# General system design questions

I haven’t been a part of the RPG scene for too long, but one major part of how I managed to get “into” it has been the 2d20 system, notably through Star Trek Adventures. And now that I’m also working on implementling STA into Foundry VTT, I get much closer to the intricacies of the system.

There are a number of things I want to homebrew using the system, notably Stargate, as neither of the official versions that exist at this point are satisfying, but also a couple of my own original ideas. Since there is no generic version of the system, nor a public SRD to dive into, I have a couple of questions about the core design and hope to find some answers.

I own three games: Star Trek, John Carter, and Dishonored. From what I know about those as well as Conan, and what I’ve read about the system in general, each game can use the 2d20 system in different ways.

Since we roll 2 or more d20s and check if each rolls equal to or below a target number comprised from, usually, two or more stats, the range is limited from 1 to 20. In Star Trek, PCs will always have the target number between 8 and 17 (Attribute 7-12, Discipline 1-5), while in Dishonored they are between 8 and 16 (Skills and Styles are 4-8). John Carter however has a theoretical range of 8 to 24 (each attribute can be 4-12), but while not having read the book in full, I would imagine no two attributes will ever be that high. However, I also know that in Star Trek at least, NPCs and ships can have stats higher than that, creating target numbers up to 19 AFAIK.

What exactly is the design “philosophy” behind this mechanic? While rolling 20s cause Complications, tests never automatically fail. So if the target number is 19, the odds of failing are significantly lower, unless the difficulty is high enough. Going 20 or higher of course makes failure impossible. I would like to understand the logic behind this, as I find dice probabilities with multiple dice difficult to understand but I would like to in order to better understand the system.

Other, much shorter, questions are:

1. What are some considerations to determine what statistics to use to for the target number and critical success, and how many “sets” to use? (e.g. Attributes and Disciplines, Skills and Styles, and Focuses).
2. How far should I go in terms of the other stats and how to measure them? (e.g. Resistance, Stress, Determination, Void Points, etc.).
3. How to decide upon the “optimal” target number range, and when to have NPCs go outside that range?
4. Finally, are there any pitfalls to look out for in terms of balancing, like Talents or similar features that can provide boosts or extra dice, as well as “Momentum/Threat”?

I know this is a lot , but this is definitely my most favorite system and I would like to understand as much as I can. For context, I’m an amateur game designer and studied game design and (video game) development.

3 Likes

I am not that versed in making my own Foundry VTT adaptations, but I want to run my next Infinity campaign on this platform. Currently I’m struggling with a lot of things for such an adaptation. So having a look at a running STA adaptation would be helpful - I hope.

2 Likes

It will effectively be the “official” unofficial system for Foundry, if things go well. But until Modiphius has had a chance to confirm what can and can’t be included and give their go-ahead, it will remain private (and I’ve barely begun).

2 Likes

I’ll put together an answer on this, but I’m in the midst of a couple of different things right now, and I don’t want to rush my answer here.

5 Likes

Oh, please, there is absolutely no rush.

2 Likes

I prefer a GOOD and informative answer, to a quick one

4 Likes

Reading the Delta Quadrant book, I also found the Voth City Ship, which has Structure 18 and Departments that go from 1 to 4. So, to my understanding, if I roll for that ship using Structure and any Department at 2 or higher the test can never fail, though still cause Complications when rolling 20s of course.

And re-reading the rules, starships can have Systems from 0-12 and Departments from 0-5, comparable to characters. But the chapter about creating NPCs clearly outline that NPC Attributes should never go above 12 and Disciplines above 5. I can’t actually remember if I saw NPCs with stats higher than that, it’s mostly starships I come across. The core book also has no specific limits for NPC starships, only player ships.

Earlier games using the 2d20 System were a little less constrained when it came to ranges of target numbers, but in general, we like PC target numbers to sit in the 8-16 range; TN 8 gives a successes-per-die rate of 40% (so, you’ll beat difficulty 1 on two dice 64% of the time), and TN 16 gives a success-per-die of 80% (beat difficulty 1 on two dice 96% of the time). From a design perspective, the TN is a measure of efficiency - how effectively can a character turn a die into a success? Focus adds to that (creating a narrow range of ultra-efficiency where one die becomes two successes), but Focus ranges in anything other than JCoM never get very high.

