Getting Comfortable Working with Stats

TMS

Guy with Mask
Veteran
Joined
Apr 17, 2012
Messages
300
Reaction score
57
First Language
English
Primarily Uses
N/A
Before I can start in on the fun part of making a game, there are a number of technical aspects to take care of. I had a feeling that I would hit something of a wall here, and so I have. It's hard to work with all the numbers that go into character and enemy stats, experience points, and bonuses for leveling up. RPG Maker gives you plenty of examples to start out with, and I know there's going to be some trial and error involved, but I was wondering how the former newbies who preceded me went about getting used to working with stuff like that. Was it all just experimentation? There don't seem to be any tutorials on the subject.
 

bgillisp

Global Moderators
Global Mod
Joined
Jul 2, 2014
Messages
13,139
Reaction score
13,640
First Language
English
Primarily Uses
RMVXA
The advice I was given when I started out was figure out the stats for your base, or average class first. Once you have those set, set your other classes around that (a spreadsheet can help). So if you have say a Mage that is high in MAT compared to the base, you set that accordingly. Though, while doing this you will want to set your minimum and maximum (for example, in my game, I used the following: 15 - very poor, 20 - average, 25 - very strong for level 1 characters and stats).

Once I had my character classes set, I then set my equipment. I did it by equipment type, and first decided on a base, or average equipment, and set the other weapons around that. Once I had an idea what my weapon ATK (or MAT or both) boosts were, I then set my armors and accessories, keeping in mind what the average (and highest) stats were for my characters and weapons.

After that I did skills, and tried to figure out how much damage I wanted the skill to do at the average point you get it in the game (that took some guessing, and I'll be rebalancing that one later, but it's a start). Enemies came last, once all of the skill/weapon/armor/classes were set.

Just keep in mind you will need to tweak it. I think I'm on my 5th rebalance run for my skills already, though at this point the tweaks are more minor (for instance, one tweak I did was raise the boost from the skill that boosted ATK from +5 to +10, so there was more benefit to casting it*).
 

Kes

Veteran
Veteran
Joined
Aug 3, 2012
Messages
22,009
Reaction score
11,347
First Language
English
Primarily Uses
RMVXA
Shinma has done something which you might find helpful. In the Game Development section, go to Program and Utility Discussion and you will find a thread about Damage Calculation Spreadsheet. Can't give you a link as I'm writing this on my phone.

Anyway, this spreadsheet enables you to experiment with your stats and see instantly how changes you make work out. It doesn't help you on things like bonuses etc., but will give you a solid basis for balancing.

Another thing you could do is this. Somewhere on the forum is a thread giving links to unencrypted games. Pick one, download it and open it up to see how it was done.
 

Diretooth

Lv. 25 Werewolf
Veteran
Joined
Mar 10, 2013
Messages
1,228
Reaction score
442
First Language
English
Primarily Uses
RMMV
The main thing with stats is you either know what you're doing, or you don't.

This is tricky, since you can only get the latter through trial and error, and maybe some help.

If you know what you're doing, then that's great, but if you don't, then you just basically mess around with it until you reach something resembling balance.
 

TheoAllen

Self-proclaimed jack of all trades
Veteran
Joined
Mar 16, 2012
Messages
5,298
Reaction score
6,055
First Language
Indonesian
Primarily Uses
RMVXA
This is the spreadsheet that ksjp17 talked about

http://forums.rpgmakerweb.com/index.php?/topic/24429-damage-calculator-spreadsheet/

I never use this kind of spreadsheet before tho,

I usually test that in actual game and have the standard like, enemy can be killed using 2-3 times of normal attack.

While most of the skill's damage usually scaled from the normal attack, like 150%

Once the standard has been made, test that out on the battle test.

If battle test is fine, try that out in actual game, because you need to know how do you feel in repetitive battles.

Is that boring? Can you survive? Enemies are too hard? or too weak?
 

Wavelength

Edge of Eternity
Global Mod
Joined
Jul 22, 2014
Messages
5,199
Reaction score
4,516
First Language
English
Primarily Uses
RMVXA
There certainly are tutorials on the subject of stats and balance.  See the following two-parter for a pretty good one:

Early on in development, though, balance does not need to be a priority.  Just err way on the easy side, and make your game.  Create the content.  Don't get slowed down by worrying about perfect (or even good) balance, because as you add more features, items, etc., that balance will get thrown off anyway.

