Sex & Romance in Role-Playing Games

PCs should be getting laid. If fiction is any guide, the three universal motivators are love/lust, greed, and revenge. They are human urges so basic that when presented in fiction they need no further justification. This is what people do and have been doing at least as long as we’ve been writing about it. Sex and violence are just good story-fodder.

The thing is, this tends not to be the case in RPGs. Yeah, we get plenty of money and violence, but where’s the sex? Where’s the romance? I’ve always wanted to find a way to encourage fictional romance at the table, but the problem seems three-fold. The first issue is that games very rarely touch the subject. When it does come up, it’s often in some form of generational play, concerning the joining of families or creation of offspring. This is more logistics management than romance. PbtA games buck this trend slightly, with Apocalypse World having moves based on sex: “When you sleep with someone, take +1hx with them”etc. This at least brings the thing into active play, rather than a background strategic element. Still, it’s... lacking. In my experience, even the sex Moves aren’t enough to pique a player’s interest unless it offers some kind of strong strategic advantage.

The second problem is that players, out of character, do not care if their PC gets laid. Players are goal-driven creatures. They want to win. They want to earn points, they want to beat the mission, they want to accomplish their goals. For the player, there is no real temptation to divert time and energy into a romantic interest that does not bring them closer to said goals. Compounding this, we’ve sort of trained players to want the exact opposite. If you do get romantically involved with an NPC, you’re just asking the GM to take it as leverage to use against you. Combine all this with the lack of mechanical attention paid to the topic and even those players who might be interested in romantic story elements may not be aware it’s even a valid option. After all, games communicate what they are interested in by the rules they choose to include.

So what do we do with this?

Sex and romance are a pretty hardcore set of biological drives for most humans, at least at some point in their lives. They are a unique area of focus that is difficult to sublimate just by focusing on other areas of development. They are such powerful drivers that we often do very dumb things in the pursuit thereof, eschewing our more rational priorities for even a fleeting glimpse of that rush (or release.) This is such a powerful factor in most of our lives  that as an audience, we don’t even balk when fictional characters risk upending their entire lives because they were lonely or the bartender was cute. Many of us have been there, man. 

In general, the way you get PCs to act like real people is by making the mechanical incentives of the game mirror the physical or psychological incentives the PC would have in-character. You are never going to get a player to really care about their PCs level of comfort or discomfort, or worry about the quality of their sleep. However, if you mechanically reward them for being exceptionally well-rested or start punishing them for getting shitty sleep on a bad mattress, the interests of player and PC align. Suddenly, the PC will be very concerned about the quality of their mattress and a good night’s sleep.

If we want players to be tempted to actively court romance or be tempted to make dumb decisions concerning their gender of choice, then we need to creative incentives for the players to do so. Getting this right, however, requires a few things:

• It needs to be a discrete subsystem, the very existence thereof drawing attention to the fact that this is a valid thing to explore in play.
• It needs to be built in such a way that it can be accessed or ignored at the player’s preference, without having to specifically build your character around taking advantage of it. This is ultimately why relegating the concept to a drive or trait doesn’t quite work. It’s fine for the P/PC who is specifically set out to do that thing, But for anyone else it may as well not exist.
• The rewards must be unique enough that you can’t get them from other systems in the game and compelling enough to tempt goal-oriented players into making trouble for themselves to do so. This is even more important when you remember the liability potential of tying yourself to an NPC.
• And ideally, we’d want it to (wherever possible) integrate into other areas of the game and plug into/make use of the existing mechanics so that it feels a natural extension of play. Accessibility is key.

You may have noticed above that I said three-fold problem and only listed two. The third and perhaps largest issue comes from the topic itself. Simply put, people have baggage. For a lot of people, sex is a significantly greater taboo than violence. The same group that will happily engage in fictional sword-fights and describe a disebowelment in gratuitous detail will get squirmy at even vague descriptions of physical intimacy. Even someone who is otherwise very comfortable with the topic of sex and their own sexuality may not be terribly interested in role-playing the details thereof with and in front of their friends at the table or watching said friends do the same. Unless, you know, you’re into that.

Existing RPGs haven't done themselves many favors. When people think of the topic, they often think about things like FATAL where the its inclusion is about simulating the act itself. Add to this the above issues of social-squick and potential prior bad experiences due to either group immaturity or deliberate bad actors and it's easy to see where this can lead to uncomfortable territory. Further add in the lack of representation of the topic in RPGs and it creates a culture where sex and romance are neither a normal part of expected RP experience, no one people know how to explore in a safe, productive fashion. No one wants to end up on r/rpghorrorstories.