The bottom end of the range can be adjusted lower if PCs are assumed to be less competent or need to start out weaker. The higher end of the range shouldn’t really go higher without a very good reason, as it can make characters too effective (turning dice into successes too easily). Still, even if you’re in a situation where you’ve got a TN of 20+ (which can happen in Infinity and Conan in some narrow cases), you still can’t really hit difficulties of 3+ without buying extra dice, so while such things stretch the game, they don’t entirely break it - you’re still reliant on focuses and on buying dice to get the higher difficulties. It does tax the Momentum/Threat economy somewhat if Momentum is too easy to cycle, but that’s a separate consideration at this point.

NPCs don’t need to be constrained as often - an NPC will roll less often than a PC (normally only in situations where they’re opposing a PC), but even then, extremely high TNs have diminishing returns - a TN of 23 is no different in practice to a TN of 21, so high attributes are supplemented by special rules such as Unnatural Attributes in MC3, Infinity, and Conan, which just grant extra successes. A high attribute may still have value if it’s used to derive other stats, such as a starship with 18 Structure using that score to determine Shields.

Ships in Trek are a little outside the normal case too, as they really only assist so their impact will be limited to a single die rather than entire skill tests. That gives us a little leeway to play outside the normal PC range of TNs - a shuttle’s scores are low enough that it may not really help a roll most of the time, as shuttles are mainly there to get you from point A to point B. The U.S.S. Defiant’s Weapons + Security TN is 18, because it’s the Defiant, so of course it’s really good for shooting things (balanced by the Defiant-class being a little less capable in other fields than another ship of the time, because it’s an ultraspecialised little gunboat).

I always aim for two sets. I’ve considered three sets, but the calculations get just fiddly enough to do in play that I’m not sure the granularity is worth the effort. Adding two relatively small numbers together in play is easy enough to ask players to do, and it’s just the right level of choice.

Earlier 2d20 games tended to do a fixed skill list - a long list of skills with fixed attributes for each. That’s probably easier for a new player because all the TNs are listed on the character sheet ready-made, but I find it less interesting, and I think there’s less depth there for players to get their teeth into. Pick two and combine creates flexibility, and I’ve found that it produces interesting combinations and helps players get into a particular mindset. I also generally like shorter skill lists from a personal standpoint - they make for neater, more compact presentation, more broadly-competent PCs, and allow the game to be more cleanly focused on what it needs to be about.

With Dishonored in particular, the Skills + Styles combination helped link to the “play it your way” notion that Arkane studios put in Dishonored originally. It also meant that every skill test could be a question - “what are you doing [skill] and how are you doing it [style]?” Conveniently, it also let us present skill tests in something closer to natural language - move quietly and tinker carefully are more natural and more evocative than “Agility + Stealth” or “Reason + Engineering”.

Resistance, I tend to define more through equipment and talents, and that does involve determining how much damage things are dealing, and should be dealing. Lots of fiddling around with spreadsheets. Below, I’ve copied my worksheet for figuring Challenge Dice odds, showing the percentage chance of scoring at least a particular number with a certain number of Challenge Dice, without and then with the Vicious quality (+1 to total per Effect rolled). The Anydice links should help with that too.