Later on, when most of the game is made and you do want to start focusing on balance and progression, that's when you'll want to test the game yourself a few times, make some initial tweaks.  This is a good time to use the guides above.  Then, once you're confident the game is balanced enough to at least be playable, hand it off to playtesters and hear (or, better yet, watch) their experience with your game.  If your game is flexible enough, they will play in ways you never expected.  You can further tweak the balance based on that.
 

bgillisp

Global Moderators
Global Mod
Joined
Jul 2, 2014
Messages
13,139
Reaction score
13,640
First Language
English
Primarily Uses
RMVXA
There certainly are tutorials on the subject of stats and balance.  See the following two-parter for a pretty good one:

Early on in development, though, balance does not need to be a priority.  Just err way on the easy side, and make your game.  Create the content.  Don't get slowed down by worrying about perfect (or even good) balance, because as you add more features, items, etc., that balance will get thrown off anyway.

Later on, when most of the game is made and you do want to start focusing on balance and progression, that's when you'll want to test the game yourself a few times, make some initial tweaks.  This is a good time to use the guides above.  Then, once you're confident the game is balanced enough to at least be playable, hand it off to playtesters and hear (or, better yet, watch) their experience with your game.  If your game is flexible enough, they will play in ways you never expected.  You can further tweak the balance based on that.
This. I did what I stated in my post above, but it was worked on as the game was made too. I would also second you should err on the side of easy, its hard to test a game when a boss is constantly killing you.

For the record, I usually put my boss stats in where I think they should be, but the HP is usually lowered in testing for the hard bosses. I usually take off a zero, so a boss that should have 2500 HP has 250 HP when testing. Easy way to make it easy for testing, and easy to switch back once done.
 

whitesphere

Veteran
Veteran
Joined
Mar 14, 2014
Messages
1,688
Reaction score
785
First Language
English
I personally never had a guide for stats and game balance.  What I did was a LOT of experimenting.

What I would do is use the nifty Battle Test button, after assigning my party members and their equipment.  Then I would compare the damage output of the monsters and party members to their max HP.  And I'd see if the damage was what I expected.  Based on that, I'd tweak the ATK or DEF of various items (or MAT or MDF for magic skills/spells).

Other posters have listed some really nifty tools that can help with game balancing.

In turn-based battle systems, the next most important attribute is AGI.  The way that works, the highest cumulative AGI goes first, then attacks go down the line.  For these, game balance is more a matter of "OK, the party's average AGI is 50, so which monsters should be faster than them or slower than them."  So I don't think it requires as much tweaking as the other stats.
 

TMS

Guy with Mask
Veteran
Joined
Apr 17, 2012
Messages
300
Reaction score
57
First Language
English
Primarily Uses
N/A
Thank you for the suggestions, everyone. Today I've started working in earnest with this kind of thing, and as you can imagine, it's been a slow process. Much as I'd like to make the game handle stat growth individualistically, I'm starting to think that it might save me some headaches just to base things like class parameter curves on the examples given by the program. Is that a good idea? Has it worked for people in the past? Is it a sign of weakness?
 

Kes

Veteran
Veteran
Joined
Aug 3, 2012
Messages
22,009
Reaction score
11,347
First Language
English
Primarily Uses
RMVXA
I think there are significant weaknesses in the default set up, not least of which is the massive numbers you are dealing with.  In my experience, it is much easier to balance things when stats are a lot smaller.  So instead of starting off with HP around 600, mine typically start around 120 or so.  Going back to a suggestion I made in an earlier post, here is the thread with the unencrypted games http://forums.rpgmakerweb.com/index.php?/topic/13580-list-of-unencrypted-games/  You can learn a lot by seeing how other people have tackled this and by playing through at least a portion of the game you will be able to see how those decisions work out in reality.
 

TMS

Guy with Mask
Veteran
Joined
Apr 17, 2012
Messages
300
Reaction score
57
First Language
English
Primarily Uses
N/A
You know, it's funny. When I tried adapting the stats to suit my game, 120 HP is exactly what I had the main character start out with.

But when I started trying to generate parameter curves, I didn't really want characters to much more than double their natural stats over the course of a game (it wouldn't seem very realistic), so sometimes they'd go several levels without really gaining anything. Of course, that was assuming they'd be level 99 by the end of the game, when really I have no idea what level they're likely to end up at. Maybe I could eliminate the problem if I capped their growth at level 50 or something.

