Random vs Non-Random Char-Gen

Best Selling RPGs - Available Now @ DriveThruRPG.com
I like mostly-random character generation because I can start with an archetypal or generic character and get to know them through play, and because it takes me out of my comfort zone by making me adapt to something unfamiliar. With non-random character generation I feel like I have to know everything about the character before play starts (their background, their personality, all of their quirks and distinctive bits) which is not how I like to play, plus it usually means that I end up falling back on one of a handful of "stock" characters, which creates extra frustration because the system usually doesn't give enough points to create the character as-envisioned (or requires deep system familiarity to know how to do it - including pre-planning an advancement "build") so you have to compromise and create a character that's not quite what you wanted to play, which to me feels like sort of the worst of both worlds.
This tends to be one of the big issues with point buy. In part I think this is more an issue in practice than it is in principle (not entirely, but I think mostly).

For some reason there's a distinct tendency to be parsimonious in the points given. In part I think this comes due to long standing assumptions about gaming. The idea that you need to protect niches*, and you need to leave space for development. However, it often means neutering the strengths of the very approach to character generation you have chosen to use.

*Which in itself I think comes from a lot of unstated and unexamined assumptions about gaming.
 
Last edited:
This tends to be one of the big issues with point buy. In part I think this is more an issue in practice than it is in principle (not entirely, but I think mostly).

For some reason there's a distinct tendency to be parsimonious in the points given. In part I think this comes due to long standing assumptions about gaming. The idea that you need to protect niches*, and you need to leave space for development. However, it often means neutering the strengths of the very approach to character generation you have chosen to use.

*Which in itself I think comes from a lot of unstated and unexamined assumptions about gaming.

I admittedly tend to fall back into the "room for growth' assumption a lot, and think that there's merit to it. But most point buy games take it too far, and barely give you enough points to build a character that's competent in just one narrow area, with no room for spreading out into different fields or covering side skills or background stuff.
 
I’m kind of a toss up. Sometimes I like random, sometimes I like point buy. But I want to note, even why I do random it is rarely “straight 3d6 in order”.

We’ve always, in the 3+ decades that I’ve played, done something like 4d6 minus lowest place them where you want. In V&V, we used random but made some allowances that stopped Flying Amoeba IceFire boy! Unless you wanted to play him.

Part of this is because it normally does suck to play low stat man in a party of “all 14+ on 3d6 really?” guys. Then someone says “but the low stats give your guys character” and that can be true. But if the game system punishes you for the low stats, you don’t really want the “character” you get from bad rolls to be “loses at all the rolls”.
 
With the original books, the skill list is so short, it shouldn't take too much to make almost any skill useful. It also helps if the skills are not treated such that not having a skill makes it hard to do anything. That marine with Vacc Suit has an important skill that most characters won't have. In my game, Vacc Suit is also useful for avoiding penalties in zero-G. Tactics has direct input to the encounter and combat system. I've played Army one-term wonders with 6 skills which is pretty cool and leaves potential benefit from being so young.
Having ex-Army characters with only one term under their belt never bothered me - they were guaranteed at least one decent combat skill, and the Traveller we played always had a fight at some point so Mr. ex-Army was always going to be useful at some point. Single term Navy was much more of a problem - not too unlikely a result, and no guarantee of anything useful. This was one reason we played a lot more Space Opera despite it's convoluted chargen and slow combat - you had a great deal of control over your character's skills.
 
This tends to be one of the big issues with point buy. In part I think this is more an issue in practice than it is in principle (not entirely, but I think mostly).

For some reason there's a distinct tendency to be parsimonious in the points given. In part I think this comes due to long standing assumptions about gaming. The idea that you need to protect niches*, and you need to leave space for development. However, it often means neutering the strengths of the very approach to character generation you have chosen to use.

*Which in itself I think comes from a lot of unstated and unexamined assumptions about gaming.
Or it comes from a cynical view of players derived from many years of running games for them. I've a player that makes me wish I'd built some strong niche protection into my current game or something other than GURPS to run it in.
 
I’m kind of a toss up. Sometimes I like random, sometimes I like point buy. But I want to note, even why I do random it is rarely “straight 3d6 in order”.

We’ve always, in the 3+ decades that I’ve played, done something like 4d6 minus lowest place them where you want. In V&V, we used random but made some allowances that stopped Flying Amoeba IceFire boy! Unless you wanted to play him.

Part of this is because it normally does suck to play low stat man in a party of “all 14+ on 3d6 really?” guys. Then someone says “but the low stats give your guys character” and that can be true. But if the game system punishes you for the low stats, you don’t really want the “character” you get from bad rolls to be “loses at all the rolls”.
When I hear things like that it drives me nuts. Characters (and people in real life) do not need to be flawed to be interesting, and they certainly don't need mechanical flaws in their core competencies (and let's face it, in a game like D&D about any time your character is rolling, it's to do with your character's chosen profession of adventuring).
 
