5 levels? It never gets to 0?) and construct a damage formula.

3 - Use your attack and defense values as variables in that formula to determine what they have to be at any given level to produce the numbers you want.

If you're not sure how to evaluate or decide upon a damage formula, there are a number of threads where people have posted samples and described how they work mathematically.

Honestly, if you're not mathematically inclined, there's no reason not to just use the built-in A/B/C/D/E curves that RPG Maker offers.

#### ATT_Turan

##### Forewarner of the Black Wind
first, default maker stats are wack:
I mean, it's fine for you to not like them, but why "wack" and "bad to compare"? A hit at 1st level will do an average of 3.5% of an actor's health, and at level 99 it'll do an average of 2.5%, which is pretty close to consistent. (presuming the default damage formula, of course)

Agility and luck don't inherently mean much of anything in combat (well, none of the parameters do, but I don't think there's much expectation that they'd mean anything in a damage formula) so it really doesn't matter if they scale at the same values.

I don't think the fact that they aren't, say, even multiples of 100 is a bad thing to most people...if anything, it kind of makes them look less artificial (for as much sense as that makes ).

#### LordOfPotatos

##### Veteran
I mean, it's fine for you to not like them, but why "wack" and "bad to compare"? A hit at 1st level will do an average of 3.5% of an actor's health, and at level 99 it'll do an average of 2.5%, which is pretty close to consistent. (presuming the default damage formula, of course)

Agility and luck don't inherently mean much of anything in combat (well, none of the parameters do, but I don't think there's much expectation that they'd mean anything in a damage formula) so it really doesn't matter if they scale at the same values.

I don't think the fact that they aren't, say, even multiples of 100 is a bad thing to most people...if anything, it kind of makes them look less artificial (for as much sense as that makes ).

by default maker stat logic this guy is not fast.
also he can't hit anything because his luck(dex in this example) is halved.
look at me in the eye and tell me that is not a violation of common sense e.e
the bars make it more blatant but looking at the equip screen is just as dumb.

also, "Agility and luck don't inherently mean much of anything in combat (well, none of the parameters do)" ?

#### ATT_Turan

##### Forewarner of the Black Wind
by default maker stat logic this guy is not fast.
You're using some kind of custom menu, by default there is no display that shows parameters as bars like that. If you want it to look better, you'd change what the gauges are considering to be their max value.

That's not any kind of problem with the numbers, it's a configuration of that display. According to the numbers you posted before, the numbers all start around 11.5% of their maximum values...so if the gauges were configured properly, they would be in the same place.

also he can't hit anything because his luck(dex in this example) is halved.
That's, again, not a thing in the default code.

also, "Agility and luck don't inherently mean much of anything in combat (well, none of the parameters do)" ?
...yes? Why do you think otherwise?

The only thing agility does in the code is get combined with the speed of a skill to determine when it happens in a turn; and get compared party vs. troop when determining preemptive/surprise attacks and escape. So those will work as expected so long as agility values are comparable to each other between actors and enemies, it has nothing to do with any other parameters.

And the only thing luck does is have an insignificant effect on whether you get a state applied to you from an enemy.

#### LordOfPotatos

##### Veteran
You're using some kind of custom menu, by default there is no display that shows parameters as bars like that. If you want it to look better, you'd change what the gauges are considering to be their max value.

That's not any kind of problem with the numbers, it's a configuration of that display. According to the numbers you posted before, the numbers all start around 11.5% of their maximum values...so if the gauges were configured properly, they would be in the same place.

That's, again, not a thing in the default code.

...yes? Why do you think otherwise?

The only thing agility does in the code is get combined with the speed of a skill to determine when it happens in a turn; and get compared party vs. troop when determining preemptive/surprise attacks and escape. So those will work as expected so long as agility values are comparable to each other between actors and enemies, it has nothing to do with any other parameters.