The thing is, all of this latter point is a red herring. Most of the discomfort (and prior bad experience) comes from the idea of playing out sex-scenes at the table.  As ludicrously amusing as it might be to develop a God of War-style sex mini-game out of Full Contest (how else would you know who won?), that's not only unnecessary, it misses the point. The reason that sex and romance are valuable as a story element is through the exploration of character motivations and the relationship dynamics it creates. While the gratuitous nudity is good for HBO's ratings, the physical act of sex is not where the drama lies. From a dramatic standpoint, it's all about the lead up and the consequences to the relationships involved.

From a game standpoint, the purpose of such a system would be both to demonstrate sex and romance as a legitimate are of play and encourage PCs to behave in ways that feel more natural to reality and fiction. In so doing, it helps close the gap between the kinds of stories that S&S generates and the kinds of stories S&S seeks to emulate. 

Creating such a system would be a delicate balance. To feel legitimate, the system would require enough structure and guidance that it was teaching players the right way to use these elements in a story and explore them safely through play. That will include a necessary emphasis on the relationships themselves, in and out of character, and handling the subject matter in a way that's comfortable for everyone involved. Such as keeping the actual climactic details safely veiled and off-screen.

I don’t have more specifics, yet, but it does occur to me that this is a very close cousin to the social conflict topics I’m currently developing. One could easily plug into or effect the other. It’s thus worth a bit of consideration, as I’m solidifying the latter. It’s also a very, very big element in the kind of fiction that S&S seeks to emulate. How many Scoundrel-esq characters practically live out of a brothel?

At this stage, I’m seeking input. Any thoughts, questions, ideas, or suggestions? Have you seen sex/romance used as a part of play at the table? How did it go? How do you feel about the topic and its inclusion? Is this an area of development that interests you, or something you’d prefer left to HBO?

OSR Project #2: The Danger of Skills

There's an argument floating around that will claim that skills expand what your character is capable of by letting you invest in a bunch of different things, rather than being confined to an archetype like a class. I'm here going to have to argue that the opposite is true.

Skills by definition impose limitations on what your character is capable of.

When OD&D was released, there were only three classes - Fighting Man, Cleric, and Magic User. All of these characters were assumed to be dungeon-delving adventurers and have all the skills and abilities appropriate of such. Later, Greyhawk was released as the first D&D supplement and it added Thief as a class. Suddenly, the dynamic of the game changed.

The thief is essentially a class based around having skills. Climb walls, hide in shadows, move silently, pick locks, disarm/detect traps, etc. The Thief as originally written is Dungeoneer: The Class. This creates a lot of weirdness compared to how the game was played before. If you listen to Arneson talk about his early D&D experiences, everyone acted like a thief before the thief showed up. Everyone was sneaking around, picking locks, hiding and setting up ambushes, disarming traps, etc. The creation of the thief changed the way the distribution of abilities was perceived. Because the thief has a mechanical means of doing these things on their sheet, suddenly those things became the domain of the thief. Worse, because the thief had a mechanical ruleset for doing these tasks, it gave the implication that because no one else had access to these mechanics the thief was the only one who could do it. (There is a very interesting argument to be made that the nature and wording of the thief's abilities was supposed to imply a slightly supernatural character to them, thus everyone could sneak but a thief could literally disappear in a shadow. I actually prefer that interpretation, but it is outside of what I want to discuss here.)

When there is no mechanic for a thing, it's in the common domain. Anyone can attempt it by navigating the fiction. When you introduce a mechanic for a thing, you codify it and ultimately limit it in some way. This isn't necessarily a bad thing, but it's an inherent side-effect of having a mechanic for it.

Now to make this less esoteric, consider for a moment Aragorn as a Ranger. In OD&D or AD&D, I can stand on the wall at Helm's Deep and make an impassioned speech because there is no Oration or Speechcraft skill. I can get up there, do my thing and feel pretty damned heroic about it because of course I can do it. I'm a big deal adventurer. The GM might nod and approve. The GM might even let me roll something to see if it has a mechanical impact on the fight. Who knows.. But it's entirely within my wheelhouse. Because there is no mechanic for it, it's up for grabs.