When I hear things like that it drives me nuts. Characters (and people in real life) do not need to be flawed to be interesting, and they certainly don't need mechanical flaws in their core competencies (and let's face it, in a game like D&D about any time your character is rolling, it's to do with your character's chosen profession of adventuring).
Of course they don’t but they also don’t have to be uber competent to be interesting.
 
I think with D&D this is one of the fundamental differences between Gygaxian and post Gygaxian D&D. It does creep in there, for example a low INT Magic User in AD&D will be at a disadvantage learning spells, but aside from that in the earlier editions low attributes aren’t a game stopper.

All the more reason I’m glad I left D&D behind in the 80’s. I’ll play AD&D 1E now and again out of nostalgia and will play OSE or Pathfinder whatever if that’s what my friends want to run because I’m a social gamer but they aren’t the games I like to play.
 
...

Part of this is because it normally does suck to play low stat man in a party of “all 14+ on 3d6 really?” guys. Then someone says “but the low stats give your guys character” and that can be true. But if the game system punishes you for the low stats, you don’t really want the “character” you get from bad rolls to be “loses at all the rolls”.

Low stats can sometimes be interesting when your character has like a single low stat s/he's known for--like a low Int nitwit, or a sickly low Con character. But when all scores mediocre or low, without a single shining stat, your character just sucks--specially in a party full of uber stat characters, with multiple high scores.
 
That has never been my experience, usually there is one player that got lucky and rolled lights out but most randomly generated characters are in the mid range with some good and some bad attributes.

edit to add again I’m talking Gygaxian D&D such as AD&D 1E, BECMI and B/X not the later versions where apparently you need all high stats to be able to have fun.
 
An issue I've run into with point buy is where you start having players "top" each other in an arms race for strongest, quickest, best xyz. Went through a period with HERO where the PCs were becoming very ineffective one trick ponies putting massive amounts into one area as the players vie for top spot. That eventually sort of fixed itself with that group as the parties were simply not functional, but it tends to happen when a new group gets together or a new genre is explored. Sometimes it self corrects, sometimes everything goes down in flames.

This is why most modern incarnations of point build systems usually have a secondary capping mechanism; it doesn't entirely prevent people trying to end run around the cap, but it at least tells people "If Attack Value X isn't enough for you, we're having a campaign mismatch here".

The other side of that is you tend to see a lot of generalists making the PCs all kind of the same. In fact it can take quite a bit of system knowledge to make a capable PC who is effective, and not a bland cookie cutter template.

Neither of these are usually long term game killers (hopefully players learn), but it can be a major turn off and does provide some appreciation for random systems letting the dice decide who is the strongest, fastest etc.

This is my experience as somebody to generally prefers point buy.

You're not wrong, but that's usually a case of people either not knowing what they want, or not paying attention to the fact they aren't the only player in the campaign.
 
Sure, that ongoing value of a higher INT may make a difference, though it doesn't come much into play until you have pretty high skills. Ans yes, rolling an 8 INT would be bad. Really if someone managed to roll an 8 INT, I'd just have them re-roll these days. Optionally I could make INT 2d6+7 (a 19 INT really is just 1 point better than an 18). I think I'll just tell anyone who rolls an 8 to re-roll...

But notice how you're having to progressively iterate away the gaps here.

Making different weapons viable is very tricky. Some systems make all weapons equal, but then there's no actual differentiation. In RQ1, a bastard sword does 1d10 vs broadsword at 1d8+1 so not much incentive to change to something with a low base skill. In my game, spears are very useful which at least makes one alternate weapon. We also have an axe fighter who is doing pretty well.

The default in RQ2 at least was one handed bastardsword was 1D10+1 to 1D8+1 for broadsword, so there was reasons to try and move over as opportunity permitted (keeping in mind that a noble could walk in with one as easily as the other, and people working their way up from non-combat backgrounds it was about as easy to do one as the other if you weren't going to go for the early base bonus and carry a mace). Spears always benefited/suffered from being impaling weapons, which, given the rules on same, meant no one tended to use them as a primary weapon.

Well, if non-combat stuff isn't meaningful in the campaign, then yes, everyone has to have a good combat role.

Its not lack of meaning (things like Lores can absolutely be pretty meaningful, as can things like First Aid) but that they're sort of one-and-done (you make one roll in the skill and it either gives you what you need or it doesn't) and when they aren't there's still not much engagement with them (you make a lot of decisions in a combat with a system with anything better than a hit-and-do-damage combat system, but with a lot of other skills you're just, well, applying them). And of course when a combat starts unless you can just hide, everyone is going to participate anyway, while with some other skills there's often an incentive to keep everyone not capable (or even less capable) out of the process.