And the only thing luck does is have an insignificant effect on whether you get a state applied to you from an enemy.

like I said, it IS a problem with the numbers. the bars just make it obvious.

if these values are only used proportionally to the opponent (because they DO mean something in default, you just explained them) then there is no downside to making them equal to the other stats.

but there is a downside to having them like default, such as having to take the disproportion into account in stat graphic plugins or damage formulas and making stat recaps unintuitive like in the screenshot.

therefore having uniformy proportioned stats is objectively better.
like these:

#### ATT_Turan

##### Forewarner of the Black Wind
like I said, it IS a problem with the numbers. the bars just make it obvious.
No, a problem means something is wrong. There is nothing wrong in the screenshot you showed - different is not wrong. It might suit your emotional preferences, but that's not objectively wrong - that's why it can be easily changed in the editor.

if these values are only used proportionally to the opponent...then there is no downside to making them equal to the other stats.
I never said there was. There is also, however, no downside them being unequal. It doesn't mean they're "whack and bad to work with."

but there is a downside to having them like default, such as having to take the disproportion into account in stat graphic plugins or damage formulas and making stat recaps unintuitive like in the screenshot.
Most damage formulae do not use agility or luck, so their rate of scaling doesn't matter. If you're making a custom damage formula then you can just as easily make a custom parameter growth curve - the point is, the default settings all work just fine with each other.

As far as adding graphic plugins, well, if you're adding third party material to customize your game, you should expect to have to configure and customize it. None of this provides any reason for telling new developers that the default numbers are bad.

therefore having uniformy proportioned stats is objectively better.
No, that is not an objective truth. The parameters that are expected to work together are already uniformly proportioned.

You can find many game systems that are professionally made and popular where all of the numbers used with a given character are not using the same scale. If anything, it helps to illustrate that they don't interact with each other.

In any case, your logic is clearly not my logic - the OP has sufficient information to work with, so I'm gonna peace out. Just to mention it:

(because they DO mean something in default, you just explained them)
This appears to be intentionally and aggressively misinterpreting me, and I don't appreciate it. My original post said that agility and luck don't matter in combat, and that is exactly what I explained after you questioned me.

Agility affecting your encounters and escaping is not anything to do with resolving combat - it doesn't affect chance to hit or damage or anything.

And luck, as I said, is insignificant. To throw that at my face and say "it doesn't matter but it exists in the code so it DOES mean something" is clearly just argumentative. It certainly doesn't have any meaningful effect like your claim of "he can't hit anything because of his luck."

I was civil to you at all times, and merely wanted to know why you were telling someone new to the engine that these numbers were bad to use. There's no call for that kind of response.

#### LordOfPotatos

##### Veteran
"This appears to be intentionally and aggressively misinterpreting me, and I don't appreciate it. My original post said that agility and luck don't matter in combat, and that is exactly what I explained after you questioned me.
"
"The only thing agility does in the code is get combined with the speed of a skill to determine when it happens in a turn"

I'll point that out and peace out as well.

#### Sword_of_Dusk

##### Ace Attorney
therefore having uniformy proportioned stats is objectively better.
Do you even know what objective means? Because that is not an example of an objective thing. That's subjective as hell.

#### woootbm

##### Super Sand Legend
Seen people ask similar questions a lot. What a strange question. There are a million things players can hate about your game. Damage formulae is not one of them (unless they're a crazy person like me).

Ultimately, numbers are arbitrary; you can just guess and fiddle until it feels good. But if you want a real discussion:

So the first thing LordOfPotatoes gets at (I think) is the visual presentation/language of how the numbers look on screen. He seems to want numbers that are nicely divisible by 100 and symmetrical. This is, indeed, subjective. You've already got a leaning toward smaller numbers. Not sure what being your "first project" has to do with that, but smaller numbers are easier for users to digest quickly. And larger numbers tend to feel "diluted." Anyway, this is a whole philosophy that a lot of people (dev's included) don't often think about. So I'll just stop on this one.

The other is the actual effect on combat. By trying to decide on a "nice curve" first, you're trying to reverse engineer things. You should be thinking about how many hits an enemy takes to kill. How many more hits should a low damage character take to kill the same enemy, and how many a high damage character would. What numbers you use to achieve your design are irrelevant. You can have an enemy with 10 HP. Decide he should die in 2 hits on average. 3 on the low dps character, 1 on the high. So you might decide the average character does 5-9 damage, the low deals 4 damage, the high does 10+. Or you could give the enemy 1000 HP. Just multiply all those previous numbers by 100. The only advantage here is that you have have more randomness because there are so many more number possibilities between 500 and 999 than between 5 and 9. The actual outcome of combat would be the same, though.

