What is the role of magic in your world? 

Most tabletop games are really designed to be played in high magic settings. There are plenty of reasons to run a high magic world where common enchanters solve common problems with their overflow of magics. There are plenty of reasons to run a gritty world where magic is rare, never the go to solution, and a spell going off is enough power to cause awe and fear in equal measures. 

This may be the most important question for people engaged in world building or writing RPG mechanics. If you want magic to be rare but powerful and awe inspiring, then having 4 of 5 party members being able to cast unlimited cantrips tells the opposite story of your world. This is not a minor world building issue, but it is the most common mismatch I see in games. 

That said, if magic is widely available, then your world would not be like the European middle ages. Those magics should change the world dramatically and, once again, having those magics available but not recognized by or folded into the world building is story breaking. It is using mechanics to tell one story while the DM’s narration tells an entirely different story. If prestidigitation is everywhere, then you don’t need public baths. If healing potions are in every shop, why do towns have doctors and are people still going to churches to pray for healing? 

This thread is core to your world building. It determines if every town has an enchanter or if only the King can afford a court wizard, and a dubious court wizard at that. Is your world’s magic a larger interconnected system or a mix of odd abilities cobbled together by mad sorcerers playing with powers they don’t understand. The former needs a library or a school and those would have heavy political influence or be controlled by those who do. The latter needs a scattering of odd abilities and nebulous power sources. Each approach requires different histories and areas in your world. 

How powerful do you want it to be in combat? In non-combat challenges? In support? 

Regardless of whether this is an author challenging their main character or a DM challenging their players, part of what you have to determine is what kind of questions does magic contain the answers for. A story without a challenge is a crappy story. So if magic is always available and always solves everything without a cost then there is really no story. Magic can destroy your story… so be careful. Don’t build a system that is so cool that it makes your story or world suck. 

There are two approaches to solve for this: 

• Limit the amount spells that can be used, either by cost or limit charges 
• Limit the power of spells, so they help to solve more that entirely solve the problem 

While this pair of approaches is common, many games end up with both limitations breaking down due to power scaling. While there are many ways to control the amount of spells, the bigger problem is often limiting the power of spells, specifically utility spells. Trying to keep a limit on those spells so they are still useful compared to other spells while also not just entirely solving the problem at hand. 

As an example, older versions of D&D had really detailed and nuanced travel mechanics that forced critical game decisions… unless you got Good Berry, Leomund’s Tiny Hut, and Pass without a Trace at which point travel challenges lost all meaning. Resources no longer mattered and all strategic concerns are removed. As a DM, if you want travel to be an issue in your world then you have to get rid of those three spells. If you think travel is a pain in the butt and a waste of time, then give those spells away as much as you can. 

Now this is also where we run into the “whimsical high magics”. These are worlds like Harry Potter or many old Disney movies. The spells are very specific, very silly, and are generally played for a laugh or a smile. This level of flavor but not power is often really engaging. Let owls deliver the mail or brooms sweep the floors themselves. Let a princess sing so birds fly in and give her a french braid. This gives the world a magical feel without giving players god-like powers. 

Vancian Magic 

This was the original system used in D&D and variations of it are still core to most tabletop RPGs games. It is a highly formal system. The idea is that magic users have a limit on how many spells they can have memorized at a time. This limit is not just the total number of spells, but also takes into account the power of those spells. When they cast a spell, they “forget” that spell. The spell list limited both the quantity and quality of spells available. 

Spells in this system are traditionally discreet and consistent between casts and casting magic is reliable. You are able to take multiple copies of the same spell but this limits your versatility as a caster. 

Vancian Variant 0.1 – the 3rd edition Cleric 

So this was less a system and more a special rule. Clerics used the traditional vancian system but with one exception. Life Clerics could mark off any spell and instead cast Cure Wounds. Death Clerics could do the same thing but it would be Inflict Wounds. In a game without flexible casting, this was seen as very powerful. 

But there is no reason this can’t be done with other classes using a Vancian system. Basically it is just picking a first level spell that can scale with level and making that your “signature spell”. I see no reasons why a specialist wizard should not have the same ability. An illusionist gains silent image. Evocation gains burning hands or magic missile. All of this is super simple to house rule. 