I get you are hard pressed to see value in randomization, and that's fine. To me it definitely adds value.

My point was, what happens to someone in a game that forces randomization who doesn't get value out of it?

With the original books, the skill list is so short, it shouldn't take too much to make almost any skill useful. It also helps if the skills are not treated such that not having a skill makes it hard to do anything. That marine with Vacc Suit has an important skill that most characters won't have. In my game, Vacc Suit is also useful for avoiding penalties in zero-G. Tactics has direct input to the encounter and combat system. I've played Army one-term wonders with 6 skills which is pretty cool and leaves potential benefit from being so young.

Yeah, but how regularly does the Vacc Suit skill need to be used? Its obviously very useful in a few situations, but you can go a long time without that being the case (and some of the potential use-cases aren't really good for having only one guy good at it). Tactics does have a positive benefit in combat of course, but how necessary is it? Its just a case of it not being clear why they're even dragging this guy around; he has a couple of specialty skills that, unless they're planning to do a lot of work on outside work on inhospitable planets or regularly do combat, just don't pay for themselves the way a lot of others do.
 
So I didn't get very far, but let's say you pair STR/INT, DEX/WIS, CON/CHA (I really don't think the standard D&D attributes work well with this idea). You then roll 3d6 for each of STR, DEX, and CON. Then INT = 21 - STR, WIS = 21 - DEX, CHA = 21 - CON. So if you rolled 12 STR, 16 DEX, and 8 CON, you would get 9 INT, 5 WIS, and 13 CHA. The trick is first you need attributes that can be truly opposed and second, you need to make it just as meaningful to have 10 STR, 11 INT as 3 STR, 18 INT as 18 STR, 3 INT.

Ah, I understand. The only thing I'd think here is you probably don't really want to do that with attributes that have too strong an impact either; I'd wince at using that in Mythras because for all that, say, Dex and Int are both valuable (and they both really are), I'd not want to have one at 4 because I had rolled a 17 on the other.
 
I can of course only speak from my experience where most people seem able to engage with the character they wind up with. I've made some minor adjustments at chargen time, and certainly suggested someone re-roll a few times (more often in my OD&D play by post).

Well, I gave an set of examples in the city/country focused campaigns in my earlier exchange with you, but its hard to say if you ever have run campaigns where its an issue. I absolutely have.

Ah, yes, I see in RQ1 your chance of improvement is increased 3% for every point of INT over 12. Somehow I don't use that an instead, INT is your minimum chance of improvement. But I do allow you to add your ability bonus (where an 18 INT nets you +10%, less than the +18% from RAW...). Yea, that rule as written is a serious double dip.

Was that how it worked? I'm remembering one version where you added your INT as a value to the roll to advance (which, in practice, acted like your skill was -INT for purposes of advancement). But I've played so many BRP derivatives over the years I could be conflating it with something else.

Yea, but that impacts build as well as random. And at least in RQ, skills are mostly player choice.

Sure. That was mostly commenting on skills you have no choice about in some systems (though you can still have the issue where all your attributes mostly contribute to skills that aren't too important to the campaign at hand in some systems, too).

Sure, those are problems. One cool thing about RQ is you get a lot of skill development in play, which of course is guided by what you actually do (experience) but preparation (training) also plays a major role.

Yeah, but that doesn't make starting out bad in the relevant skills and good in the mostly irrelevant feel good. As you say that's usually not an issue in RQ (though the wilderness campaign where your best attributes are POW and CHR pushing you to be good at things like Oratory, Fast Talk and Debate can make you feel like the dice thought you were in a different campaign).

Traveller certainly is more likely to cause problems with useful skills, and I agree that there is an issue there. I've played ground pounders in a more space oriented campaign and that can be a bummer (most everyone has a chance of getting gun combat at least, and failing that you can try the self improvement system to get some gun combat). But that's where the roll 2 or 3 characters and pick one would help a lot (and if you manage to roll 3 army types, well, roll another one or two...). In other words, my thought for injecting more player choice in Traveller is to just generate a stable of characters. Or use the online generator and use the options to instruct it to make your choice of Scout, Merchant, or Navy if you want a ship skilled character.

Well, Traveler is admittedly an outlier in that it can end up pushing you into skills that aren't really relevant to the game at hand. Even games that demand you roll attributes don't usually do that (though as I noted, in games with strong attribute contribution, it can still make a situation where all the things you'd be really good at are things that are minor factors in the campaign).

I guess in the end, I agree with you, some level of player direction to the type of character they get is important. But I like that to be tempered with some randomness. With any random based system, you can always temper the randomness with player direction simply by generating several characters and picking one. As others have said, it can be hard to inject randomness into a build system (though I have randomized which attributes get which values from an array).

