Labyrinths & Liontaurs
Contents

Notes

Labyrinths & Liontaurs Manifesto

In designing a new version of a game, I think it is important to make changes only because design goals require them, not because the re-designer had a clever idea or simply desired novelty. Having made this declaration, let me admit that I have fallen short. The liontaur race, the break skill, and several other options are included in Labyrinths & Liontaurs just because I liked them, for example, not out of philosophical conviction. Still. In general I have tried to stick only to changes that are required by the demands of my game philosophy. Here is that philosophy.

THE GAMES WITHIN THE GAME

It has been said that Third Edition D&D and its near relations really consist of several games. You may have heard of the E6 ruleset, or the P6 version in Pathfinder. In these games, play proceeds through to level 6, and then tops out, with only very slow and minimal advancement thereafter. These games seek to preserve a sense that the player characters live close to the edge, that they are mortals battling forces greater than themselves, that PCs are heroes, not superheroes.

I agree with that basic idea, and I’ve taken it further, because D&D breaks into four different games, actually, with four different styles of play:

From character levels 1-5, you make your way in a gritty, "realistic" world in which you are just a couple lucky hits away from permanent death. You are not far removed from humble beginnings. You are a callow neophyte, a tyro.

At levels 6-10, you are a competent adventurer, but your concerns are still local. You can get raised if you make it to the right cleric in time. There are plenty of knights and ancient mages and other creatures more powerful than you. The word of the local duke or king means a lot.

At levels 11-15, you ARE the local duke or king, concerned with top tier threats, like armies of orcs and dragons and vampires. You are the priest who raises the dead. You are a famous hero of the land.

At levels 16-20, the game changes yet again. Now you are like a legend out of myth, like Hercules, Beowolf, and Gilgamesh, fighting demon lords and liches, traveling the planes, finding artifacts, and dealing with gods as you once dealt with kings.

In Labyrinths & Liontaurs, these varieties of play are baked into class mechanics. Each of these modes of play is called a “tier.” The four tiers are tyro tier, adventurer tier, hero tier, legend tier.

I think that this model of D&D is very insightful and useful -- and the game should reflect it. In E6 and P6, the game ends at level 6, with some slow advancement after. Back in First Edition, a character reached “name” level around level 9-11 and stopped rolling hit dice at that point (gaining just a few hit points per level). Pathfinder's published "adventure paths" tend to end well before level 20 (13-17 is typical). For D&D, in my vision as a better game, the breakpoints at levels 5-6, 10-11, and 15-16 can all be used as opportunities to cease rapid advancement and accept slow advancement for a little while until it is time to retire characters. The E6 game says, "take the level 1-5 path and cap it with "epic" rules at level 6." I say, "offer the chance to cap or retire play at any of the breakpoints." This model also allows slow “epic” advancement after level 20.

I have instituted other mechanics to reflect this model. If a feat grants a numeric bonus, then the PC can take the feat and stack it with itself, but only once per tier. (Thus, Pathfinder’s feats that advance with level -- Toughness, Skill Focus, etc -- do not function that way in L&L.) Each class grants a signature ability as a tier is gained. And characters gain a set of class powers and character powers in a repeating pattern with each tier.

Instead of front-loading lots of good abilities at level 1 and ending a 20-level run with a “capstone,” the way the game does now, L&L features singular "signature" powers at levels 1, 6, 11, and 16. I considered placing signature powers at 5, 10, 15, and 20, but what good is a capstone at level 20 that you never get to use? I like 1, 6, 11, and 16 better -- getting a “name” ability at the start of the tier is just more fun when you get to play with it throughout the tier.

THE QUANTUM LEVEL

The underlying core philosophy of 3E is that the quantum unit is the level -- that is, character level as the sum of class levels, The level is the chunk that cannot be broken down further. Moreover, in theory you can combine class levels freely, and the game will remain balanced.