Vancian Variant 1 – Mana 

The first variant method is so common it is actually in the 5e D&D DMG and thus an official variant, that is the spell point system or mana system. This is often written as a conversion option for magic users in the game as it exists today. Spell slots are converted to a pool of mana points. Spells are given a mana point cost. Total number of known spells is generally lower because in the original Vancian system players would often take multiple copies of their favorite spells. So if you have 10 spell slots, maybe learn about 6 spells. 

This system is often seen as a more flexible and less complex version of the original. The problem here is it does often encourage one trick pony casters. The guy who only casts fireball loves this system… but that does get old. If your spell list has any imbalances between spells, then this system often emphasizes that problem. 

Vancian Variant 2 – Spell Save System 

This is a fairly common variant in the OSR world. Instead of spells being guaranteed to be successfully casted each time, you roll for success or fail. On success the spell goes off and you still remember that spell. On a fail, no spell happens and you forget the spell. This changes the risk reward profile of magic, making it far more hit and miss. 

Like variant 1, this system can simply be applied on top of an existing system without the need to change everything around. It does add a level of randomness to your game, so it will be more unbalanced. The good news is it does produce both overpowered and underpowered outcomes naturally for a larger dungeon. The problem is in moment to moment encounters, this kind of casting can make them entirely trivial or the party is basically down a player. 

This goes back to the question “what is magic in your world” because if you want rare but powerful magic, then this does a great job of that as casters chain a series of big spells together. If you want rare magic, then just wait until 2 or 3 spells fizzle in a row. Magic will feel rare and unreliable. 

Vancian Variant 2.1 – Mis-spell System 

Similar to above but the negative consequence is not you forgetting the spell but miscasting it. So this is a tongue-firmly-in-cheek system. So for example, the spell “send message” on a fail would instead turn into “send massage” and the target would get a nice back rub instead of critical information. “Speak to Dead” could become “Speak to Dad” and trying to cover the floor in “Grease” could instead covers it in “Geese” and that flock starts to run around and pull Untitled Goose Game shenanigans for the rest of the session. This takes a bit of extra work from the DM and players but it does add humor to the table. For balance reasons, maybe you lose the original spell for an hour or so, but not the normal length of time.

I actually really like this system, but not for all worlds. This does not fit in a gritty setting, but is great for a one shot in a magic castle. I am also unsure how well the jokes will play out over a long campaign. 

I first saw this idea here: https://www.prismaticwasteland.com/blog/spelling-errors-a-magic-miscast-alternative and I kind of love it. So credit where credit is due. 

Vancian Variant 1.1 or 2.2 – Shadowrun’s Drain 

So in the kitchen sink of a cyberpunk game, Shadowrun uses a drain system. This game basically has two different health pools and some abilities, like magic, damage the second pool. If you are a fighter type, this second health pool is just extra health. If you are a magic or special skills user, this is your mana pool, AKA variant 1.1. Generally you can assign incoming damage from enemies to either health pool so you are always playing a mini risk vs. reward game negating incoming damage vs. having mana to spend on attacks next turn. 

Here the spell always works… but you roll to resist stun damage. So while the power of the spells is far more consistent, here the wildness of the system created by a failed roll is the cost you paid to cast the spell, thus making it a bit like a spell save system, AKA variant 2.2. This magic system fits into the narrative of that extreme setting as spells always work, something spells get a ton of extra damage, and something they render the caster unconscious afterwards because they dealt too much damage. It is all very metal and very Shadowrun. 

Vancian Variant 3 – what the heck is 5e’s system exactly? 

So 5e D&D doesn’t really use any of the 3 previous interchangeable systems. They have kind of a mix of Original Vancian and Variant 1. In effect, they have multiple tiny mana pools which overlap down and multiple tiny spell lists with overlap up. Casters are able to cast any 2nd level spell with any 2nd spell slot. That sounds simple. But the nuance here is that many 2nd level spells can also be cast with higher level spell slots for extra kick and higher level spells slots can be used to cast lower level spells, thus higher quality slots actually give you both more power and more versatility. 