I still kind of argue against that; its not that hard to do some randomness within attributes withing a point distribution system. You just, well, randomize the point distribution. If you have 60 attribute points to spend, no one cares if you decide on how those are distributed or roll some dice to determine so. You might end up not wanting to go too low anyway, but that just means you randomize, say, 30 of the 60. With a full point distribution system (that is to say one that takes everything from a single pool rather than, say, having separate pools for attributes, skills and talents, which isn't particularly uncommon) you can still do it, though you might need to make the randomization a little more complicated (since you probably want a minimum amount in skills and a minimum amount in attributes), but again, no one cares how you do that other than whatever degree of expectation they have for you to have a functional character at the end.
 
It's not so much that I judged them to be inflexible... I just came to recognize our differences of tastes and interests were pretty wide. But I also learned not to join groups if I can discern those sort of differences exist.

Yeah, I'm not sad I haven't had to get into a gaming group blind in decades.

Right now there's a group I'm in that I want to quit... and I probably won't bother to explain it to them either. It's not them, it's me.

Well, its not like its your fault either; you just have a mismatch.
 
This tends to be one of the big issues with point buy. In part I think this is more an issue in practice than it is in principle (not entirely, but I think mostly).

For some reason there's a distinct tendency to be parsimonious in the points given. In part I think this comes due to long standing assumptions about gaming. The idea that you need to protect niches*, and you need to leave space for development. However, it often means neutering the strengths of the very approach to character generation you have chosen to use.

*Which in itself I think comes from a lot of unstated and unexamined assumptions about gaming.

Its probably related to the zero-to-hero compulsion you see in a lot of games too (particularly in the D&D sphere).
 
I'm actually a "middle way" when it comes to this. I have come to loathe highly randomized methods that create a disparity in PC power and do nothing to give the players ideas. At the same time, point buy methods that require the player to game the system to make an effective character or struggle to get what they want (Pathfinder 1e, D&D 3.x, I'm looking at you) also infuriate me.

I appreciate methods that use randomness as a tool for inspiration (which I call "Dice as a Muse"). This is where Sentinel Comics RPG falls. Old school Traveller has a similar vibe, but definitely falls into the trap of creating a disparity in PC ability.

I also appreciate the more randomized creation sometimes forces players out of the mindset of creating the same character every time they sit down.
 
Traveller is quite extreme in its randomness out of the box. I experimented with doing hybrid random/choice based systems where the player rolled some of the skills and chose the others, and added some background skills so you could do a one or two-term character without being horribly disadvantaged.
 
MegaTraveller doesn’t get much love (for plenty of mostly-valid reasons) but I liked the way it maintained the core of Classic’s random char-gen but added more elements of choice in the form of “cascade” skills where instead of the table roll giving you a specific skill it would often give a “cascade” where you would then pick one from a list of usually 3 or 4 related skills.

Between that and a couple more things I house ruled in from later editions (allowing s couple points of “background skills” not related to career, and allowing higher education options for all characters, not just “advanced” ones) felt like I’d gotten to a nice sweet spot between random and choice-based char-gen in the mid-90s.
 
But notice how you're having to progressively iterate away the gaps here.
True. The truth with RQ is I really like the game and there are things I don't want to try and change about it. Because of the way attributes work in the system, point buy doesn't work. So I use random generation, with a bit of GM oversight to protect against crappy characters. It works and I have plenty of players who enjoy the game.

The default in RQ2 at least was one handed bastardsword was 1D10+1 to 1D8+1 for broadsword, so there was reasons to try and move over as opportunity permitted (keeping in mind that a noble could walk in with one as easily as the other, and people working their way up from non-combat backgrounds it was about as easy to do one as the other if you weren't going to go for the early base bonus and carry a mace). Spears always benefited/suffered from being impaling weapons, which, given the rules on same, meant no one tended to use them as a primary weapon.
Yea, so the weapons table changes are one reason I don't like RQ2... On the other hand, I really don't have a problem with stronger PCs doing more damage (though I do appreciate that Cold Iron avoids the double dipping, while there is a damage bonus for high strength, it's for having a significantly higher strength that your weapon requires).