The promise of this philosophy is that a multiclassed character can take as many levels of different classes as he or she wants because every level is equal. It should not matter that a character is a wizard5/cleric5, or a fighter2/rogue3/paladin2/monk3, or a bard 10, for that matter -- because all levels are equal, and any combination is equally valid.

Now, we know of course that it does not work that way in reality. Third Edition failed and fell short of that promise. In 3E, the first level of a class is often the best level. Some levels are notoriously underpowered, like 3E’s Fighter 5. And combining spellcaster classes is an especially bad idea, albeit somewhat patched in 3.5 with prestige classes like the Mystic Theurge, the Arcane Trickster, and the Eldritch Knight.

PF made a valiant effort to fix this by adding more powers at a variety of formerly unexciting levels across a character’s career. L&L follows this model: Every class level grants one interesting power.

D&D 3E’s failure to make each level equally worthwhile extends beyond classes to feat and ability score progression. Look at the rate of feat acquisition in 3.5 D&D. You get new feats at levels 1, 3, 6, 9, etc. What's wrong with that? There is a two level gap between 1 and 3, and a three level gap between all the others. That's not elegant. Moreover, if all levels should be equally valuable, then levels at which boosts are granted should be spaced equally apart.

Ability boosts, on the other hand, are gained at levels 4, 8, 12, etc. That's better, right?

Not really. When you put the two progressions together, you get "good" levels at 1, 3, 4, 6, 8, 9, 12, 15, 16, 18, and 20, over a 20-level career. But that makes levels 2, 5, 7, 10, 11, 13, 14, 17, and 19 "empty" or "bad." And it makes level 12 "supergood," since you get both a feat and an ability score boost then. Ugh.

Look, most powers your character gets in 3.5 Dungeons and Dragons -- hit points, attack bonuses, saves, class powers -- come from your class. If all class levels are equal (and they are not, but if they were), then the other things you use to improve your PC (feats and ability score boosts) should also be equal.

To achieve that goal, in L&L feats are gained twice per tier. Ability boosts are gained once per tier. But that still leaves two "empty" levels. Therefore, in L&L, over the course of advancing one tier, you also gain a background-based power, called a Trait [based on PF1 Advanced Players Guide traits] and a race-affiliated power.

Non-Class-Based Character Progression

Tier 1: TYRO Tier 2: ADVENTURER Tier 3: HERO Tier 4: LEGEND
Lvl 1: Trait Lvl 6: Trait Lvl 11: Trait Lvl 16: Trait
Lvl 2: Feat Lvl 7: Feat Lvl 12: Feat Lvl 17: Feat
Lvl 3: Racial Ability Lvl 8: Racial Ability Lvl 13: Racial Ability Lvl 18: Racial Ability
Lvl 4: Feat Lvl 9: Feat Lvl 14: Feat Lvl 19: Feat
Lvl 5: Ability Score Boost Lvl 10: Ability Score Boost Lvl 15: Ability Score Boost Lvl 20: Ability Score Boost

In this way, with every level gained, a character gets one class-based power and one non-class-based power. No one level offers much more than any other; no levels disappoint.

THE PROBLEM OF ABILITY SCORE GRANULARITY AND OPTIMIZATION

In a similar way, ability score progression is also bumpy in both 3E and PF. Even numbers grant a bump; odd ones do not. In a game that grants ability score increases in +1 increments, that’s bad. So in an echo of 1E design, for L&L, some ability-based bonuses progress at odd levels (Dexterity bonuses to reflex saves and to attack), while others progress at even levels (Dexterity bonuses to AC and initiative), from a base of +0 at ability score 10.

In L&L, the ability boost that all characters gain at levels 5, 10, 15, and 20 can grant +1 to one’s highest stat OR +1 to a different stat in addition to +2 to lionheart points. This encourages players not to keep stacking +1 on their highest score.

There is no point-buy option in L&L for ability score generation -- such systems only lead to over-optimization. However, characters should not be penalized for rolling poorly, especially given the goal of more equal character power from player to player. Therefore, the number of lionheart points the game awards is inversely related to your ability scores. If you have lower stats, you have more lionheart points; higher stats, fewer lionheart points.