Now pretend we're playing 3.5e instead. Now there's a Diplomacy skill (or whatever it's called in 3.5) that exists for trying to influence NPCs. The existence of this skill means that in order for my character to be good at the thing governed by the skill, I need to invest mechanical resources into making them do so (in this case, skill points). What was something that I might have done and could have done because I thought it was cool and might have made a memorable scene becomes an area of the game I can no longer meaningfully interact with unless I spec out my character specifically to do so. I have to buy back the thing I could have done before, had the skill not existed.

Worse, because there is now a mechanic attached, if I do give an impassioned speech at the walls of Helm's Deep, the GM might make me roll for it anyway and because I don't have the skill I've introduced a risk. I might botch the roll and the GM penalizes the troops for my good intention. The GM might decide that I get up there and somehow drop my speghetti because a lot of games are written with the assumption that a bad roll means the character fucked up.

Where before this was a fun and optional thing that at worst would have been neat to play and at best might have given me some kind of fun bonus from a GM trying to encourage such things, I now have to weigh the risks of even attempting a thing that the game says I'm mechanically bad at because I don't have skill points invested in it. At best, I look unheroic and dumb, at worst, I might accidentally penalize my troops for having tried.

In the 3.5 ranger's case this is even more punitive because I've spent all of my skill points buying the abilities that previous editions gave me just for being part of my class.. And we'll not even talk about how different classes in 3.5 don't have access to certain skills and thus you have to invest even more character resources if you want to do something like be a fighter who also knows how to talk to people.

If you want an even more banal example.. There is no AD&D, B/x, or OD&D character who can't ride a horse. And yet, in 3.0/3.5/pf, Ride is a skill. You can technically ride without it, but if you want to do anything with the horse or avoid any perils of said horse, you now need to drop points in ride.

Making something a skill inherently walls the thing off from the open domain of play. If there's no cooking skill and you want to cook something, you are generally assumed to be able to do it because it's beneath what the game cares to simulate. The moment you create a cooking skill, you are mechanically a shitty cook until you invest resources in being able to do something you otherwise would have been able to do for free.

Lateral Advancement for Lamentations of the Flame Princess and Other Old School Fantasy Games

One of the strengths of OSR games is how simple the advancement scheme tends to be. No worrying about builds or feats or whatever. You just gain your xp, roll your Hit Dice, and go. The B/x style class scheme tends to help this quite a bit. They are nice and broad, defining your character by your skill set rather than forcing specific fluff to make its way into your character. Your Fighter could be a wandering mercenary, an archer, a tribal barbarian, or a chivalrous knight depending on how you want to picture them.

The big four (Fighter, Magic User, Cleric, and Thief) do an excellent job both of providing broad umbrellas to build characters under and providing some niche protection. As I've written before, though, there are certain character concepts that just don't fit comfortably under any of the above. How do you handle a Ranger who can actually fight? How would you build Conan under such a system? Or Elric?

This brings me to my idea for Lateral Advancement for LotFP. OSR already works with the understanding that not everyone advances at the same rate. More powerful classes already cost more XP to advance, and in AD&D multi-classing was literally just "take both classes, divide XP between them." The premise "choose to advance slower in exchange for more power" is baked in from the start. I think there was a kernel of something useful in there, and I've expanded it here.

Lateral Advancement

Any time a character would gain a level they can choose instead to purchase a lateral advancement, burning the XP they would have used to gain said level in exchange for one of the following non-standard benefits.

• Increase an Ability Score: the character may increase any one ability score by 1 point if they can explain how their character improved it. No score can ever be improved in this fashion by more than 2 (read: a character who rolled 15 strength can only purchase two increases bringing it to 17), but may be purchased separately for each of the character's ability scores as desired. 
• Increase a Skill: the character may add a single pip to any one skill if they can explain how they learned it. Obviously, this is useless for Specialists. 
• Increase Attack Bonus: the character may increase their base attack bonus by 1. Obviously, this is useless for Fighters.
• Improve a Save: the character may improve any one category of saving throw by 1. This may be purchased up to twice for any single save, but may be purchased separately for each save as desired. Obviously, less useful for demi-humans.
• Learn a new class ability: In my home campaigns, I often design rituals and other special abilities that can be gained through play but not necessarily earned automatically as a class ability. If the Fighter got inducted into a berzerkergang cult, this would be his way to buy the ability.   

Characters who purchase a lateral advancement are immediately set back to the minimum amount of XP required to have reached their previous level.

 In LotFP, a Fighter gains level 4 at 8,000xp and 5 at 16,000xp. Upon gaining 16,000xp the Fighter could choose to take a lateral advancement and would then be back at 8,000xp again having burnt the remaining as the cost of advancement. 