Its not lack of meaning (things like Lores can absolutely be pretty meaningful, as can things like First Aid) but that they're sort of one-and-done (you make one roll in the skill and it either gives you what you need or it doesn't) and when they aren't there's still not much engagement with them (you make a lot of decisions in a combat with a system with anything better than a hit-and-do-damage combat system, but with a lot of other skills you're just, well, applying them). And of course when a combat starts unless you can just hide, everyone is going to participate anyway, while with some other skills there's often an incentive to keep everyone not capable (or even less capable) out of the process.
OK, so I DO have one player who is unhappy with their ability to contribute to combat, and doesn't feel like their other capabilities are much use. But overall that has not been an issue with my RQ campaign. Yes, combat does take a significant amount of time, but one nice thing about RQ is everyone can do magic. If you don't have the stats to be a front line fighter, start using some magic. But this is a role balance issue, not a random vs. build issue. And yes, if the game is only rewarding if your character is good in combat, the generation system better make everyone good in combat...

My point was, what happens to someone in a game that forces randomization who doesn't get value out of it?
Don't play in my game, play in another. I run my games on Roll20 and draw from across the US (and sometimes across the world - I've had players from the UK, Australia, and New Zealand). There are plenty of other games being run out there.

Yeah, but how regularly does the Vacc Suit skill need to be used? Its obviously very useful in a few situations, but you can go a long time without that being the case (and some of the potential use-cases aren't really good for having only one guy good at it). Tactics does have a positive benefit in combat of course, but how necessary is it? Its just a case of it not being clear why they're even dragging this guy around; he has a couple of specialty skills that, unless they're planning to do a lot of work on outside work on inhospitable planets or regularly do combat, just don't pay for themselves the way a lot of others do.
So Tactics gives +1 on surprise rolls (military also gives +1 - so that Marine with Tactics has +2 on surprise rolls - that's major). Zero-g could be very relevant. Also, emergencies without Vacc Suit skill are almost sure to be disastrous (-4 for no skill...). BTW, in Traveller, if you have a decent DEX you can get +1 or +2 with a weapon. One of the best combatants in my last Traveller campaign was the Doctor because he got +2 with Laser Rifle... And again, don't have a combat skill? You start a self improvement program, roll 8+ on 2d6 and immediately get level 1 in two combat skills. After 4 years, they are permanent. The roll was a bit easier in my game, I wanted to make training a bit more effective so the +1 or +2 you are eligible for using the weapon is added to the roll for the 8+ for self improvement.

But also - my Traveller campaign had combat in less than half the game sessions... And skills were less important that might seem, a lot of things were resolved through role play not skill rolls.
 
Ah, I understand. The only thing I'd think here is you probably don't really want to do that with attributes that have too strong an impact either; I'd wince at using that in Mythras because for all that, say, Dex and Int are both valuable (and they both really are), I'd not want to have one at 4 because I had rolled a 17 on the other.
Oh, absolutely. I actually would make the attributes not traditional attributes, but more like some kind of role. Maybe one aspect is fighter vs. magic user. And if you conceive it right, the middle is interesting too. Of course you are not playing a random character concept for sure...
 
Well, I gave an set of examples in the city/country focused campaigns in my earlier exchange with you, but its hard to say if you ever have run campaigns where its an issue. I absolutely have.
I have had issues in the past. I've had problems myself as a player (I've commented on those - and those inform how I run my games). I've probably even lost some folks from my current campaigns because of issues. But I know I have plenty of players in my current campaigns who are just fine with playing what the dice gave them.

Was that how it worked? I'm remembering one version where you added your INT as a value to the roll to advance (which, in practice, acted like your skill was -INT for purposes of advancement). But I've played so many BRP derivatives over the years I could be conflating it with something else.
My memory was faulty, so I looked up the actual rule in RQ1 (and I just confirmed the rule is the same in RQ2). It's probably different for RQ3 and different for various other BRP games. But whether it's +INT or +(INT - 12) * 3 there's a serious double dip and yes INT is super important.

Sure. That was mostly commenting on skills you have no choice about in some systems (though you can still have the issue where all your attributes mostly contribute to skills that aren't too important to the campaign at hand in some systems, too).
The only game I play in this vein is Traveller. And ultimately I have a love-hate relationship with SF (or really any modern) game, partially because of skills. It wasn't too much of an issue in my recent Classic Traveller gaming. But I also mitigate by allowing players to generate several PCs and pick one they would like to run. If they really couldn't find one they liked, I'd have them keep rolling. Or we would switch to the online generator and I'd ask them one thing (skill or something else) they really wanted, and use the directives the online generator has to generate a PC with that one thing.

Yeah, but that doesn't make starting out bad in the relevant skills and good in the mostly irrelevant feel good. As you say that's usually not an issue in RQ (though the wilderness campaign where your best attributes are POW and CHR pushing you to be good at things like Oratory, Fast Talk and Debate can make you feel like the dice thought you were in a different campaign).
Well, RQ1 only has Oratory in the "Charisma" skills department... (I've added several more). They PCs have used Oratory more often in the wilderness than in the city... But I will grant that there are systems that produce characters that will be of limited value outside of certain places. And yes, some ability to direct character rather than randomize the character is important to keep that from happening. And sure, that could be a serious issue with Traveller. Again, my work around to continue to use random vs. build is generate several characters and pick one. Also, the players and GM should be working together to make sure the skills that PCs have are useful. If I was running an exclusively city or exclusively wilderness campaign, I would also direct players to not make choices that render their character useless. And again, I'd do something to keep randomization from ruining a player's experience.