Rolling randomly for ability scores means that most PCs will not start with a maximal ability score in the 18-20 range. A score of 20, entirely likely for a first level 3E or PF player character, especially using a point buy system, is impossible at first level in L&L. Only one in eight characters will have the option to start with a 19, and then only if they put their highest number into the stat boosted by a +2 racial bonus.

THE FIRST LEVEL BUMP

First-level characters in 3E D&D are fragile, sure. But the game softens the blow a little, by granting full hit points to first level characters, and by awarding quadruple the normal number of skill points as well. Both of these benefits only accrue at first level.

The problem is that these benefits -- granted only at character level 1, not when you take the first level of a class -- end up creating a balance issue for multiclass characters: The class you take first matters.

For example, if you want to make a skill-focussed cleric/bard, the rules strongly encourage you to take the bard class at first level, since that maximizes your skill ranks. And compare the fighter1/rogue1 (16 skill ranks, 13.5 hp on average for an Int 10, Con 10 PC) to a rogue1/fighter1 (34 skill ranks, 11.5 hp on average). If you wanted to make a barbarian/cleric, there is absolutely no reason to take the cleric level first. The barbarian1/cleric1 has 16.5 hp and 18 skill ranks; the cleric1/barbarian1 has 14.5 hp and 12 skill ranks.

The game gives starting characters another bonus: an extra boost to saving throws at first level. But this bonus applies to higher level PCs when they take a second class -- giving multi-class characters a saving throw boost that singletons do not enjoy. A cleric2/monk2/ranger2 has base saves of Will+6, Fort+9, and Reflex+6 at character level 6. A singleton monk does not equal that until level 8 for Will and Reflex, and not until level 14 for Fortitude. And monks have the best saves in the game.

So looking at these three game mechanics that give the Level One Bump -- hit points, skill ranks, and saves at first level -- the system clearly breaks down a little for multiclass characters.

The promise of 3E's multiclassing rules is the promise of level equivalence, that is, no one combination of classes should be more or less powerful than another. Yes, with any span of choices, some of those choices will be more optimal than others -- but the game should act to minimize the imbalance. The Level One Bump undermines that philosophy, because it contributes to some level choices being more powerful than others.

So we have to consider revising the Level One Bump so that it neither advantages nor disadvantages multiclassed characters. The Level One Bump serves an important purpose. Brand new characters ARE fragile, and they need a bit of a boost. There's nothing wrong with a bump at first level, so long as it applies equally to all characters, and so long as it does not create artifacts like favoring one multiclass option over another, or over a single class option.

Here is my solution: Make the bump a factor of race, not class. Yes, give starting characters extra save bonuses, skill ranks, and hit points ... but make those benefits a consequence of race selection, not class choice. You only pick a race once, and only at first level, making race an ideal carrier for a one-time bump. In L&L, starting bonuses to skills, hit points, and saving throws are all linked to race.

THE PROBLEM OF CASTER POWER

One criticism of 3E class balance is that fighters, especially at high level, end up underpowered. Rogues, monks, and caster classes are all better warriors that they were under 2E, but high level spell casters especially outshine fighters. See this post for an interesting analysis of this issue.

I’ll note that Pathfinder exacerbated this trend by boosting hit die types for several classes (especially d6s for arcanists), plus a few other boosts. L&L sees this problem and takes some effort to fix things:

I wanted very much to vary nonproficiency penalties by class, say, for wizards, -5; for fighters, -1. But I could not make the mechanics of it work for multi-class characters realistically or elegantly.

D&D 5E takes the goal of capping caster power very seriously. The use of concentration makes it much harder to combine buffs. Caster level never boosts variable effects for spells. Casters gain fewer spells. Spell DCs start lower and are harder to raise. Ability scores are hard capped at 20. There are only three magic item slots. I personally think 5E goes too far, but there are echoes of these ideas in L&L, since the goal of nerfing caster power is a valid one.