A really good idea is to make a combat prototype. A prototype is just an ugly demo you build that allows you to see how things work. Emphasis on ugly, because you want to spend ZERO time on making it presentable. Your focus is purely on testing mechanics. So just throw default assets in there so you can fiddle with numbers all day without worrying about anything else.

#### Lord Vectra

##### Master Eventer
Oooo, math! I love math! I spend a lot of time on damage formulas but for multiple reasons (I'm the type to put conditions and skill effects in my damage formula).

Tip 1: Don't go too high or too low
20 - 50 are good numbers to always start with. If you do stuff like 5, it get's harder to balance because a single +1 makes them +25% stronger than before. If you go too high like 1000, it makes it harder for the player to comprehend what the numbers mean, and it makes divisons/multiplications really wacky due to the nature of large numbers.

20 is good because it's low but not too low. You can add/subtract or multiply/divide with less issues (usually). 50 is good because it's divisible by 100 making it good for certain situations. An example is an ATB system. Setting AGI to 50 by default makes it easier for the player to decide whose faster and by roughly how much.

But basically, for the developer, its good to go 20 - 50 for balancing. For the player, if number comprehension is important (discussed in tip 4), the same 20 - 50 applies but is, imo, more important to stay in 20 - 50 so the player can more easily comprehend what the numbers are doing.

Tip 2: Excel is your friend
Excel is indeed your friend in this endeavor. I use it a lot ot decide average damage, HP, how much a spell should cost, etc. I used to do the "play until it feels right", but I have adopted the formulaic route.

Tip 3: Go-To Damage Formulas
One decent damage formula to start:
ATK * (ATK / (DEF + ATK))
This means the ATK is multiplied by the ratio of the attacker's ATK and defender's DEF + your ATK. The reason your attack is added to their DEF is so if their DEF is 0, the player just deals damage equal to their ATK. If their ATK is, for example, 20, it's just 20 * (20/20) which is 20. if enemy's DEF is also 20, then you'd get 20 * (20/40) which goes to 10.

Found it in a game I play, and I kind of like it.

There is the simple way
ATK - DEF with a minimum damage forced in the code
it's the lazy way of doing things, but it is something that you can do. I wouldn't do the ATK * 2 - DEF that RPG Maker does because that is legit a terrible formula imo.

My current but complicated formula for my project
ATK / ((PDEF / 100) + 1) with minimum damage of 15% of ATK.
This makes it so if PDEF is 0, it's just ATK / 1. If PDEF is 100, then its ATK / 2. Each 1 point of PDEF decreases the ATK by roughly 1%. This makes it so higher PDEF have smaller returns the higher than ATK it is but since it's division, ATK increase also means the more ATK that is being lost from the PDEF.

Like others said, numbers are pretty arbritrary. I go the formulaic approach so that it's easier to organize and decide how much ATK a weapon should have, etc etc. Best you can really do is put things in place for numbers to be easier to manage for the future. When it comes to starting, it really amounts to how comprehendable you want the numbers to be.

For example, Mobile games (a lot of them) have incomprehendable numbers. You'll get a monster that has 3000 ATK and 2735 DEF and you get something that adds 234 DEF and 835 ATK and you deal 12,567 damage, but the player has no idea why. They just know big number leads to big number.

For some games, this is fine like Skyrim, for example. A Sword does 300 damage but you don't know the formula relationship of damage and armor rating but it doesn't matter for the game. Then there are games like Dungeons and Dragons where knowing the numbers IS the game..

### Latest Profile Posts

All the artists on here always share their art, and I wanted to partake in the sharing. I wanted to show off the album art for my game's soundtrack while also showing off this song. I wanted to take a break from all the pianos/strings/jazz. I experimented with some side-chaining techniques to modulate the dub-step like synth.
Crazy idea to help people be more productive. We all upload a photo of someone who is good at nagging, cajoling, guilting, etc. you into doing what you should be doing. Then, after you have posted X amount of times in a defined period, that photo pops up with the text of "Shouldn't you be working on your game?" Just for fun.
That is a really cute avatar concept that you have there! I like it :3
Well, here's a screenshot of just another game that might never be completed...