This system gives you some of the flexibility of the Variant 1 with some of the rigidness of the original still in there. When you get into the details, it gets fairly complex very quickly and that gives you some strange decisions at various points in time. This system has a nice mix of spell balance and flexibility, but the strange overlaps do end up feeling a bit less like a flexible design and more like it’s wobbly enough that players can make it fit if they shove on it hard enough. 

—

Thank you all for reading, we will pick up next week with part 2 where we talk about alternatives to the Vancian variants and see what happens when we try to combine all the ideas into one system.

Blizzard’s new hardcore World of Warcraft servers have been really interesting to play on for the last 2 weeks. This server is classic wow but with the twist that characters can’t come back to life. Because it is classic wow, a game that is really well understood and as expected, a hundred videos and blog posts popped up about what to do and which route is best and what classes you should play. 

… and almost all of those have been proven to be completely wrong. Why? Why is advice about a game from 2004 wrong? Is nearly 20 years and millions of nerds not enough to understand it? 

So in this case the issue is clear: people misunderstand the problem. Almost all of that advice was based on how quick you can get to level 60 or what class was the best at level 60. Currently, the average character dies at level 13. This completely changes the dynamics of the game. 

Turns out the end game is not the hardest part of wow, it is surviving the early levels. And this makes a lot of sense. Characters at level 60 have a large and well rounded set of skills to deal with a range of situations. This is something that content makers know and spend a lot of time talking about those classes in the level 60 context as the end game is the part they play regularly. But those skills are slowly given out as you level up. It doesn’t matter if a hunter has the best survival skill in the game if they don’t get that skill until level 30. 79% of the hunters die before level 30 and thus will never see that skill. In fact, most hunters never see their pet… which is what the entire class is built around. 

The second great mistake is all the leveling guides. These are all built around efficiency of time to xp… but in hardcore they really need to focus on risk mitigation. Quests are designed to push you to the next zone or go fight a deadly monster… which is the least safe way to play. Picking a spot with the most escape routes and steadily but quickly killing enemies a few levels lower than you gives you the most money and most xp in the safest way. This is an MMORPG where the quests are bad. In many ways this is a return to the old everquest style of grinding out levels against random enemies. 

Now Blizzard made very little changes to the game itself. In fact, they were so lazy they left in the resurrection spells but just disabled their effect. The one thing that has been added in is an announcement system via add-ons. If someone dies in your guild, you learn who, at what level, and what killed them. Generally this is followed by lots of “F” in the chat to show respect. But this also really brings the guild together in an odd way. Everyone learns to avoid certain areas. Everyone shares the pain when a higher level player dies. In a social game, this does really change how the game feels. 

This design also forces you to ask yourself “what is powerful in this game?’. And in this case it is not the traditional dps meter (overall offense) or even best tank/healing abilities. It is the pairing of the panic button abilities that let you stay alive just 3 more seconds to run away and awareness by the player to know they need to run away. This forces much of the traditional wisdom out the door. 

For example, Paladins are slow levelers so everyone on classic goes Retribution spec to get just a bit more damage on a normal server. But by doing so, you skip past some of the Paladin’s best survival mechanisms. Turns out to play well in hardcore, you have to unlearn much of what you did originally because the priority is risk mitigation now. 

This also leads us to an important note. The best class in the game… maybe a profession. At low levels, the target dummy from engineering is an amazingly strong panic button ability. In fact it is better than what most classes have at this level. And because there are so many low level characters starting over again there are plenty of cheap materials on the auction house for you. At level 40, the gnomish invisibility device gives you 10 seconds of invisibility, which is one of the strongest survival options in the game. Alchemists get a similar potion that doesn’t require the profession, so keep an eye out. These two options are better than what most classes get for the whole of the game. 

So the following words are not my own. They are the introduction by Dave Arneson to his Blackmoor setting. The odd piece of history that exists between post-Chainmail and pre-D&D. It has some odd ideas but is also a brilliant piece of game design. I see large complex design issues being solved by combining these pieces together that I would normally not think to do.