https://anydice.com/program/1a3a2 % to at least…
Challenge Dice 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Average Maximum Effects
1 67% 17% 0% 0% 0% 0% 0% 0% 0% 0% 0.833333333 2 0
2 89% 56% 19% 3% 0% 0% 0% 0% 0% 0% 1.666666667 4 1
3 96% 80% 49% 20% 5% 0% 0% 0% 0% 0% 2.5 6 1
4 99% 91% 72% 44% 20% 6% 1% 0% 0% 0% 3.333333333 8 1
5 100% 97% 86% 66% 41% 19% 7% 1% 0% 0% 4.166666667 10 2
6 100% 99% 94% 81% 61% 38% 19% 7% 2% 0% 0% 0% 5 12 2
7 100% 99% 97% 90% 77% 57% 35% 18% 7% 2% 1% 0% 0% 0% 5.833333333 14 2
8 100% 100% 99% 95% 87% 72% 53% 33% 19% 7% 2% 1% 0% 0% 0% 0% 6.666666667 16 3
9 100% 100% 99% 98% 93% 83% 68% 49% 31% 17% 7% 3% 1% 0% 0% 0% 0% 0% 7.5 18 3
10 100% 100% 100% 99% 96% 91% 80% 65% 46% 29% 16% 7% 3% 1% 0% 0% 0% 0% 0% 0% 8.333333333 20 3
11 100% 100% 100% 100% 98% 95% 88% 77% 61% 44% 28% 15% 7% 3% 1% 0% 0% 0% 0% 0% 9.166666667 22 4
12 100% 100% 100% 100% 99% 97% 93% 85% 73% 58% 41% 26% 15% 7% 3% 1% 0% 0% 0% 0% 10 24 4
13 100% 100% 100% 100% 100% 99% 96% 91% 83% 70% 55% 39% 25% 14% 7% 3% 1% 0% 0% 0% 10.83333333 26 4
14 100% 100% 100% 100% 100% 99% 98% 95% 89% 80% 67% 52% 37% 24% 14% 7% 3% 1% 0% 0% 11.66666667 28 5
15 100% 100% 100% 100% 100% 100% 99% 97% 94% 87% 77% 64% 50% 35% 23% 13% 7% 3% 1% 0% 12.5 30 5
https://anydice.com/program/1a36c
…with Vicious 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Average Maximum Effects
1 67% 50% 0% 0% 0% 0% 0% 0% 0% 0% 1.166666667 2 0
2 89% 78% 42% 25% 0% 0% 0% 0% 0% 0% 2.333333333 4 1
3 96% 91% 71% 54% 25% 13% 0% 0% 0% 0% 3.5 6 1
4 99% 96% 87% 75% 53% 35% 15% 6% 0% 0% 4.666666667 8 1
5 100% 99% 94% 88% 74% 58% 37% 22% 8% 3% 5.833333333 10 2
6 100% 99% 98% 94% 86% 76% 59% 43% 25% 13% 5% 2% 7 12 2
7 100% 100% 99% 97% 93% 87% 75% 62% 45% 30% 16% 8% 3% 1% 8.166666667 14 2
8 100% 100% 100% 99% 97% 93% 86% 77% 63% 48% 33% 20% 10% 5% 1% 0% 9.333333333 16 3
9 100% 100% 100% 100% 99% 97% 93% 87% 77% 65% 51% 37% 23% 14% 6% 3% 1% 0% 10.5 18 3
10 100% 100% 100% 100% 99% 98% 96% 93% 86% 78% 66% 53% 39% 27% 16% 9% 4% 2% 0% 0% 11.66666667 20 3
11 100% 100% 100% 100% 100% 99% 98% 96% 92% 87% 78% 68% 55% 42% 29% 19% 11% 6% 2% 1% 12.83333333 22 4
12 100% 100% 100% 100% 100% 100% 99% 98% 96% 92% 87% 79% 69% 57% 44% 32% 21% 13% 7% 3% 14 24 4
13 100% 100% 100% 100% 100% 100% 100% 99% 98% 96% 92% 87% 79% 70% 59% 47% 35% 24% 15% 9% 15.16666667 26 4
14 100% 100% 100% 100% 100% 100% 100% 100% 99% 98% 96% 92% 87% 80% 71% 60% 49% 37% 26% 18% 16.33333333 28 5
15 100% 100% 100% 100% 100% 100% 100% 100% 99% 99% 98% 96% 92% 87% 81% 72% 62% 51% 39% 29% 17.5 30 5

Stress, I try to base on one or more of the scores from the character’s stats and skills or equivalent, as that normally feels right.