I'll take a look at those unencrypted games and see if any have genres similar enough to serve as role models. In the meantime, any other advice is welcome.
 

kerbonklin

Hiatus King
Veteran
Joined
Jan 6, 2013
Messages
1,728
Reaction score
282
First Language
English
Primarily Uses
RMMV
What I did for my stat balancing between 4 PCs and 12 classes was capping the minimum total stat (excluding HP/MP) for each class (stats gained from levels depends on current class) at level 1. Every class at level 1 has the same total stat points but different curves, each class specializing a certain stat or has a balanced amount. At level 99, the maximum total stat value of each class is 850 worth. Class A would have 206 attack, 175 def, 154 m-attack, 157 m-def, 158 agi, and 50 luck, all of this totals to 850. Class B would have different values but still total to 850.

Also as a side thing, I made my level up stat gain increases permanent through class-switching, allowing players to make definite preferred stat builds by getting their levels as specific classes. Unless two different players never change their PC's classes and level straight to 99 (which is not recommended), their end-game stats on their PCs will always be different.
 

bgillisp

Global Moderators
Global Mod
Joined
Jul 2, 2014
Messages
13,139
Reaction score
13,640
First Language
English
Primarily Uses
RMVXA
I actually did something like that too. My level 1 characters I made the stats add to 120 (HP/MP excluded), so a character could be all 20's, or it could be 15 ATK, 25 DEF, 15 MAT, 15 MDF, 25 AGL, 25 LUK (just for example). Similarly, level 99 adds up to 480.

Exceptions to these rules were made for classes with unusually high HP or MP (they might add up to a lower number to compensate), or classes with special features. To give an example, one of my classes has 50% to earned experience, but in return the stats add to 140 at level 1 and 560 at level 99 to compensate (so as to make it a class that is really powerful compared to the others, but slow to get there).
 

Harmill

Veteran
Veteran
Joined
May 3, 2012
Messages
295
Reaction score
130
First Language
English
Primarily Uses
Having every "stat spread" total to the same amount implies that all stats are of equal value. However, unless you use an ATB or CTB system, AGL is too much just a binary value. You are either faster or slower than your enemies. Your enemies would need to be balanced so that there are some enemies that only the "fastest AGL growth class" could beat, otherwise, his extra AGL stats are useless when compared to the "second fastest AGL growth class". Or you have to start adding extra factors to AGL that are not typical of the stat, such as creating skills that scale with AGL instead of STR or MAG.

Then there is the LUCK stat. Just how noticeable is a high LUCK stat in any RPG? I'm sure we can customize it to do whatever we want and MAKE it noticeable, but if I see a class that favours LUCK, it's going to feel like it just has a weaker stat spread, even if they all add up to the same total.

Personally, I don't use formulas or anything of the sort for my own stat balancing. I'm super obsessive over recording my character's stats in the RPGs I play, and so when making my own RPGs, I just go berserk testing different values until I find something that works. Formulas are too rigid for my taste.
 

Tai_MT

Veteran
Veteran
Joined
May 1, 2013
Messages
5,423
Reaction score
4,737
First Language
English
Primarily Uses
RMMV
Okay, I read a few posts here and I'm going to have to take issue with a few of them.  Not because I think those people are wrong...  But, because I don't think they're addressing your real question and you're not quite clear on how to ask your real question.

Your question boils down into two questions.

"How do I determine proper progression in my game via stats?" and "What stat spreads are important to the game I'm making?"

The problem is that those are really only things you can answer.  Let me clarify.

Progression will be determined by planned length of game, planned stats (are you going to use higher or lower values?), and damage formulas.  If your game is only a few hours long, progression won't be that useful, especially in terms of stats.  If your game clocks in at 70 hours, then you're going to want to keep progression at least steady over those hours to ensure players don't suddenly start having issues with battles.  If you plan on using higher level stats, then you can use more complex formulas for damage and points in each stat will actually mean a lot.  If you plan on using lower level stats, then your damage formulas will be fairly simplistic and regular stat gain may not mean very much.  If you want to use multiplicative damage formulas, then you'll have to consider slower progression so that these formulas do not unbalance your game.  If you want to use additive formulas, then you'll have to consider a fairly quick progression as your stats likely won't mean a whole lot in terms of damage variation or growth.