I rather like the above as a setup. It allows characters to dip their toes into different fields without being in any danger of overshadowing anyone else's niche as well as giving the DM a way to let characters earn special abilities they want without just handing them out willy-nilly. You want your Fighter to be a Paladin? Do good deeds for the church  and in the course of it, I'll give you the ability to learn some paladiny abilities. Now I can do that without being unfair to the other fighter who is perfectly happy just smacking things with a sword. It's also perfect for when you have a character who just wants a skill to fit a concept (survival comes up a lot with fighters who want to be barbarians or rangers) without having to do some kind of multi-classing setup or hacking together an entirely new concept. Want to be a Witch-Hunting Specialist? Increase your Save vs. Magic a couple times to keep an edge on your prey.

If you're wondering why there's no "purchase magic" to go along with the options for skills and attack bonuses, it's because I'm already using Ten Foot Polemic's Arcana Skill.

Would you use it? Tell me what you think.

Statistics in the OSR market

My OSR house rules have slowly evolved into a game of their own, and I've been kicking around the idea of publishing them after 'Bastards is all grown up and finding its own way in the world. To this end, I've done some amateur market research by way of straw polls. I'm sure none of these are statistically significant by way of sample size, but here they are.

Screencaps are here from the time at writing, but if you want to weigh in, I'll leave links to each below the poll itself.

http://strawpoll.me/7025086/

This poll had the most results, and it confirms to what I predicted. The most popular edition chosen was Moldvay Basic, but nearly twice as many people weighed in and said that they didn't care one way or another about the specific edition. What did surprise me was that OD&D beat out BECMI/RC and AD&D by a decent margin. I can't be sure if this is because OD&D is more popular than I thought, or because more people thought it was underrepresented than the other editions.

Since the leading result didn't care what edition it was, that leads to the next question: how much do we care about faithfulness to the core material?

http://strawpoll.me/7044557/

The results here are more definitive than the first question, but we have an even smaller sample size. That said, it completely conforms to my expectations. Because OSR stuff is so broadly compatible, the overwhelming majority seem more interested in seeing innovation than faithfulness to the old rules.

http://strawpoll.me/7139460/

This last one I've been kicking around for a while. Book formatting is something I've written about before and it's an interesting topic for OSR. Depending on the edition and game, we've seen it done a number of ways. The results here surprised me.

I've never gotten to experience LBB OD&D, but I've always loved the physical presentation of it. The amateurish doodles give it a naive charm, and the nostalgia factor is high. The actual booklets themselves have a kind of versatility that you lose in single-volume sets, as they allow you to have multiple people looking different things up simultaneously. If you can acquire them separately, it also means you can have multiple copies of the character creation section if you need it to speed up the process. I think it may also be less intimidating to initiate a new player into the game with a half-sized 30-page booklet than handing them the 300-page player's guide from certain other editions of the game.

For the first day or so that I monitored this poll, all four options were at a dead heat, with at most a vote or two between them. I assumed that the single volume 8.5x11 would ultimately win out, as it is the most common form RPG books take. I was pleasantly surprised that the 6x9 took the lead. I've written about how much I love this format ever since I picked up the Burning Wheel Gold book.

Even if the above changes significantly after the time of posting, I think I will stick with the 6x9 as the target size for a new book. This is not just because I genuinely enjoy the form factor for it, but because it creates a secondary possibility as well. It's easy to lay out the book at 5.5x8.5 (half-page LBB size) first and create the game in such a way as it can stand as multiple booklets and then just slightly upscale the same layout to fit a single-volume 6x9 without changing page numbers. Print on demand being what it is, it would be easy to offer both for people to pick at their preference. I can easily see someone using the single-volume hardback and a couple extra character-creation booklets at the table.

The Problem With Assassination in Role Playing Games

This is mirrored from a discussion the GH forums, but I thought I'd throw it up here as well. The discussion is specific to the way Band of Bastards works, but it applies to any game with moderate-to-high potential lethality.

As I've been hammering out the rules for 'Bastards, a topic has been eating me for a while. Since Higgins and I were unable to come up with a good solution, I thought I would offer it up for discussion here.

We are going to define the word assassination in this discussion to mean "a situation in which the target is dead without a meaningful chance to prevent or react to it." The primary examples to be discussed are sniper shots and poison.

Both of these things are extremely effective in killing people. That is why they exist. It's realistic and good simulation that they should kill people pretty effortlessly when they happen, and it's fitting that the rules should encourage players to make use of them (if it is the sort of thing their character would do). After all, getting in a fair fight is usually a very poor tactic utilized by the foolish, idealistic, or desperate.