From the first excursions into the dark depths of Blackmoor Castle’s Dungeon, it became apparent that these first hardy bands of adventurers would soon seek out new worlds to pillage. From the castle itself the small town of Blackmoor grew, then the surrounding countryside became filled with new holes to explore and beyond that talk was already spreading about visiting the Egg of Coot. Each of these steps entailed a great deal of work upon a naive Judge who felt that there was already more than enough trouble already available to satisfy any band of adventurers, a phrase no doubt heard rather frequently since then, in other areas. In general, a fairly loose procedure was set up for the establishment of each of these new areas, with a great deal of emphasis being placed on the players themselves setting up new Dungeons, with my original Dungeonmaster role evolving more into the job of co-ordinating the various operations that were underway at any given moment. At the height of my participation as chief co-ordinator, there were six Dungeons and over 100 detailed player characters to be kept track of at any one time.

Each area had to mesh with those areas that were around it, in so far as setting up the various monsters, etc, were concerned. It was also readily apparent, from previous experience running a “Conventional” Napoleonic Wargames campaign that some sort of Overall Background would have to be constructed to provide a framework within which the players could work. Thus the overall concept of the Evil Egg of Coot and the Great Kingdom was born. These two entities could prove to be the source of great events outside of the actual campaign, a source of new recruits and monsters, and give the stimulus, in the way of quests and adventures to give the players more of a motive than just looting the Dungeon. Also with such powerful and potentially aggressive neighbors, the locals decided that at least some taxes should be collected to provide for the common defense. This was a good plan but one which failed to take into account the drain placed on the local manpower pool by the repeated sorties into the Dungeon areas. So it was with the Dungeon of Blackmoor. It began with only the basic monsters in Chainmail and was only some six levels deep. Six levels was chosen since it allowed random placement with six-sided dice (no funny dice back then) (sic). So even in the Dungeon it became quickly apparent that there was a need for a greater variety of monsters, more definition even within the type of monsters, and certainly a deeper Dungeon.

So there were now different types of Dragons (by Size) and other new creatures, like Gargoyles, from standard mythology. AC was determined by description of the creature (Hide, scales, etc.) and how impervious it was in the accounts given in mythology about it. HD [read: hit dice or health] was determined pretty much on the size of the creature physically and, again, some regard for it’s mythical properties. For regular animals that were simply made larger, like Beetles, a standard text book provided interesting facts about the critters and all were given HD proportionate to their size, relative to other Beetles for instance. Insects were all given about the same AC with additions, again, for unique properties.

Character motivation was solved by stating that you did not get Experience Points until the money had been spent on your area of interest. This often led to additional adventures as players would order special cargos from off the board and then have to go and guard them so that the cargo would reach their lodging and then the player would get the Experience Points. More than one poor fellow found that his special motivators would literally run him ragged and get him killed before he got anything.

Combat was quite simple at first and then got progressively complicated with the addition of Hit Location, etc.. as the players first rolled for characteristics, the number of Hits a body could take ran from 0-100. As the player progressed, he did not receive additional Hit Points, but rather he became harder to Hit. All normal attacks were carried out in the usual fashion but the player revived a “Saving Throw” against any Hit that he received. Thus, although he might be “Hit” several times during a melee round, in actuality he might not take any damage at all. Only Fighters gained advantages [read: leveled up] in these melee Saving Throws. Clerics and Magicians progressed in their own areas, which might or might not modify their Saving Throws. And so it went, Hit Location so that even the mighty Smaug could fall to a single arrow in the right place (very unlikely), height differentiation, so that the little guys could run around more and the big ones could kill more, etc. Still these were guidelines, Hit Location was generally used only for the bigger critters, and only on a man to man level were all the options thrown in. This allowed play to progress quickly even if the poor monsters suffered more from it.

By the end of the Fourth year of continuous play Blackmoor covered hundreds of square miles, had a dozen castles, and three separate Judges as my own involvement decreased due to other commitments. But by then, it was more than able to run itself as a Fantasy campaign and keep more than a hundred people and a dozen Judges as busy then as they are today. Whether there will ever be the co-ordination of all the area Dungeons in the future as they were way back in “the Good Old Days” is unlikely, but already there are 20-30 people meeting every 4th Saturday to do Blackmoor and other Fantasy in related areas, so who can tell…after all, the keynote is that “Anything is Possible”, just that some are more likely than others.