Well, Traveler is admittedly an outlier in that it can end up pushing you into skills that aren't really relevant to the game at hand. Even games that demand you roll attributes don't usually do that (though as I noted, in games with strong attribute contribution, it can still make a situation where all the things you'd be really good at are things that are minor factors in the campaign).
Yes, Traveller is pretty extreme in this way. Following some of the guidance folks have posted on the web (see my Traveller reading list for some blogs, particularly Tales to Astound - https://ffilz.github.io/Gaming/traveller-reading.html for some of these thoughts) can be used to make it much less of an issue with Classic Traveller. And then rolling up a few characters and picking one really does help the rest.

I still kind of argue against that; its not that hard to do some randomness within attributes withing a point distribution system. You just, well, randomize the point distribution. If you have 60 attribute points to spend, no one cares if you decide on how those are distributed or roll some dice to determine so. You might end up not wanting to go too low anyway, but that just means you randomize, say, 30 of the 60. With a full point distribution system (that is to say one that takes everything from a single pool rather than, say, having separate pools for attributes, skills and talents, which isn't particularly uncommon) you can still do it, though you might need to make the randomization a little more complicated (since you probably want a minimum amount in skills and a minimum amount in attributes), but again, no one cares how you do that other than whatever degree of expectation they have for you to have a functional character at the end.
OK, I think you're arguing against the "roll several characters and pick one." That has worked out well in Traveller games I've been GM or player in. In an AD&D play by post, the GM gave us each a list of like 10 sets of attribute rolls. I picked one that looked interesting.

But I get that some folks want much more control over their character and that's just fine. I have come to dislike a lot of what I saw coming out of character build systems and thus I'm less enthralled with them. Burning Wheel character build though is cool. Bushido's is OK, though VERY break point prone. But D&D 3.x? Hero? GURPS? Those I've lost patience with. I don't have much experience with much else (other sometime in the 1980s, my Traveller campaign then switched to characters getting an array of skill levels and then they picked skills rather than random skill determination - but that game is so distant to what my gaming experience is today that it's not worth much in informing my current gaming.
 
MegaTraveller doesn’t get much love (for plenty of mostly-valid reasons) but I liked the way it maintained the core of Classic’s random char-gen but added more elements of choice in the form of “cascade” skills where instead of the table roll giving you a specific skill it would often give a “cascade” where you would then pick one from a list of usually 3 or 4 related skills.

Between that and a couple more things I house ruled in from later editions (allowing s couple points of “background skills” not related to career, and allowing higher education options for all characters, not just “advanced” ones) felt like I’d gotten to a nice sweet spot between random and choice-based char-gen in the mid-90s.
I've come to not like the extensive skill lists of later Traveller (Book 4 and beyond...) but yes, Traveller is very amenable to rules hacking that moderates the randomness. Another simple one is make your 1d6 roll then pick the table giving you a choice of 3 or 4 improvements (one of which may be an attribute improvement rather than a skill).
 
Interesting podcast by two creators in the “indie new OSR” scene, I don’t know what labels mean what anymore to be honest. Any way at 38:30 minutes in they launch a discussion of character creation and the idea of random vs choosing. It is coming from the indie side not traditional games.
 
Or it comes from a cynical view of players derived from many years of running games for them. I've a player that makes me wish I'd built some strong niche protection into my current game or something other than GURPS to run it in.
That's somewhat circular though. You only need to protect niche protection if you assume you need it.

(And sometimes you may want it, but it's very rarely designed for very well anyway).
 
Last edited:
I don't do random chargen, period. I like certain character types.
Random doesn't preclude this. Roll 4d6k3 6 times and arrange as you choose allows a lot of choice in character type. Working with the GM to guide the random generation to get the character type you want works also.

Take a look back through the thread. Many, including me, have offered ways that allow choice of character type meshed with random generation.
 
Random doesn't preclude this. Roll 4d6k3 6 times and arrange as you choose allows a lot of choice in character type. Working with the GM to guide the random generation to get the character type you want works also.

Take a look back through the thread. Many, including me, have offered ways that allow choice of character type meshed with random generation.
Yes, but this doesn't really allow much useful randomness as I see it.

Mostly, you end up with pretty much the same characters as point buy but with a bigger spread of overall ability. If you roll down the line you least get certain characters come up that usually wouldn't if ability scores were being assigned.