As for your stat spreads and which are important...  Well, that's largely based upon what kind of game you're designing.  How valuable your stats are is determined by the formulas you'll be using to determine damage.  Many designers think of the "Luck" stat as fairly useless.  Well, if you only consider what it does by default to the VX Ace engine... then yeah, it's not that useful.  It determines the chance a state will be inflicted (if I remember correctly).  However, in the hands of someone with more imagination, it can be used as an attack stat or defense stat.  It can be renamed and used in different formulas to be useful in its own way.  Same with the AGL stat.  It determines turn order.  Okay, so what?  Well, if you make as many skills that run off of it that run off of Magic or Attack skills, then it's suddenly fairly valuable as a stat.

Before you get into all the questions about "How long should it take to get stat progression?" and "how many xp points should I assign per level/monster?" you need to decide on what kind of game you want first.  You cannot delve into your stats and battles unless you know exactly what you're aiming for in a game first.

Let me toot my own horn here for a minute and give you examples of what I'm doing.  In my RPG, it takes a fairly long time to level up because I didn't really want players "grinding levels".  I also disconnected stat gains from level ups to ensure that "grinding levels" wasn't the way your characters became more powerful.  Instead, I use "level ups" as a means to get my players to explore the map each time they gain a level just to see what's changed.  Gaining a level opens up shortcuts or sometimes adds new things to locations.  Perhaps a ladder appears to let you take a shortcut up the mountain.  Maybe some debris vanishes to let you through on a road.  I use level ups as a means to get players to explore the map again and again.  To tread the same ground, but still feel the delight of "exploring a new place".  Likewise, I linked stat gains to only items.  Items I only give out upon the completion of Quests.  I wanted players to engage in the narrative and not care so much about grinding for xp or money.  I wanted them to get engaged in the characters, the places, the lore, the backstories, and the events going on around them.  These items that give stats can be used on anyone so that a player can distribute stats however they want amongst their party.  Likewise, I keep stats fairly low in order to keep the nightmare of battle balancing under control.  20 HP to start with doesn't seem like much, but if it still takes 10 hits to kill you from basic enemies, it's a decent amount of HP.  20 MP doesn't seem like much until you realize that most basic spells are cast for 1-3 MP.  10 Defense doesn't seem like much, but if the enemy only has 10 Attack to throw at you, you don't take any damage at all.  I also made my "small stats" matter by changing many of the formulas.  Attack/Defense pairing, Magic/Magic Defense pairing, and Speed/Luck pairing.  I gave each character a specialization in one of those three pairings so that they'd have an initial stat that would actually matter to the player.  Enemies with low Luck attributes (or player characters) would take a lot of damage to skills that used Speed in their formula.  Equipment in the game also gives the same fairly small stat bonuses that using those items gives.  The starting equipment generally only gives one or two points.  Some of the more expensive starting gear can give as many as ten.

It's about knowing what kind of game you want to make and how to set about doing that with your stats.  What do you want players to be doing?  Do you want them to know how the formulas for your skills work?  Do you want to lessen the workload on yourself?  Do you want to make intricate formulas to keep players guessing just how the heck their damage is calculated based on their stats?  You need to decide on what kind of game you're making to decide how you want stat progression and balancing to work.
 
  • Like
Reactions: TMS

TMS

Guy with Mask
Veteran
Joined
Apr 17, 2012
Messages
300
Reaction score
57
First Language
English
Primarily Uses
N/A
You bring up some good points. Part of my problem is that I don't know how long the game will be. Sure, I have a general idea - a list of locations and objectives, and a desire to keep it relatively short since it's my first game - but it's hard to guess with any accuracy how many "level ups" we're talking about. I'm beginning to think that, as Wavelength said, issues of balance don't matter much when development of the game begins. Of course, the problem then becomes knowing when to start shifting the focus more towards the balancing/difficulty aspects of it.

When it comes to what the stats affect, I won't be doing anything creative, since it's the part of game design that interests me least. They make enough sense as they are, so there's no need to create extra work for myself.

As for what kind of game I want to make, it's something that can't be put into a few words. But I certainly don't want players to spend the game preoccupied with numbers and formulas. That's not really up to me, though. I couldn't enjoy that way of playing a game, but I know there are players out there who love to figure out formulas and take advantage of them. What I want  is for people to enjoy exploring the game's world, hopefully feeling a sense of wonder (and occasionally dread) and taking a liking to the characters in the process. Your idea of connecting map changes to leveling up sounds interesting, though I won't be using it myself.
 