– Dave Arneson

So lots of interesting ideas in there to unpack. First off, I think every DM can appreciate the world growing too fast and the players running off in the wrong direction… c’est la vie. It is also worth noting the Dave did not try to DM everything. He quickly let other people run things and made himself more of a WM (World Master) with others running dungeons as is needed.

Paragraph 4 has the best idea I have heard in a while and it is a details that is often missed. Yes, players want gold. But they gain xp by spending that gold. You don’t kill the dragon to gain a level and become powerful. Instead you kill the dragon, loots its’ horde, and then spend that horde on new armor and training… which makes you powerful in the story and gains you the level in the class.

Player are looking for a gold sink to gain levels with. Spending was not a second source of power to be controlled, it was the primary source of power so why not spend it in a way that helped your class. Players were trying to empty their pockets as fast as they could. Cleric would make scarifies to their Gods… because it was a gold sink, not for some boon. Yes, the bard going to the brothel is a legit gold sink to gain levels. And to add a level of story, the bard is not pious and the cleric stays pure, so they don’t get each other’s gold sinks. Both gold sinks have a set cost and thus growth is limited so players would spread it out over multiple sessions in town.

This inverts the tradition rules of player incentive. Fights are skilled in all weapons… so they can carry one of each, thus giving them more gold sinks. Players want to pay 5,000 gold for a +1 sword, but not for the sword. This also means gaining a title or rights to build a castle is a really big deal. This is why old school players would always build a keep and raise an army. The is how you gain the higher levels.

This also changes how the players interact with the world. The thief does not want to steal from vendors because that is lost xp. The town does not have unlimited supplies, so the players want to town to grow. Securing supply lines means more items means more gold can be spent means you gain a level. The party wants to save the town blacksmith when he is kidnapped by goblins… because if they don’t then they are limiting their own leveling. This solves for all kinds of in game problems.

The “saving throw” is really more an active dodge mechanic. I think it would slow the game down… but I like the feel. It allows for health to be kept lower so there is still a thrill. Maybe both roll at once, one to hit and the other to save to keep the game moving. Also limiting it to the fighters would work. This is also why armor originally went down instead of up. It was leaning into not an armor mechanic but a dodge mechanic.

If you start to look into the math underlying this systems, or at least as much of the math as we still know since Dave didn’t take notes and was always changing the rules, what you are looking at is:

1. Hero, Fighter: 4 health, coin flip to dodge, coin flip to hit enemies with a killing blow (mostly)
2. Hero, Cleric and Magic User: 2 health, coin flip to hit
   • Cleric get healing and some defense to stay alive
   • Magic User get invisibility and some movement spells to stay alive
3. Basic Enemy: 1 health, coin flip to hit for 1 damage
4. Big Enemy: more health but could still be killed on a lucking hit, player roll d100 against this enemy instead of a coin flip, the big enemy still does a coin flip to hit the heroes for 1 damage (or bonus effect)

This gets you a combat where the fighter can take out a half dozen basic enemies before they need healing. It is super fast… and just kind of works. Yes that example is over simplified to the point of inaccuracy, but if you want to do a quick test on the old system then use those numbers and see what you think.

Giving big enemies their own rule set is odd by modern standards, but is makes sense going from a war gamer. One-off rules to handle things like siege weapons or tanks were common and this does give the game that video game feel when the boss’ health bar appears on screen and you know things just got serious.

So about a year ago I started to lay the groundwork for my new tabletop campaign. I knew I wanted to wrap up my current campaign, it was a homebrew fantasy setting using 5e Dungeons and Dragons, and those kinds of campaigns need finality to the character arcs. Also, I think 5e is irredeemably broken past a certain power level and we were well past that power level. So I started to aim that campaign to an ending point. I also used this time to revisit my cyberpunk system, Burning Light, which I have written about previously. 

Originally, I had planned to build out many themes and then let DM’s include whichever themes fit their world(s). I actually backed off of that plan because of how that often works at the table. Anytime WotC released a new expansion to D&D people wanted to play those classes/races even if they did not fit the world at all. In fantasy, including an odd race is slightly jarring but is often hand waved and still works ok, but in sci-fi introducing radically different technology can derail the entire setting. Recognizing that this disconnect was incoming, I pivoted. 