THE PROBLEM OF WEALTH/ITEM CREEP

One of the reasons that many Pathfinder campaigns top out at level 15 is because there is too much power creep at higher levels -- and because of that it becomes very hard to create balanced challenging encounters at high level. A big part of the problem is that characters use plentiful wealth to create and buy a wide range of magic items. As a game master, there are things you can do to work around the high level power creep problem. But L&L includes several mechanical adjustments to disincentivize too strong a reliance on wealth and magic gear.

The “Wealth By Level” system is baked into 3E, and to strip it out (as the game designers did in 5E with its rarity and attunement rules) is a bridge too far. Moreover, because the game incentivizes players to look for numeric maxima, players tend to focus on ability score boosters, rings of protection, cloaks of resistance, etc etc. That means a lot of really nifty items are disdained by players. The Island of Misfit 3E Magic Items is well-populated.

To reduce power creep and encourage the use of more flavorful items, in L&L, I’ve halved the Wealth By Level guidelines for the total value of character possessions (with even larger reductions at the highest levels). However, the price of all items that do not grant numeric bonuses is also halved. Any flavorful item that requires an enemy to make a saving throw uses the PC’s caster level and relevant ability score to set the save DC (like a staff does). Creating new combination-power items is outlawed, so character can’t stack the effects of multiple items into one slot -- ability boosters only benefit one stat, not two or three, and each stat booster item has its own slot, as in 3E. Thus, to boost six stats one needs six items. Those ability booster items advance in +1 increments, not +2, since odd stats see advantages; and they max out at +4, not +6. I’m removing the stat booster ioun stones and magic books that raise ability scores.

I’ve also retained 3E’s use of experience point loss as an important prerequisite to magic item creation. I’ve reduced the feat cost by including only four item creation feats, at one per tier. Since all classes have casting, all characters can use experience points and take a feat to craft magic items. Forcing characters to use experience points instead of gold for item creation results in fewer items in the party and in the world.

THE PROBLEM OF POWER INFLATION

All traditional RPGs give player characters options to increase in power over the life of the PC. That advancement in power gives players a feeling of progress and meaning. The problem comes when that power advances too fast and too far. Power inflation is an issue when PC power becomes so great that it becomes hard to create good challenges, and when power discrepancies among PCs are too large.

First Edition D&D, for all its flaws, did not suffer from level-based power inflation. Hit dice maxed out in the 9-11 range. Advancement at upper levels did not continue to pile on abilities and goodies (well, bards and monks aside). It was arbitrary, but it was not inflated.

Compare to Third Edition. Skill synergy bonuses, for example, stacked too high. Hit dice and powers advancement never capped. And as more and more splat books came into play, those threw the power curve even more out of whack. Pathfinder had similar problems: feats that scaled with level, characters whose magical gear is more important than their class abilities, too many ways to add cohorts, companions, and followers.

In L&L, I have two solutions to this problem. First, it is expected that characters will be powerful at the Hero and Legend Tiers, so there are rules and advice on how to create adventures and challenges that are more suitable at these power levels. Second, I am decreasing the power curve in several notable ways. I’ve already mentioned bringing down spell caster power and magic item power. Another example: I’m removing all level-based advancement from feats.

THE PROBLEM OF MURDER-HOBOES AND THE PRISONER DILEMMA

Traditional D&D was created as a murder-hobo game. Wandering "adventurers" are home invaders who bash in the door, kill everyone inside, and take all their wealth. Then do it again in the next room. And for some reason this is OK because these "heroes" are invading "dungeons" populated with "monsters." Now, I've been in the hobby since it started; I remember using the OD&D booklets, and I remember the exciting times when "Advanced" D&D was released. Sadly, that means I also remember too well the moral panic against D&D: the stupid accusations that a game with devils meant RPGers were Satanists; that when your character died, weak-minded players would end up committing suicide; not to mention the Tom Hanks movie that demonized the game; the printed propaganda that they handed out in churches; and more. All of that was just pure bunk, of course, but there was a kernel of truth in there: Some players have enjoyed playing evil characters, honoring evil gods, and telling stories in which the "heroes" were the ones committing atrocities.