The problem is that this blade that should cut both ways, but in practice doesn't. Mechanically speaking, there's nothing stopping me as a GM from rolling some dice and announcing "Your conversation is cut short when the window shatters. Blood and brains fly out the back of your skull and splatter the wall behind you. A half-second later, your friends hear the report of a high-caliber rifle some distance away." The rules support it just like they do if it were a player character doing the same to an NPC.

The problem is that any GM who would actually do that out of the blue is ..well.. probably an asshole. For a game like 'Bastards to work, you have to have some kind of trust in your GM and killing a PC off without any meaningful chance to prevent it is a gross abuse of that trust. Poison is the same way. An option is always some kind of perception test to spot the sniper or detect the poison first, but even that seems grossly unfair when the fail-state of the test is "you die."

Thus in practice, this becomes an extremely strong tool for the PCs to use, but one that they are essentially plot-armored from ever receiving. At best, the GM is going to have to either intentionally miss the first shot, or an NPC winds up eating a bullet to alert the players to the danger.

In our discussion, Higgins did point out that we are primarily simulating fiction - not reality - and that protagonists very rarely die suddenly and without warning from assassination attempts (George R. R. Martin notwithstanding). This is a good point, but the counterpoint I offered was that the protagonists in fiction aren't aware of their own plot-armor. By contrast, players can generally trust that their GM isn't just going to kill them without a warning or a chance to prevent it - which has a strong tendency to translate into player-characters operating under the same assumption.

So the question we are forced to grapple with is "how does one balance assassination as a useful and effective tool for the PCs without making it so effective that it can never be used against them?"

Following this up with some broader-applicability to other games - I find this isn't an issue in OSR games specifically, or D&D games in general. After first the first couple levels, you generally have enough HP that a single crossbow bolt to the face isn't going to kill you, and even then in most D&D games death isn't permanent. Poisons lose a lot of their efficacy, because you can still try to make saves for them. In general, dungeon-crawling also usually means the characters are always on the lookout for something trying to kill them anyway. 

House Rule: Duel Wielding

I've never liked the idea that duel-wielding gives you twice as many attacks. Having two weapons doesn't make you attack twice as quickly, and having two attacks per round is by comparison such a good option that it's almost always better than having the +1AC or whatever that a shield would have given you.

My current compromise:

When using two weapons, roll your attack as normal. If you succeed, roll the damage of each weapon. Keep the highest.

This gives a slight bonus without being ridiculously better than a two-handed weapon or shield. I'd like to incentivize using smaller weapons with this - or at least, using something like "Sword and dagger" rather than "I'm duel-wielding longswords." I'd prefer to do it without having to resort to some kind of "one weapon has to be shorter than the other" kind of rule.

Under/Over

Continuing my LotFP tinkerfest:
My biggest bugbear with the current setup is that some rolls are roll-over, others are roll-under. This bothers me, as players get confused as to which way the rolls should go.

As it stands, attack rolls and saves are roll-over. Ability rolls and skills are roll-under. I'd like to align them all in the same direction:

Option A: Roll Over

Move everything to a roll-over system. Attack rolls can stay the same (d20+AB vs AC). Saves would be the same (equal to or greater than your save number).

Ability Score would have to be roll 1d20+Ability. 21+ is success.
Skills would have to be 1d6+skill. 7 is success.

Benefits:

• Rolling high is by virtue of language and conditioning intuitively good. "Natural 20" automatically triggers something in most gamers.
• Ability Scores can be directly opposed, with each side rolling d20+Ability to determine the winner.

Drawbacks:

• Double digit+double digit math is something that will produce an "uhhhh" every time it is rolled.
• Defeats the point of recording skills in pips, as part of the charm was the visual representation on the sheet actually looking like the d6, and needing that or under.
• 21 and 7 sound arbitrary (even if they aren't), and begin to stray into TN/DC territory, which I'm not a fan of for this kind of game. The simplicity of roll-under is that what you see is what you need.
• Save numbers go down to improve instead of going up, which is slightly counter-intuitive. 

Option B: Roll Under

Attribute and Skill rolls remain as they are now. Saves are easily converted to roll-under: you just take what they are now and subtract them from 20. A sheet could have the updated saves and no one would notice the difference.

Attack Bonus would need renamed (it's no longer a bonus), but is functionally the same. If you had +2 before, it would now just be a roll-under of 2. AC would go back to TSR era "lower is better."

