Homebrew System Pitfalls

Sup/tg/ I'm about a year into the development of a homebrew system, its not revolutionary or genius, I just wanted to give it a go to tailor a system to my GMing style, what pitfalls should I look out for when designing a system?

>what pitfalls should I look out for when designing a system?
making a homebrew

>what pitfalls should I look out for when designing a system?
Could you be a bit less vague?

I expected this

Im thinking of anything obvious I should avoid, other than forcefully shoving it down peoples throats

We're going to need a little more information than that.

What style, what genre and how much simulation-to-fun ratio are you looking for?

I want as much depth for as little rolls per turn as possible, with a pretty heavy realism focus

It's hard to say much when I don't know anything about what you are trying to make in any way, but I guess

If you are designing classes, you want them to be broad enough that you don't have like 20 classes that either aren't that different, or where some are great and some are just shitty and take up space. You also need each class to have a purpose. It does X, and X is useful, and something other classes don't do. Otherwise, it's a shitty class.

You should have a unified die mechanic that works for pretty much everything. Try and pick something that has little or no addition/subtraction, and avoid escalating die pools where you end up rolling 15 dice or some crap.

You want to do your work on the math in terms of minimal, average, and maximum damage output compared to health, so that you can properly judge where the players are in their advancement and how much damage they should each be doing within a range, and how much health they should have in a range. This is important so that you can avoid one guy doing 50 damage a hit while everyone else does 5, and for knowing at what levels (or points or whatever) characters will be doing X average damage, and can take X average damage, for quick balancing encounters.

You should try and avoid unnecessary numbers and distinctions. The practice of separating Attributes and Skills, for example, is very common but largely unnecessary. DnD style counting, where 12=+1 and 13=+1 and 14=+2, is redundant. The actual numbers are not needed, you can just have the +/- values. You should consider this sort of thing and decide how to do it most efficiently.

I would avoid simulationist design. Don't try to make things that emulate reality. Come up with a way the mechanics work that is simple and consistent, and then figure out how to match what you want to that system. Basically, avoid doing what GURPS does, giving you different rules for every thing in an effort to be realistic.

This is good, I'm going for a roll under mechanic on a d20, I like how the roll under works on like CoC and 40krpg but the big numbers confuse some of my players so a d20 would be more manageable, I'm also going for a classless system where anyone can purchase whatever, I'm all about player agency over balance personally

Pitfalls to avoid:

> Throwing in far too much because you have no idea what you're aiming for.
> Trying to do something another system already does well.
Write up a short list of what you want your system to do that other systems do not do. Be specific. Have a specific setting in mind*, with specific themes you want campaigns run in your system to have.

Things you want your system to not do are also worth including.

Then give us that list so we can point out any systems you've missed.

*Even if you plan to replace/remove the setting later. Having a specific setting in mind will help focus you on what rules are important and what aren't.

>Lack of playtesting
You should have started playtesting your system already. The best playtesting is when you give someone else a copy of the rules and they GM a game. Make sure they agree to run the rules as written, no matter how stupid a rule is. No houserules.

One thing you are testing for is how well you have written the rules. If there is a disagreement over a rule or if the group interprets the rules wrong, then that is a problem with how you wrote the rules. Take a note to update the rule in the next revision.

Do not speak up to correct anyone during the session. Do not answer if anyone asks you to clarify a rule. Doing either of these risks the group guessing your intent and going with that, lessening their ability to show you when the RAW is badly written.

The only time you speak up is if they are using an old version of a rule you have updated.

The players will want to have clear ideas for what is effective and deliberate in your game mechanics, they won't like picking some option only to discover later that it's useless or never used in the game.

>Basically, avoid doing what GURPS does, giving you different rules for every thing in an effort to be realistic.
It doesn't. It gives you a lot of rules to suit the style of your campaign. That's why you have things like Wildcard Skills, Destiny Points, Bleeding, Accumulated Wounds, Simplified Range Bands, Super-Strength, Crippling Injuries, Bulletproof Nudity, Damage to Armor, Ham Clause, Harsh Realism for Unarmed Fighters, Unarmed Etiquette, Flesh Wounds, Cinematic Explosions, Hit Locations, Cinematic Knockback, and so many other rules that fit a wide variety of games. Almost every rule that is "realistic" has an unrealistic alternative.