And even if you did play true heroes who show mercy and spare the lives of defeated foes, then what? D&D has its own prisoner dilemma: unless there's a civilization with a criminal justice system somewhere close by, either you slay defeated foes in cold blood or let them go to do more evil another day. For a hobby that's supposed to be fun, rewarding mercy with poor choices seems harsh. Is this the ethical morass we want in our gaming?

Heck, I recall debating my "D&D morality" with my own father, a devout Catholic. Those were his words. Back then, I argued that a D&D morality was a good thing, inspiring teamwork and creativity and collaborative storytelling. I did not mention that D&D could also include immoral stories. I did not mention the time my friend, angry at an innocent NPC, decided to "gut the wench" -- and when we challenged him on it, declared, "fine, I only HALF-gut her." We laughed at the absurdity of that, but in truth we were pretty tame compared to RPGers who embraced misogyny and rape, torture, and casual murder as part of the stories they told.

My question is this: Should RPGs embrace all styles of play, both good and evil, presenting all options as equally valid? D&D and Pathfinder implicitly do. But as for me, since L&L is the game I want to play, I am including options and mechanics to discourage evil play and to reward those who choose to be heroes. Good implies a respect for life, and a reluctance to end it except when unavoidable. Evil implies a delight in cruelty, a wanton disregard for life. In Labyrinths & Liontaurs, there are good options to defeat foes without killing, and using those options is rewarded. I offer ways to deal with prisoners other than death or freedom -- by using spells, oaths, and diplomacy to curb bad behavior and even to alter alignment. You don't have to be a murder-hobo.

TOWARDS BETTER CHOICES

D&D 3E and Pathfinder 1E share a problem: It is too easy to make bad choices and end up with sub-optimal characters. Similarly, it is too easy for clever players to make optimized choices and end up with overpowered characters. There is also the “Fourth Edition” problem -- in D&D’s Fourth Edition, the designers went overboard in several ways, but most notably, in making all characters more alike. If mages are like fighters, they will be balanced. But that’s an unsatisfying way to approach the problem; rather, the better solution is to make choices both meaningful and balanced; to makes options distinctive and still offer fewer suboptimal choices. I’m trying to do that in L&L.

Here are a few ways to help solve these problems:

Make characters of equal class level more equal in power. To do that, make sure that the class abilities granted per level are comparable across classes. Also, when choices are available, try to set up power selection so that one player cannot take exclusively synergistic combat powers while another player takes RP-based or non-synergistic choices.

Decrease granularity for all classes and levels. For example in 1E, Fighter 7 was a real breakpoint because one was granted three attacks every two rounds. In every edition of the game, Wizard 5 is powerful (Fireball!) while Wizard 4 and Wizard 6 are meh. This can be fixed, for example, by granting a new level of spell at every level, so that the game ranged from level 1 spells to level 20 spells (not the traditional 1 to 9).

Revise multiclassing so that combining levels from different classes does not result in severely underpowered or overpowered characters. The cleric5/wizard5 should be in the same ballpark as the wizard 10 or the cleric 10. Such is not the case in 3E or PF.

HOWEVER, it is important to also avoid the danger of 4E, that is, making classes too interchangeable, boringly so. I’ve done that with L&L’s spell lists, for example. There is no overlap among the lists. No wizard spell can be cast by a cleric, no cleric spell by a druid. And I have created a limited list of inherent spells for fighters, barbarians, rogues, and monks to use. This makes the choice of class more meaningful, and it offers multiclass casters options that single-class casters do not have.

I’ve also done this with the use of Third Edition (not Pathfinder) multiclassing rules, including an experience point penalty for uneven multiclassing. Moreover, my extra racial abilities gained at higher tiers, and tying the first level bump to race instead of class make the choice of race more meaningful (but also hopefully balanced, if I’ve done my job correctly).