Now sometimes this has certain benefits in increasing the range of possible characters if the player rolls particularly well* - but that's mostly to do with the issue of overparsimonious point buy budgets, and that tends to go away if you fix that issue.

*Eg you might occasionally roll well enough that you can put an extra high score in Intelligence without reducing the effectiveness of a Fighter overall and play a very smart Fighter - something which pretty much doesn't exist in point buy games - but this is a rare and unlikely result.
 
Yes, but this doesn't really allow much useful randomness as I see it.

Mostly, you end up with pretty much the same characters as point buy but with a bigger spread of overall ability. If you roll down the line you least get certain characters come up that usually wouldn't if ability scores were being assigned.

Now sometimes this has certain benefits in increasing the range of possible characters if the player rolls particularly well* - but that's mostly to do with the issue of overparsimonious point buy budgets, and that tends to go away if you fix that issue.

*Eg you might occasionally roll well enough that you can put an extra high score in Intelligence without reducing the effectiveness of a Fighter overall and play a very smart Fighter - something which pretty much doesn't exist in point buy games - but this is a rare and unlikely result.
These days I don't tend to do roll and arrange any more. I do allow roll and swap for Cold Iron. But roll and arrange doesn't necessarily produce the same results as point buy. With point buy I might not pick any really low attributes, or I might not pick one 18 and the rest middling (depending on how point buy works). RQ has to be roll down the line for me because with my use of 2d6+6 INT for humans, no race has 3d6 down the line.

It can be tricky to provide the right point buy budget. Actually, Bushido I think WOULD have a bit more variation in characters with a bigger attribute budget. One the other hand, if your attribute budget is enough to buy an 18, a couple 16s and nothing else below 12, then it may lead to a feeling of all characters being super characters.

But actually one of the things the whole business of attribute buying or rolling brings up is whether the way attributes are done is REALLY a good way to define characters. Too many systems don't actually make different arrangements equally interesting to play. If I was to take a hard look at Cold Iron and aim it at character building rather than rolling randomly, I might actually radically change how attributes are generated. But the question then is does that remove some of what actually is stuff I like about the system? The same could actually be said about RQ.

Burning Wheel has relatively little attribute variation among characters. In my experience, most characters have 3 or 4 in each of the attributes with much more of the character definition in the skills and traits.

But I have yet to see a system without attributes that really appeals to me. But then I also haven't actually seen that many such systems... One problem is that it seems that along with dispensing with the "modelling of something" by using attributes, the systems without (traditional) attributes also dispense with other "modelling" and become such abstract systems they no longer appeal to me.
 
That last post got me thinking how my feeling about random vs. build goes, or really what degree of randomness in chargen do I like...

Ultimately what it comes down to is that after my gaming experiences of the 2000s (give or take), I have returned to the games that most appealed to me in the 80s. Classic Traveller, RuneQuest, and Cold Iron, with a dose of pre-2.0 D&D. Bushido has entered my stable. And I guess there's a few others I would play, but certainly for running, it's Classic Traveller, RuneQuest, and Cold Iron. Those three games work best with some level of random generation. Now I get that telling someone to make the rolls exactly as defined in the rules, and play the character so generated is not appealing to many. So I apply various things. I allow rolling more than one character and picking. I tell people to toss characters with really bad attribute rolls. In games other than Traveller I have players roll an extra die for each attribute and drop the lowest (or pick - SOMETIMES it actually makes sense to pick not the highest). I allow a swap or two sometimes. I bump an attribute up. Whatever. I fudge somewhere if it helps a player get a character they are satisfied with. But no one ends up with exactly the same character because I'm tweaking, not overriding. This works for me as a GM. It works for enough players that my RQ campaigns are both full, and if someone drops out, I'll fill the slot soon enough. I've got players I enjoy gaming with. My games aren't for everyone. But that's OK.

Now I can also happily discuss ideas on how to do random better or differently or merits of various build systems.

And someday I will at least take another stab at Burning Wheel and completely embrace it's totally non-random generation.

I will probably never again run GURPS or Hero, and I probably wouldn't play either. My play experience with TFT is enough exposure to that system. I'd probably play L5R again, at least with the GM I played the one campaign with. I'll happily continue to play Bushido, but at this point, I'd rather keep at Cold Iron Samurai Adventures than try running Bushido again.
 
But actually one of the things the whole business of attribute buying or rolling brings up is whether the way attributes are done is REALLY a good way to define characters. Too many systems don't actually make different arrangements equally interesting to play. If I was to take a hard look at Cold Iron and aim it at character building rather than rolling randomly, I might actually radically change how attributes are generated. But the question then is does that remove some of what actually is stuff I like about the system? The same could actually be said about RQ.
No I think rolling attributes is simply not the best point of randomisation. It's better to randomly determine something else and then use that to determine attributes. That way there is more useful information given by the randomness (at the very least).