You should give GURPS Lite a read, if only to steal material from for your homebrew. GURPS is extremely comprehensive in traits (Lite less so) that can make up a player character, and you can cut down to size.

Homebrew systems belong to the trashcan.

Study. Research. Read. Find ten games that do the thing you want to do. Run them all. Then run another ten games that do completely different things. Then, and only then, you'll maybe have an inkling of what a game needs.

I hate the word "gurps", it sounds like a baby vomiting I'd never play it just because of that.

Don't expect people to regularly multiply two numbers above ten for any reason. For some reason I've seen a lot of homebrews where people are expected to multiply two d20s by each other or whatever. Kills the pacing.

Ive been a foreverDM for 10 years, Ran AD&D, 3.5, CoC, WoD, 40kRPG, Pathfinder, and none of them quite hit the spot, but I think im fairly adept at identifying why they dont, to myself at least

Noted, cheers mate, I dont actually have any multiplication in mechanics, just basic addition and subtraction

I know you guys who like GURPS always say that, but I've played in a good number of GURPS games and the biggest problem with all of them has been that something happens, and the GM says, "oh, that's in one of these books" and starts to look for some chart or rule or specific mechanic for how cocaine works or a list of the damages done by vehicles or how much a horse would have cost in 1805 or how far you have to fall to reach terminal velocity or what the percentange likelihood of someone with narcolepsy falling asleep in a round is or something. And that takes up huge amounts of time.

Yes it is true that you don't have to use all the rules. I have NEVER found a person who actually ran GURPS who followed that advice, though. The people who don't feel the need to follow all the rules as written are not using GURPS to begin with.

>the GM says, "oh, that's in one of these books" and starts to look
They're bad GMs. You're never supposed to look at the book during play, for any system. Anything not on your GM screen is not worth looking up.

So you ran basically six versions of the same game, only with different dice from time to time. That doesn't help your case.

> Trying to do something another system already does well

Don't use good ideas is somebody else does them.
Jim wipes his as with toilet paper. I shouldn't.
Disregarded.

So is gurus just a bunch of overpriced books + whatever your DM pulls from his ass + freeform? Sounds bad.

Is English your first language? I know we have a lot of foreigners on this board, so I need to know if I have to talk to you like a child. If you carefully read the post you linked, you'll notice that I was talking about every system, meaning that your statement applies to your favorite system as well. Do you understand that?

Crafting Systems
Action Point Systems
Rolled Movement

Avoid all these things

>Action Point Systems
>Rolled Movement
explain

Action Point example: Nechronica
>Combat takes place on counts, and you act when your AP equals the count
>Every character has an AP total from 6 to 15+
>Every action a character takes costs AP
>Some actions may be taken when it is not your count as Rapid actions
It's a great, highly tactical system that works beautifully for zombie and mech combat.

Rolled Movement example: Savage Worlds...?
When you run in Savage Worlds, you roll a die and add that much to your movement value. But your default movement value is static. I've got no clue if any system has you always roll your movement value every time you want to move.

The most frequent ones I see around:
>avoiding a tool because you don't like another game that uses it (the d20 for example)
>using 3d6 because of the bell curve distribution when your system only checks pass/fail
>underestimating the availability of other dice than the d6 for your target audience
>mistaking granularity for complexity (20 self evidently named skills using the base resolution vs 5 perks with unique and conditional rules)
>going so rules light that you might as well forgo the system for rolling the die of choice and eyeball according to result (the "uncanny valley" of rpg systems)

Implying I have to prove a case to you or I will lose sleep tonight
This is useful thanks,could you explain what you mean with the bell curve of the 3d6? I was thinking about using those other than a d20

>could you explain what you mean with the bell curve of the 3d6?
The bell-curve is the distribution pattern of results when you sum up the three d6. It makes a bell shape when shown as a diagram.
This distribution pattern is regarded as realistic because it appears in many statistics taken from nature.
But when you just binary check for pass or fail you only have two outcomes and can't utilize this realistic distribution of results because you simply don't have the "resolution" to register it.

The way I have designed the core mechanic is that the amount that you succeed or fail by matters a huge deal, you reckon I would be better off going with a 3d6?