So I removed all the class/theme limitations from the game entirely and retitled the system “Star Chaser”. Anyone can be anything and use anything. And that worked really well. I asked my player to “stay adjacent to your character idea” but also over multiple sessions “grow your character”. They did this without a problem. I think it has more to do with specific players learning to love specific weapons and items… which is fine. Specific characters have developed a specific feel to them and a tactical specialization that the player maneuvered around without the need to create classes. Gear as class not only worked, it is better because the exact details of that balance shift somewhat session to session. 

This approach to players of “no, you tell me what you want to play and I will match you with options” really opened up player creativity. I would not have made half the ideas and class-like roles that the player naturally came up with. This open call to create something new and weird really worked. 

In Burning Light, we had the pawn shop function which was little more than me dealing the top 6 cards from my stack of 50 or so 3×5 cards, but in StarChaser there are way, way more items. From about 50 items in Burning Light to 3,000+ in StarChaser. That said, I still leveraged that pawn shop function. I am only in the second version of it, but currently it is 3 random weapons/armors, 3 random utility times, and 3 items with a shared idea like all are relics from a specific faction or all ground vehicles or all new crew members.  

Originally I had some crafting options to act as a backup to the pawn shop. I have since removed them. Partly, this was because the pawn shop generation needed some work as it was generally providing under powered or uninteresting items. Secondly, the alternative crafting method was way too powerful. The players were leaning on the alternative method almost entirely. At the end of the day, I wanted power gain to be connected to the world, not the character, and people trying new ideas, not a strategic level up process. Players have some control over the process by where they complete missions at. Either way, this puts the players dependent on the world to power up and not a rule set.

The long term goal here is this allows the DM to control gear/classes/power-features of the game directly via the pawn shop. If the DM does not like an item, reroll the option if it comes up in the pawnshop. This level of control is still session by session and not all the start of the game. So this is not a DM saying “low magic” during a campaign’s session zero or banning specific classes or builds. Instead the DM is holding back the magic items until much later on in the campaign or simply not letting a class’s mechanics be available in play. I.e., instead of a “No Monks Rule” the party just never seems to find Monk items. This is a more organic, and entirely behind the screen, balancing mechanic for the DM to use if needed. This also allows the DM to keep their world focused on the themes they want to play around. It stops that random idea derailing your world because a designer for a completely different setting published a fun book that works for their world but not yours. If players pull their power from class/source books, that different designer is a threat. If players can only pull their power from engaging in your world, then the DM can pick and choose the ideas that fit their world from that designer. That threat is now a buffet. 

Power scaling in gear is still something I am pinning down. In Burning Light, there was generally only one level of gear. A few pieces had an upgraded version. Originally in StarChaser I had 10 levels of power. Currently I have scaled that down to 4. The “new” four are just levels 1, 4, 7, and 10 of the previous system. That said the overall power scaling may still be too much. I may later scale them to 1, 3, 5, and 7. That said, my players have not really played a ton at the higher level and those items all contain drawbacks. We will have to see how hard that drawback hits as to if we need to scale those items back. 

I think I can get the weapon options a bit more flesh out within the current constraints, but 2 more options would really help. A lot of relic choices have not been fully leveraged so that will partially fill this gap. I also think this system will open up a bit when the drawbacks start to kick in. Some of the movement driven options are also only now being fully utilized. That said, even just 2 truly new weapon types may add in a lot of extra diversity to the combat options. 

So with 6 months of game play behind us, I am happy with mechanics overall. Improvements are needed in places, but happy overall. I think the real challenge of this system has been the world building. 

So where to even start? First off, I am used to building a world made up of cities, but in this system I needed to build a galaxy made up of systems made up of worlds made up of cities. I was two exponential scales larger than I was used to so I would have to invent new solutions. 