Regarding those multiclassing limits. In Pathfinder, which removed the xp penalty for uneven multiclassing, it is theoretically possible to take any combination of classes. This is not much of a problem, because it is an obviously suboptimal choice in the majority of cases. You can’t shop around too much in PF or 3E for particular powers because all you will end up with are a slew of low level abilities.

A Pathfinder Druid1/Wizard1/Bard1/Cleric1 is a poor choice compared with a Sorcerer4, but in L&L, since caster levels stack, a Druid1/Wizard1/Bard1/Cleric1 is more viable, having caster level 4 and access to 4th level spell slots. Since save DCs are based on spell slot level, not spell level, those slots are useful, even if only used to cast first level spells.

FRACTIONAL ADVANCEMENT

One disadvantage of a d20 system is that the smallest number you can add to the roll of a 20-sided die is +1. That might not sound so bad, but in practice, it means that the chance to succeed or fail in any endeavor must be incremented in 5 percent chunks. That's because the odds of rolling any particular number on a d20 is five percent; for a span of numbers from 1 to 20, you have a 1 in 20 chance of rolling a given number, that is, 5 in 100, that is, 5 percent.

So as d20 fighters improve in fighting by +1 every level, then a base attack progression by less able characters needs to advance by less than +1 per level. That means you have to use fractions of one sort or another.

A wizard, for example, progresses in fighting by +1 every two levels. That's the same as +0.5 per level on average over an even number of levels. A cleric progresses by +3 every four levels; that's 0.75 per level, on average. But the designers of 3.5 chose to keep the attack increments as +1s, not as fractional bonuses. Over enough levels, that is fine for a single-class character, but for a multi-classed character, one can end up with bad rounding. Too many +0’s and not enough +1’s. The solution is to include the fractional improvement for use by multiclass characters.

And following that model, in which all classes contribute fractional advancement to saves and base attack bonus, so too in L&L all classes also contribute to caster level. Thus, caster level and spells per day are not tied to class, just as BAB is not tied to class. A wizard/druid should have a single caster level that is the sum of both caster levels, NOT a wizard caster level and a seperate druid caster level.

Just as some classes have weak attacks, some classes also have weak casting. Fighters and barbarians contribute only +0.25 per class level. Rogues, monks, paladins, and rangers contribute +0.50 per class level. Bards, +0.75. And the three prime caster classes contribute +1 per class level to caster level.

This common caster level would be used for all classes for all the usual purposes -- determining effects such as duration, range, and damage for some spells. Also for concentration checks and caster level checks to dispel magic or penetrate spell resistance.

If one's caster level is the sum of all caster levels from all classes, then the multiclass caster, although still penalized by a lack of high level spells, at least can use higher level spell slots to cast low level spells effectively. The trade-off as a multiclassed caster is to gain access to a wider variety of spells and a large number of low level spells but to lose access to high level spells.

NOTE: Forcing players to add up fractions is poor game design. In L&L, pure single-classed characters can use the same class progress table that they have used all along, with no need to add up fractions. It is only when you multiclass that you have to start adding up fractional bonuses. And since all fractions are all in quarters, halves, and three-quarters increments, even that is not so bad. Compare with 5E, in which some casters advance by thirds!

CUSTOMIZATION VS COMBAT OPTIMIZATION VS BALANCED CHARACTERS

In 1E, there was almost no way to customize your character. If you were a fighter, you had the fighter array of powers. Ditto thief, magic user, and all the rest. The only customization, to set your cleric off from other ones, was through weapon choice, magic items, and spells chosen. Magic-users and druids could customize via familiars and animal friendship, but that was not much.

2E made character creation more customizable, with non-weapon proficiencies, cleric spell specialization, fighter weapon specialization, etc. 3E, even more so, with ranger combat style, rogue special abilities, etc. And PF1 went even farther, with sorcerer bloodlines, paladin mercies, traits, etc.