Determination and Void Points are essentially the same mechanic with some game-specific tweaks, but I don’t tie those to other stats - rather, the limits for those are fairly constrained to encourage players to use them rather than hoard them, and accruing them is tied more to the game being played than anything else.

As I noted above, 8-16 is my baseline range, I’ll nudge the upper end up a little (17, maybe 18) if characters are meant to be especially competent, but high-skill can be covered in other ways, such as talents and focus range, so there’s not too much need to play there. I’ll nudge the low range down if characters are meant to be lack skill in some areas or start out with more room to grow. I personally don’t like to go below TN 6 here, but sometimes it can’t be helped depending on the game’s brief. Fallout, I think, goes as low as 4 in some edge cases.

The more broadly you apply Focus, the more skilful characters will feel. Extra dice for free should only be available in very specific cases - the easier it is to get a free extra die, the less a player has to engage with the Momentum/Threat economy - and re-rolls of d20s should be limited to a single die unless they apply only in specific niche cases or come with a significant cost. A lot of that comes with experience, and different games demand different types of Talent, though I generally prefer ones that are evocative or encourage interesting play more than flat power boosts. Talents that give PCs interesting things to do at the cost of adding to Threat are fun to design, because they tempt players to play around with Threat when they might otherwise be reluctant, and can help with a character who is narratively brash and reckless (while Momentum-powered talents give a feeling of collaboration or careful planning).

Any more specific questions?

4 Likes

Wow, thanks so much! This is more than I hoped I’d get, and it’s extremely informative. At this point I’ve got enough inspiration to give it try myself, so I have no more questions right now. I feel very inspired to get started and see how it goes first.

Again, thank you! I’m very grateful for this.

I am wondering something else though. This is probably a question that can’t be given a straight answer from Modiphius staff as it could include “insider” information, so this is more “in general” or aimed at others versed in game design that read this.

I come mainly from a digital game development side, so I am well versed in the techniques of using the Game Design Document to describe the design of a video game before writing a line of code or building levels. However, how would a tabletop RPG be designed? What’s the process? Is it all going in a master document that forms the basis of the final rulebook, or do designers write a sort of design document first, like a reference document, to design the mechanics/rules for internal discussions and playtesting, before putting it all in the actual rulebook?

What I would do is first outline the idea of what I want to achieve (and in this case how I would adapt the 2d20 system to fit my particular game idea), then outline the mechanics to achieve that. Finally, I would just put everything straight in a manuscript for the actual rulebook. I just feel it’s a big step going from a design outline/brief to a full manuscript.

More or less, we’ll have early meetings and decisions made as to what we actually want the game to achieve, driven both by the licensor’s expectations and by whomever is in charge of the product creatively. For 2d20, we have an SRD (System Resource Document) that I’ve designed (and which gets periodically updated) for internal use as a core rules reference for all 2d20 System games, but each game differs enough that we can’t just use that text as-is in each time. I’ll typically put together bullet point notes as to what I want the game to do mechanically (based on my own take on the IP and direction from those earlier discussions), which we’ll throw back and forth a bit before those notes get fleshed out into either a beta draft or a first manuscript draft (broken down by chapter, because it keeps individual documents from getting unwieldy), and then that gets refined into the final work (examples of play added, tidying up presentation and language, tweaks based on feedback, editing, etc.)

That’s my process, at least.

4 Likes

Generally speaking, it is a big step. Whether it’s mechanics or prose, going from an outline or design document of a few thousand words to a full manuscript that could be 100,000 words or more, depending on the product, is a major step in the process.

2 Likes

there are many that have been asking for a generic 2D20 system for their own home brewing of systems

2 Likes

It’s nice to hear I’m not far off with my assumption. Seems I was already on the right track in the first place. Thanks again!

1 Like

Good information thanks for sharing
vmware

Amen. I’d love to see one.

It’s nice to read that someone tries to bring STA to foundry vtt.