So I started at the largest scale and worked down. I spent a bit of time seeing if there was a way to make a functional 3D map but that was a rabbit hole with no easy solutions. So I decided on a flat map with about 100 solar systems. That is an awful idea and I was committed to it. I also wanted some randomness in the locations and pathing. To do this I spent some time in excel, basically using a random number function for 1 to 4 and then removing all results for 2-4. Wanting a 100 systems, that means I needed a 20 by 20 grind. I generated a few of these and picked the result with the more interesting set of patterns. This ended up being 94 systems. 

I then set out to create a random solar system generator. I did some research of what Nasa has found in its search for exo-planets and then applied some of the ideas into a random number generator. The idea here was to create different mixes of stars and then place planets in different bands. I ended up generating the star types first because there are S and P styles of binary star orbits. P types are close together and the planets rotate around both stars. S types are far apart and do not share planets. Each S type star with all their planets rotate around a mutual point between the two stars. 

For single stars and P-types, I used multiple bands beyond the star(s) with the possibility of empty results, asteroid belts, and gas giants with their own moons. The inner system always had 1-2 habitable planets, a low possibility of a gas giant, and a high possibility of a rock world in the first band. The out ring had a moderate possibility of a gas giant with moons in the first few spots and a mix of none, asteroid belts, or small rock world in the last few spots. For S star systems, I just ran the inner system twice and removed the outer system. 

I tweaked this system a bit and ran it a few times, once again picking the option that produced the most interesting mix overall. I also then decided to hand make the “core world” as those systems are actually the systems closest to the real earth, so when good quality data is available, that overwrites the randomly generated system. 

This is actually an approach I think works really well. Use random generation to create something different and unique, then as the DM respond to it and edits it as is needed. Do not hold the random result sacred, instead let it be an inspiration point. The random result will do the bulk of the work for you, but you will tie everything together. 

From here I realized I had just created 780+ worlds. I thought about all the work I had put into just one world in D&D and then released how much work may be in front of me. So take a deep breath, divide up the content, and go at it. I started by just working on the bottom right corner of the map. Sure enough, the solar systems created were interesting enough in their own right, I was able to ask myself “how would this system work?” and was able to come up with answers inspired by the systems. Some stories were smaller, some systems were denser and messy. Some systems were rich empires and other systems were barely scraping by. 

Part of this process was going out and finding every image of a planet I could find. Movies, video games, fan art, all of it was taken and categorized. When building the solar systems I then moved through the various rocks, forest worlds, etc. as they came up in order. This was a quick and dirty solution that worked really well. 

With the galaxy, solar systems, and planets taken care of, I was only left with cities… which has actually turned out to be the hardest part. I was spoiled when running fantasy games and there are so many great village maps and even full one village generators. Neither is true for sci-fi cities. I was kind of able to use some of the village generators… some of the time but only after a lot of manual manipulation. Right now I am working on a project to grab assets and be able to build out the cities in the raw directly on d20. I am not a huge fan of that solution so stay tuned to see how well that goes. 

So today we are going to get very esoteric and do some high level game design analysis: why are all RPGs unbalanced? Seriously, show me a MMORPG that does not have a hundred reddit posts complaining about balance. Or long winded blog posts complaining about balance on tabletop RPGs. Or dark scroll style games that don’t develop cheesy strategies or underwhelming/overwhelming character builds. Why are really skilled game designers getting this wrong everywhere all the time? 

I blame Gygax. I respect the hell out of Gygax, but it is also his design that is causing this problem. To be fair, other designers have had 50 years to improve the Gygax’s underlying design but instead they keep following the same structure dogmatically. So let’s walk through that. 

There is a concept in design known as an Iron Triangle. It is one of those “you can pick any 2 of 3… but you can’t get all 3.” So “service, quality, and price” is the classic example of this idea. I argue that we face a similar trade-off in the RPG design space: player input, balanced output, and deterministic outputs. 

The core of the Gygaxian design that underlines most MMORPGs, tabletop RPGs, and Dark Soul style games is centered on player input and deterministic outputs. The more you can tweak your character, specifically in little details, and those systems directly translate all numbers (which is so ingrained it is assumed) the more room there is for players to find an unbalanced outcome. As these games are highly complex, there are always combos of options where some numbers will not align as expected. I know every designer says to themselves “yea, but this time I’m doing it so this time we will get it right!” and once again we get another unbalanced RPG. 