Your attack roll would then be a d20  with the goal of rolling under your Attack Score+EnemyAC. 

Bonuses can universally be assumed to apply to the score itself, rather than the roll, so a +1 bonus is still good.

Benefits:

• No math required anywhere except the attack roll, but that always needs math unless you want to make a chart (Note to self: consider chart). What you see is what you need.
• Pips on dice retain their charm.
• All Stats go up to improve - except for AC? Hrm.
• AC listed in other retroclones/TSR material can be used more or less as-is. 

Drawbacks:

• "Roll low to succeed" is inherently less intuitive. "Natural 1" doesn't hit the brain the same way as "Natural 20."
• Enemy AC now modifies your roll, instead of being the target you need to meet. Not sure if this is less intuitive. Could be fixed through a THAC0 style table, but that goes back into "is this more or less intuitive than the alternative? If you did go that route though, AC could be positive as well.

At present, I'm liking B more than anything, but debating whether a return to a THAC0 style chart is better or worse than weird AC mechanics.

Classes in Lamentations of the Flame Princess

We started a new Lamentations of the Flame Princess campaign the other day. I really love the system, and it's become my go-to for OSR style play. As my players get into it though, I find the itch to tinker again. I always have the itch to tinker. Right now, my attention is on classes.

(TL;DR, skip down to "What I'm Thinking.")

What I love about LotFP Classes:

There is an extremely fine balance between the classes in the game. Each of the four human classes are good at and completely own a specific niche.

• Clerics are the only ones with access to clerical magic, which already grants them a lot of really cool stuff including the extremely important healing abilities and turn undead. 
• Fighters are the only class that gains AB as they go, as well as having access to better combat options (AC+4 on parry, Press and Defensive fighting). This is the only version of a D&D style game I've played where Fighters really feel indispensable as the fighting class.
• Magic-users get all of the eldritch goodies you'd expect, and some extremely flavorful spells.
• Specialists are the best thief variant I've seen in an OSR game. The ability to put dots where you please lets you play them as the ranger, the cat-burgler, the explorer, the assassin, or whatever else. More importantly, they are the only class who gets points to spend in increasing abilities. 

The demi-human classes are somewhat less impressive, but each feel like a hybrid of something above.

• Elves are essentially fighter-mages, as they usually are. The interesting thing is that they don't get advancing AB the way Fighters do, but they do get the extra melee options.They do however get an ascending Search die as they level.
• Dwarves are treated much the same. No advancing AB, but they get the extra melee options, a larger hit-die, and can carry around more stuff. Stonecunning takes the form of an advancing Architecture skill.
• Halflings.. uh.. are. halflings. In LotFP, they wind up with a 5-in-6 Stealth skill, an ascending Bushcraft skill, and a bonus to their dex and AC. [Edit: Raggi himself dropped a comment on my G+ for leaving out the halfling's saving throw bonuses! Woops.]

What I Hate about LotFP Classes:

Each class is indisputably the best at it's specific niche, but those niches are fairly narrowly defined. In normal B/x, it's pretty easy to add in new classes because every class already gets better at multiple things. In LotFP as written, it would be extremely hard to introduce another combat class that wouldn't step on the Fighter's toes. Giving other classes much in the way of skills winds up making the Specialist a little less.. special.

What I Want to Do:

I've got a few things I want to tinker around with but have held off because LotFP is on its own a pretty well-oiled machine. For instance - letting a character have some additional points to spend in their skills based on Intelligence modifier (Fighters wanting bushcraft is an obvious one. A dwarf in our current campaign made that exact request). I've also wanted to tinker in letting players specialize their character in one way or another - a Specialist variant that gets more offensive capabilities in exchange for reduced skill points; Paladin, Ranger, or Barbarian being Fighter variants that gain extra abilities in exchange for giving up some Fighter prowess; Variant or specialized magic users or clerics, and so on.

What I'm Thinking:

Fighter and Specialist are the two classes I'm most concerned with, as their whole value comes from being the sole class that advances in their niche. If I give another class AB, or another class skills, their relative value goes down. To solve this, I'm considering the following:

When the class in question is doing the thing they are supposed to do, they roll two dice and take the better result. Fighters would roll 2d20 on their attack rolls, keeping the higher. Specialists would get 2d6 on every skill roll, only failing if both fail.