That's why I think lifepath systems work better for random character generation.
 
It can be tricky to provide the right point buy budget. Actually, Bushido I think WOULD have a bit more variation in characters with a bigger attribute budget. One the other hand, if your attribute budget is enough to buy an 18, a couple 16s and nothing else below 12, then it may lead to a feeling of all characters being super characters.
It depends on the game, but generally I think want you want to do is cap the high end* But allow for the lower end to spread more widely. A lot of them suffer from not just thinking clearly about the desired range of results. One of the frequent issues is to try and budget the point buy so to intrinsically do the former which ends up shortchanging the latter.

*Unless the game is perfectly fine with one PC being the best swordsman in the world from the get go - which some games may be.
 
Yeah, I'm not sad I haven't had to get into a gaming group blind in decades.

I think this flavors perceptions far greater than many understand. I envy people who have rarely had to hunt for a gaming group. Beyond high school I have rarely had such a luxury. I've had a few really solid groups, but life issues made them fairly short lived (employment can be such a bother).
 
It depends on the game, but generally I think want you want to do is cap the high end* But allow for the lower end to spread more widely. A lot of them suffer from not just thinking clearly about the desired range of results. One of the frequent issues is to try and budget the point buy so to intrinsically do the former which ends up shortchanging the latter.

*Unless the game is perfectly fine with one PC being the best swordsman in the world from the get go - which some games may be.
What kind of cap and spread are you thinking?

One thing I would like to figure out is how to really make STR fighter and DEX fighter work. RQ does have some significant advantage to DEX fighter (and they can be awesome archers with two shots per round and you just need enough STR to use a composite bow). Cold Iron doesn't work great for DEX fighter since you lose damage, you're better off having your points in STR but there ARE some ways to make a high DEX low STR character viable using a spear so you can avoid being in the front line, but then you ALSO want a low CON so you can be small (so your dodge is maximized). I had a GMPC in my first Cold Iron campaign that had a serious glass jaw, but enough STR to do reasonable damage with a spear, and a kick-ass DEX. Her dodge was amazing.

One of the problems with point buy though for Cold Iron is that actually having a character that can be a good fighter AND a good caster is hard, because then you want almost all the attributes to be good. And if you are mediocre on any of the attributes you end up not being a good fighter OR a good caster. Random chargen of course can give the occasional character who lucks out, and the way attribute improvement works and the way I handle skills keeps that awesome-sauce character from being among the best fighters. And a fighter-caster is going to take a different spell load than a pure caster so pure casters still get to feel like they have a place.

One way to do build in Cold Iron would actually be for me to lay out some attribute distributions that create desirable characters of different types, but aren't balanced by any point buy system. So I could lay out a distribution for a fighting cleric that leaves the character decent in a fight, but the pure fighter is better. I could also lay out STR and DEX fighters that would feel equally effective. Maybe that's what I should look into. And then maybe add a random "benefits" table that gave some random minor bumps to attributes (like 1 point here or there) or other minor benefits that mesh with the system. Then two characters who took the same distribution could be different even if they chose the same weapons or cleric cult or both were magic users.

The same could probably be done for RQ also.
 
What kind of cap and spread are you thinking?
Depends on the game and what it's trying to do. The cap depends on how good you want a starting character to be able to be, and how much, if any, room needs to be left for growth.

I'm thinking more skills based here - I think part of the problem with heavily attribute based systems is how much they tend in practice to limit certain character types. The trade offs are just too blunt. Random rolling can remove those trade-offs sometimes, but I don't think it's particularly a good solution because it only does so if you roll well enough to be ahead of the curve.

One of the issues is the over-weighting of combat abilities in many games that depend heavily on attributes.
 
Last edited:
That's somewhat circular though. You only need to protect niche protection if you assume you need it.

(And sometimes you may want it, but it's very rarely designed for very well anyway).
In this case lack of niche protection means constantly having to rein in that particular player when he tries to join other players in the spotlight by having his character 'help', and due to lack of mechanical spotlight protection there's often no mechanical reason why the character shouldn't. Yes, this is technically a social, out of game, problem, but with stronger character niches it wouldn't come up as much, and that would be easier on me and make running the game less tiring, and maybe allow me more energy for improving the quality of my GMing in other ways.
 
One of the issues is the over-weighting of combat abilities in many games that depend heavily on attributes.
I think there's a fine line here, because if they aren't important (presumably with skills or dis/advantages taking over), you can have things like big, strong guys not being scary in brawls, and for a lot of people that's quite immersion-breaking.
 
Banner: The best cosmic horror & Cthulhu Mythos @ DriveThruRPG.com
Back
Top