So what would happen if we did this in the opposite direction? What if we start with the output? What if instead of building a system to generate the outputs… we just design the outputs, let players pick their outputs, and skip all the nonsense in between? In fact, let’s look at all three possible options: 

1. [player inputs + deterministic outputs = unbalanced game]
2. [balanced output + deterministic outputs = no room for player’s character design]
3. [player inputs + balanced output = builds are not continuous]

So most RPGs, etc., systems follow the first option. MOBAs and fighting games follow the second design mix which works great for competitive games but lacks variety. You can play Ryu, but everyone’s Ryu is the same Ryu. The 3rd is the most interesting to me. What if we predetermine all outcomes but let players select between outcomes? Yes, this removes “my warrior does +8 base damage with 11% critical hit and your warrior does +9 base damage with 9% critical hit” but is that slight level of customization worth an unbalanced game? 

And this is my pitch to game designers. We need to stop trying to make players engage in the spreadsheet game. So yes, still have customization, but instead of a contentious mix what if we just figure out how to build 12 different kinds of Ryu and then let players select between them all? We can still use the ideas behind stats… but balance each character based on those ideas. So have a slightly stronger Ryu, a slightly faster Ryu, and a Ryu with a higher critical hit chance. Each of those answers are messy in a deterministic system because each character might need a different mix of speed boost or critical hit chance. Those aren’t universal answers across different character types and when we treat them like they are we get an unbalanced game. That or we do what World of Warcraft ended up doing, which is make those numbers work differently in not just every class, but every spec within every class. A “+ critical hit” bonus becomes very powerful for some and for others it does not. That is a massive balancing project and one that only makes sense in the context of a computer game which does all the math for you and a very large and well funded design team that constantly patches their game… and the balance is still wrong in most expansions. 

So what does it look like to design to determined outputs? And how does this fit into leveling mechanics? 

My first attempt to do this is found here: https://seanwitherspoon.com/2019/09/07/alpha-test-of-my-cyberpunk-table-top-game/ and it worked well. This version tested end points with characters having access to a limited pool of gear. This version did limit gear access by a class like mechanic and gear did not level up, all progression as horizontal. It gave players lots of choices that they like and the players thought it felt really well balanced. It was also super straightforward to make a character and select options. 

While I think this was a great starting point, it is time to push it further. In my next iteration, I will be both adding in leveling elements and removing the class limitations. Both of these elements may unbalance the game, so that is something I will be watching for. Adding in leveling elements was a fun problem for me to solve for and my answer is actually fairly simple. I created a spectrum of power options for different profiles of weapons/powers and placed everything somewhere on that spectrum. This creates a really simple system to answer the question “what does the leveled up version of this weapon look like”. This spectrum has also been leveraged to build a very fluid crafting system and gear upgrade system. A nice side effect of this approach. 

The balance concern comes from two places. The weapons profiles are built around the idea that some weapons should do more damage because they are hard/risky to use and some should do less damage because they grant some other bonus. I did this with the weapons in my cyberpunk game and it worked really well, but does it scale well? Does it still feel balanced if different characters are upgrading different parts at different times? I am leaving the upgrade mechanics pretty wide open here so we will see if that gets out of hand. 

The second balance issue is I am no longer relying on classes as a balance crutch. Yes, classes are a balance crutch in most systems because if everyone can do something special then you need everyone even if 80% of the time that other class is notably sub optimal. Classes can be used as a way to provide a minimal level of balance. You are the best at least 20% of the time, so it is ok if you suck the other 80%… but it actually isn’t ok it is just bad design. Previously I had 7 classes and everyone was a combo of 2 of them. Here I will be leaving classes out entirely… but some higher level items will require training or other in game options/investments that limit access to them. The idea here is to have so many good things worth investing in that not everyone will invest in everything. This is a way to back into classes at mid levels using softer systems that players can invest in now or invest in later. 

So there you have it. In a few weeks I will start to test this new design which removes the deterministic outputs and instead focuses on pre-balanced outputs, character input being a selection between them, and player not forced to struggle through the weeds of additive stats or multiplicative class features.