A Paladin class then could take on Lay on Hands or a similar feature and still have advancing AB, but lose their 2d20 roll. Other classes could have skills, or an Assassin class could exist with its own features, but only get 1d6 for skill rolls. It would also make the Specialist better at whatever skill than any of the demi-humans using that same skill.

I can't decide if this is the way I want to do it. I'm not sure if it would make the classes too good by comparison, but it's definitely something I want to play around with.

Update: I have since come up with some potential solutions for all of this.

Hybrid Classes

Lateral Advancement

Multiclassing

D&D: Customization vs Complexity

The problem with rules-light games is that after a while, characters can begin to feel mechanically samey. The fewer the options you have to individualize them, the more likely Fighter dude A is going to look like Fighter dude B. The reverse issue is that the more options you have to differentiate Fighter dude A from B, the more rules you need to add to the game, the more reading that has to be done, and the more things that have to be remembered.

Let's say for the sake of argument that I have a specific idea I want to play with. I want to pattern my character after Logen Ninefingers, The Bloody Nine of Joe Abercrombie's First Law books. What are my options?

Specificity by Fluff: Core Archetypes Only
In OD&D, BECMI, Moldvay Basic and games that mimic these (Lamentations of the Flame Princess, for instance), Logen is just a fighter. My character customization options are: roll up stats, choose fighter class. This is simple, effective, and relatively fun. It's how people played for years, and stands up pretty well I think. The only problem is, if I want to play Logen as he is written, I'm either just role-playing his berzerker-rage, or the DM has to make something up for me (usually a rule stolen from another version of D&D). This is perfectly servicable, but you can't take house-rules into account for a discussion on game design.

Specificity by Diverse, Narrow Classes
In AD&D1e variants, the core classes give way to "sub classes," which become more specific. I can now play him as a berzerker class if I wanted, and that does give me more specificity and preserves the ease of learning / ease of play, but it still constrains my customization options to being stats+class (though AD&D generally gives me the option of race as well). Some people play this level with weapon proficiencies as well, which can be nice. The real rub here is that now his berzerkergang thing is the defining feature of his character, which can be a bit confusing when faced with the question of "why can't he do the things normal soldiers can do?" (after all, he has to lose something in order to avoid simply being better than the fighter.) What I gain in specificity, I lose in flexibility.

Specificity by Layering rules: Kits, Skills, and Powers
AD&D2e seems to have made an honest attempt to address all of this by introducing the much loathed Skills & Powers book, as well as making Non Weapon Proficiencies part of the core book. If that wasn't enough they also have the Complete Guides around for me to choose class kits, in addition to my pre-existing ability to choose sub-classes. Honestly, this all sounds pretty good in theory: I can now roll stats, choose class (or subclass) and race, choose weapon and non weapon proficiencies, fiddle with Skills and Powers, and tinker with all of the above via the class kits. In practice though, this can become a tangled mess. The various sub-systems don't interact well with each other, the implementation of kits is clumsy and notoriously unbalanced. On the bright side, it does address my original problem: Logen can now be a fighter with the berzerker kit and not feel like he's pidgeon-holed himself. On the down-side, it is an absolute crapshoot to deal with. The redeeming factor here however may be that all this clutter is optional. The new player sitting down for the first time can more or less roll up a dude without having to learn much of anythign about the system up front, just like in B/x.

Specificity by Integrated Abilities
D20 takes all of the above and integrates it into a unified system. For a lot of people, this was awesome. All of the customization options of 2e, but inherently meant to work together (and thus theoretically balanced). I'll not get into how well that succeeds. I never was much of a fan of the setup, but it does allow me an incredible amount of control over exactly what kind of character I want to build. The rub here is that for me to have all of those choices as a standard, non-optional part of the game as written, I need to read all of that and understand all of that. The first time player needs to read all of their class skills, all of their class abilities, all of their available feats and what those feats require as prerequisits and coordinate all of this so they can make sure they can get what they want down the line. Gone is the "roll up a character and go," but you do have more direct input over your character.

Specificity by Classless Point Buy
This doesn't apply to D&D, but it's the obvious other extreme counter-point to b/x. Abandon classes entirely, and make all skills and special abilities available on a customizable point-buy. It is the ultimate amount of customization, but also the most complicated player buy-in.

What we're seeing here is that you can't really have your cake and eat it too. The more you are allowed to tweak your character, the more complicated the system becomes. AD&D2e might have been the closest attempt to a compromise, but suffered from poor implementation. At the outset, it seems like 3e attempted simply to codify and integrate all of the ideas of 2e but in so doing it made them all mandatory, losing the flexibility and ease of play. 