Matter of taste really.
It is important to note that with 3d6 you get more mileage out of skill or stat investments when you are in the middle of the scale opposed to the extreme ends. And most results will be medium rolls when modifiers are neutral. This brings dependable consistency but is also predictable. And you give up the easy regular increments for calculating risks.

Further I personally don't like when the fringe results are below 1% probability. Comes up to rarely but you still have to account for those results. This means the most spectacular result are seen rarely. 0,5% is once every 200 rolls. When you only roll the most important things, when the stakes matter this takes a while.

This is false. An important feature of a bell curve is that it makes a +1 bonus nonlinear. On a single roll, in a vacuum, it doesn't matter, and isn't different from, say, rolling a 1d100 and checking to see if it's under 5, but the nonlinearity of bonuses and penalties is EXTREMELY important. The only way to model it with a d20 would be a lookup table, and that is, frankly, really stupid.

I basically criticized this lack of difference. As the benefit of having a bell curve isn't utilized when regarding binary results. The realistic bell curve is often stated as the main argument for 3d6. When you don't use that benefit you still have to account for different side consequences of you choice regardless.
The nonlinear bonus you mentioned. (and I did here )
The extra addition you have to do.
In a vacuum it also influences which difficulties you can model, as you have to adhere to the steps of the die/dice results. Can't do a 62.5% task on a d20, can't do a 70% on 3d6.

i fail to see the problem with either mechanic.

Not knowing how far you can move in advance is not fun.

AP systems tend to slow the game down, in my experience, as players try to calculate optimized AP costs, ensuring they can most effectively use all their AP and not leaving anything left over afterwards.

There isn't a problem with either mechanic when done well. Homebrewers are not going to be able to do them (crafting and action points, specifically) well. It's going to take a lot of playtesting and revising and chopping up and gutting. I'm also not a fan of randomized movement values, like , but there's no real issue with AP costs if the costs are static. Nechronica does allow you to lower/raise them for future use, but IME the combat slows down mostly due to timing resolution and a bit due to tactical discussion.

>Rolled Movement example: Savage Worlds...?
>When you run in Savage Worlds, you roll a die and add that much to your movement value. But your default movement value is static. I've got no clue if any system has you always roll your movement value every time you want to move.
I can't think of one either. It sounds like one of those FATAL "Rolling dice gives us an even bigger boner than gang rape and we didn't think it through beyond that" deals. Also, it's worth noting that even though you don't roll for every movement, a lot of SW groups still ignore rolled running movement. It really isn't a great mechanic in practice.

It's more popular in tabletop games.

>Not knowing how far you can move in advance is not fun.
well, you obviously should have a good estimate. but let's say you could run 20+1d6 meters per turn, that sounds pretty cool. because then there is a risk in whether you can reach the orc with yoru charge or not. what really sucks is if you can calculate it to the last meter as if it was a tactical boardgame.

That's honestly really annoying. It also means that when running as a group people fall ahead and behind constantly, and it's impossible to maintain a formation with optimal speeds, leading to disorganized groups.
It would be fine if you have some kind of sprint that takes extra, but something like that for default movement is incredibly annoying.

>what pitfalls should I look out for when designing
Don't spend so much time on the lore. You're writing for a game not a novel, so some depth is good but going full LoTR is just not necessary.

Give more care and attention to the aspects or areas that you're personally less interested in. If you're in general a bigger fan of magic-using characters, spend twice as long developing non-magical options as you do on magical ones. Force yourself to care--don't just check the boxes and be done with it.

Interesting advice. It's counter-intuitive but very logical.

>You're never supposed to look at the book during play, for any system.
Says who? Are you supposed to have every rule memorized before play?

>Anything not on your GM screen is not worth looking up

it's a homebrew, where the fuck is OP going to buy a GM screen?

Run playtests often, it's by far the fastest way to figure out what does and doesn't work.

Thanks for all the replies, I appreciate it

>I'm about a year into the development of a homebrew system
Then you're doing it very wrong.
No system should take more than a week or two to develop.
A rules light system can be made in an afternoon.

I do have a full time job and a family also there mate, aswell as playtesting and tweaking it all the time, sure I thought of the core mechanic in an afternoon but you are probably doing it wrong if you call it done after a couple of weeks