Tai_MT

Veteran
Veteran
Joined
May 1, 2013
Messages
5,423
Reaction score
4,737
First Language
English
Primarily Uses
RMMV
Well, here's a free tip.  Design a game you would want to play.  If you want to play it, the chances are fairly good someone else would want to play it as well.  If you want players to enjoy exploring, tailor your game experience to that.  Combat won't be as important to you and you wouldn't need to add in things like monsters or tinker with stats as much.  You could probably even save that for the end of the development process and tweak it as you go along.  It's simple to tweak a stat and then test it in gameplay.  It takes a matter of seconds.

Maybe you should worry about stats and battles after you've gotten all the stuff you do care about out of the way?  Add in monsters to locations later.  Get a generalized idea of which monsters you want where and then remember that when you come back to it and tinker with their stats/xp/gold as you need to.
 
  • Like
Reactions: TMS

TMS

Guy with Mask
Veteran
Joined
Apr 17, 2012
Messages
300
Reaction score
57
First Language
English
Primarily Uses
N/A
I'll give that a try. Thanks for the advice.
 

bgillisp

Global Moderators
Global Mod
Joined
Jul 2, 2014
Messages
13,139
Reaction score
13,640
First Language
English
Primarily Uses
RMVXA
@Harmill - True, and I made sure to implement that in my game. I actually got rid of luck and replaced it with DEX, and used Yanfly's script to make it affect to hit, and AGL affects EVA. So now AGL and DEX have importance, low DEX means you miss more, and low AGL means you are hit more.

Also AGL affecting speed is part of my strategy. I have a couple super fast classes that are meant to always go first, but they have low other stats, so you have to figure out how to use them (one way is to have them toss a potion for emergency healing. The high AGL means they will be going first, so you will get that potion off before the enemy acts). I also have a couple low AGL characters meant to go last in battles, which can also be used in a similar fashion.

But I will say the best advice is from Tai_MT: Make what you want to play. Ultimately that's what I'm doing in my game. I figure if others like it enough to buy it when I release it, great. If not, then making this game was my hobby for the last x years or so.
 

Wavelength

Edge of Eternity
Global Mod
Joined
Jul 22, 2014
Messages
5,199
Reaction score
4,516
First Language
English
Primarily Uses
RMVXA
I'm beginning to think that, as Wavelength said, issues of balance don't matter much when development of the game begins. Of course, the problem then becomes knowing when to start shifting the focus more towards the balancing/difficulty aspects of it.
If it's not something that lights your passion, and it's not meant to be something that represents the core engagement of your game, then you can wait until all the content is built and you're almost ready to playtest the entire adventure in one shot before you worry about things like XP curves and stat balancing.

One other thing you asked is whether things will "work out" if you use the defaults in the database editor.  The answer is almost definitely not - because the database editor has no idea how often your party will be fighting, which monsters will appear in which area, which weapons you'll have at a given point in the game, etc.  But as long as you're smart about it (i.e. not using a Succubus with its default stats in your game's second dungeon), you can probably get away with using the default stuff as a baseline.  You just need to be extra-careful when you test your game to note the places and reasons that imbalances are occuring, and correct them accordingly.

If you feel you have no talent for that, though, you're better off finding someone who's passionate about it - and trust me, there are a good number of us out there. ;)   Build the stuff you enjoy building, and as you get closer to being able to call it a genuine game, find someone who's good at building up the RPG mechanics and have them help you out!
 

Users Who Are Viewing This Thread (Users: 0, Guests: 1)

Latest Threads

Latest Posts

Latest Profile Posts

*when the One Map Challenge is the only time you can finish a game*
Drifty wanted to make sure the default MV spark animation lived on, so he made it in Effekseer for the contest :guffaw::guffaw::guffaw:
Thanks to a post in a forum thread we can better create MV/MZ plugins in Haxe.
Latest video of my adventures in Kingdom Hearts 3~
Imagine every unfunny joke gets a laugh track, and it doesn't top until you genuinely laugh at that unfunny joke.

Forum statistics

Threads
99,565
Messages
966,762
Members
131,199
Latest member
TheFaz
Top