The ideal system (for me, at least) would be one that featured some kind of escalating complexity. First level characters can start somewhat broad, but as they develop and earn experience points, they have some way of becoming more specified if they so choose. In a game like this, I want players to be able to get started as easily as possible. The less player buy-in required, the more likely you are to make things get going. Once everyone is moving forward, they can dip into the more advanced rules at their own leisure.

I'm kicking around some ideas on how to best make this happen, but I'm not willing to commit them to paper just yet.

Until then,

RPG Mechanics as Creative Problem Solving

When you think about it, RPG Mechanics are a bizarre concept. They are purely mental, procedural constructs that have to individually interact with one another in a very precise fashion in order to allow us to play out a story. It's like a great Rube Goldberg machine crafted entirely out of invisible pieces with the lofty goal of allowing us to go back to the same game of pretend we played instinctively throughout childhood.

I've always been fascinated by the mechanics behind role-playing games. I have a terrible habit of picking up games I know I'll almost certainly never play with the explicit purpose of seeing how they decided to build their particular machine.

Recently, I've begun really digging into Sword & Sorcery RPGs as a thing. I've always had a love for the genre, but it took a while for me to piece together why I was so drawn towards games explicitly billed as Sword & Sorcery over, say, generic fantasy games. After all, OSR games tend to cleave very close to their S&S roots without even trying. The explicit assumption is that the PCs are all adventurers in search of treasure, willing to put themselves in ridiculous danger to acquire it. Is it familiarity that has lead me to stray?

There's a broader theme, I realized. Reading through a game's mechanics isn't just about how they decided to numbers and dice to make up probabilities, or seeing if they came up with some creative and interesting innovation to make play faster/easier/more interesting. I've argued before that I believe a game's mechanics are part of the setting and define its reality. It's about the feel of the setting. After all, if the average person in your game can take 30 points of damage without suffering penalties or adverse effects, and a pistol does d6 damage, the reality of the world is that it takes an awful lot of bullets to do someone serious injury. Don't be surprised when PCs act accordingly.

A gritty crime drama or horror story should have rules as such that characters can and will die if they aren't careful. On the other hand, if you're going for a super-heroic story about crimefighters then it's totally genre appropriate for bullets to do negligable damage. Jason Statham has made a career out of seeing how much physical abuse a single action hero can take on screen.

Source

More than that though, when I look through mechanics I want to see what kind of experience they are designed to give and how they accomplish that.

There's no hiding that I was a huge The Riddle of Steel guy back in the day. It was a kind of Ah-ha! moment that really changed the way I played tabletop RPGs. Our own Band of Bastards owes a fair bit of inspiration to it.  For me, it was the definitive "game about swordsmen." It delivered that experience and then some. 

I've always held Sword & Sorcery to be one of the most evocative traditions in fiction. In digging through various Sword & Sorcery style games, I realize what I am looking for in a game is something that gives me not just the feel of an S&S story, but also the experience of one through play. It's a dragon I keep chasing, but I'm unsure if it will ever be satisfied.

One of the things I find so interesting about game design is getting to ask these questions and try to find different answers and solutions. I don't want to just create a game that has a functional mechanic. That's fairly easy: Roll <insert die or dice here> add / subtract modifier based on <skill, attribute, trait, whatever> meet over/under <number based on difficulty>. I want to read, play, and create games that do different things, that are tailored to provide specific experiences and evoke a certain feeling.

That's the magic of RPG mechanics. You aren't just looking for a way to resolve task probabilities, it's an exercise in creative problem solving. It's the purest form of thought experiment. You're building a virtual reality machine. You're dissecting what is at the heart of a particular genre, and then trying to come up with innovative and creative ways to both replicate the internal logic and feel of that world and provide an experience that brings playing that world to life.

This test is actually the reason I have a hard time getting into generic systems: d20, GURPS, Savage Worlds. Don't get me wrong, I played a great deal of Savage Worlds a while back. It's great for quick games, pick-up games, pulp games. If anything, I'd argue that it's actually tailored specifically to be a pulp game, but anything outside of that specific genre I'd generally use something else.

If you're at all interested in game design, I really urge you: read and play as many different games as possible. Take note of what they do and how they do it, and ask yourself how those decisions impacted your play. Any time you play a game that pitches itself to a specific genre, keep an eye out for what the game does and ask yourself how the mechanics help evoke the feel and experience of that genre. It's a great habit on the way to sharpening your own design skills.