This ability to customize your PC is a good thing. Granted, it makes character creation more complex, but, on the other hand, it gives the player more choices, and making choices about your character deepens your involvement and engagement with the game. Customization makes the game strategically and tactically richer. It gives power gamers opportunities to create exciting PC builds; it gives role players the chance to more precisely match their vision to the characters they create.

The danger in customization is over-optimization. That’s the pit into which PF2 has fallen. By expanding feats into the stratosphere, it becomes possible to use feats to optimize for combat without ceiling. And my understanding is that players tend to do so, at the detriment of roleplay choices and other modes of play.

In L&L, I want to encourage choice and customization, but I want to discourage combat optimization over all other choices. Here is how I’m doing that:

WAR VS SPORT

One way to look at modern game design is to consider two distinct styles of play. Combat-as-War emphasizes a more simulationist, grittier game in which combat may be lopsided. Clever tactics and planning may give players an overwhelming advantage, or they just even up a conflict that would otherwise be beyond the PC's capabilities. The world is a sandbox for players to explore, with internally consistent rules and consequences. It features non-combat solutions to problems and resource management, including careful use of healing and daily ability use. It is more realistic, in that PC death is possible if you play foolishly, or are simply unlucky. Combat-As-Sport is a gamist approach, in which enjoying balanced combats is key, without worrying about the minutia of the campaign world. Time-keeping, resource management, convoluted planning, and the like are annoyances that make the game less fun.

For more on these two philosophies of how to enjoy RPGs, check out this essay on the topic.

This relates to my points on optimizing characters for combat as opposed to seeking to create well-rounded characters. If all you seek in a game is arena combat after arena combat, then by all means play Pathfinder 2E or a game like it. If you seek varied ways to creatively approach problem-solving, then L&L encourages play in that mode. Yes, this puts me in the Combat as War camp, though of course no RPG system makes one style impossible; a talented game master can bend a system to run the campaign they prefer.

But let me show some examples of how L&L compares with other games on the War vs Sport continuum.

DESIGN GOALS

Beyond these considerations, a better rules set can prove its worth in design fundamentals. Consider these key goals for good design:

Simple -- Games should not be overly complex. Well, as a Dwarf Fortress fan, maybe I better hedge that statement. Simplicity is a feature that makes a game more accessible to a wider swathe of players, and that's a good thing. For example, 3E multiclassing is easy. When you advance a level, you pick a class and add a level from that class. That's it. The mechanic is very easy to grasp. In L&L, it is even easier -- no half-ranks for cross-class skills, a flat 10% xp penalty for uneven multiclassing, not 20% per uneven class.

Intuitive -- The 3E system is readily grasped by new players. It requires very little explanation. it is conceptually easy to figure out. I hope I have made the L&L class and spell system even more intuitive.

Make A New Thing Like An Old One -- Although I am a reformer, I see the value in conservation. For wider general appeal, there is value in keeping new rules as close to the old ones as possible. Yes, make radical alterations where required and justified by core design considerations. But resist the temptation to "fix" things that are based on personal preference. Sure, roll up Listen and Spot into one skill, but call the new skill "Spot" not "Perceive." Everyone will know what is meant by "Spot." That makes the learning curve less steep. Adapting PF Traits for use is fine, but better to call them “Traits” rather than "Accomplishments." so that Pathfinder players will figure the game out more easily. Both PF2 and 13th Age suffer too much from change for change sake. It is disorienting. For a new game seeking adoption by D&D3 and PF1 players, only change things that are essential to the required philosophical design strictures.

Elegant -- In this context, we are not talking about white tablecloths and crystal vases -- this is the mathematical sense of the word. Per the Wikipedia, "surprisingly simple yet effective and constructive." For RPGs, I take this to mean:

L&L strives to be a more elegant game. Look at the number of spells per level in the spell lists, the inverse relationship between lionheart points and ability scores, and the reduction of all companions to one set of rules, for examples of what I mean.

FOR FURTHER REFERENCE

If you have managed to get this far into my blather without falling asleep, you might be interested in a collection of spreadsheets I used in planning components of L&L. Enjoy!