Random vs Non-Random Char-Gen

Best Selling RPGs - Available Now @ DriveThruRPG.com
Not a clue. I can't run a game for the "play to find out" crowd, so my pool of anecdotes is severely biased. Some randomness combined with choices and some control have worked for us, but full randomness would most likely fall very flat.

I also think "play to find out" probably work a lot better with GM's that have a planned adventure, as the finding out can be used to give motivation to work toward the goal of the adventure. While extensive backstory most likely work better with character driven sandboxy campaigns. The former would most likely not have any momentum into the sandbox, and the later would most likely clash with the GM's planed adventure.
I don’t see how a sandbox can’t work with the idea of developing a backstory over time. Typically in a sandbox there are a variety of options the group can explore and they will slowly learn about themselves and reveal bits of their backstories as they explore the various interesting things in the region.

I’m not discounting the importance of slowly weaving character specific hooks into the game at some point but that doesn’t need to happen right away when they are deciding between checking out the haunted mine, riding shotgun to protect the stage from bandits, going to the Army fort a week away to get the smallpox vaccine or hiring on as goons with the local boss.
 
One obvious thing that jumped out at me in my unscientific survey was that very few games actually have completely random char-gen--I think maybe 2 or 3 of the 61 I looked at even had that as an option. Normally if a game includes random elements, like generation of basic statistics, it has other things you choose about the starting character, like class/profession/archetype etc., skills, gear, and so on. And very frequently games that have random generation of basic attributes allow you to generate the whole set of numbers first and put each one where you like--which deals with the problem Z ZDL mentioned of lacking the correct value for a prime requisite stat. Or they allow you to adjust rolled stats, raising one by lowering another. So even for figuring basic stats these systems are often a mix of randomness and choice.

Still, I think thebigh thebigh is right that random systems tend to produce starting characters of different power or ability levels, though how much so of course depends on how important the randomly-generated elements actually are for play. As Bunch Bunch pointed out, in OD&D stats meant very little--except for initial hit points. Given the balance issue, and that some people care a good deal about it, it's a little surprising that games with random char-gen procedures don't include more balancing mechanisms--for instance, giving characters with low attributes more points to spend on skills, or an extra 'hero point' or something similar. I'd guess this is rare because (a) many games with mandatory random elements are deliberately 'old-school,' (b) figuring out the conversion factor could be a difficult part of design, and (c) it's easier just to include a non-random alternative rather than a balancing mechanism.

As to the idea that random elements can keep players from creating the characters they want to play, I guess that's true--if the part of character generation that is randomized is really that important to the character concept. But from my perspective, that's a very minor part of the constraints that limit what characters are possible. Much more important are the basic tenets of the particular game and campaign. Even if the universal design system would allow you to create it, you can't play a resurrected Alexander the Great zombie with a magic guitar in the gritty spy game set in Central Europe just before WW2. You can't play a magic-wielding character in systems or settings where there is no magic. And so on.
 
One obvious thing that jumped out at me in my unscientific survey was that very few games actually have completely random char-gen--I think maybe 2 or 3 of the 61 I looked at even had that as an option. Normally if a game includes random elements, like generation of basic statistics, it has other things you choose about the starting character, like class/profession/archetype etc., skills, gear, and so on. And very frequently games that have random generation of basic attributes allow you to generate the whole set of numbers first and put each one where you like--which deals with the problem Z ZDL mentioned of lacking the correct value for a prime requisite stat. Or they allow you to adjust rolled stats, raising one by lowering another. So even for figuring basic stats these systems are often a mix of randomness and choice.

Still, I think thebigh thebigh is right that random systems tend to produce starting characters of different power or ability levels, though how much so of course depends on how important the randomly-generated elements actually are for play. As Bunch Bunch pointed out, in OD&D stats meant very little--except for initial hit points. Given the balance issue, and that some people care a good deal about it, it's a little surprising that games with random char-gen procedures don't include more balancing mechanisms--for instance, giving characters with low attributes more points to spend on skills, or an extra 'hero point' or something similar. I'd guess this is rare because (a) many games with mandatory random elements are deliberately 'old-school,' (b) figuring out the conversion factor could be a difficult part of design, and (c) it's easier just to include a non-random alternative rather than a balancing mechanism.

As to the idea that random elements can keep players from creating the characters they want to play, I guess that's true--if the part of character generation that is randomized is really that important to the character concept. But from my perspective, that's a very minor part of the constraints that limit what characters are possible. Much more important are the basic tenets of the particular game and campaign. Even if the universal design system would allow you to create it, you can't play a resurrected Alexander the Great zombie with a magic guitar in the gritty spy game set in Central Europe just before WW2. You can't play a magic-wielding character in systems or settings where there is no magic. And so on.
It’s interesting you mention the idea of a balancing mechanism for low stats in random generation, it was something that occurred at least with TSR back in the beginning. Early on Gygax wrote about how demi humans like dwarves were meant as consolation prizes and options for people that rolled poorly and 1975’s Boot Hill indeed gave extra points to your stats if you rolled poorly.
 
As an almost-forever GM, if I get a chance to play and the system uses random generation, I love the surprise and the weight it takes off coming up with a concept. I also enjoy life path games like Traveller and WFRP.
I like the system in WFRP 4e where you can have a fully designed character but you can get exp bonuses by randomly rolling it as well. (I have heard a handful of complaints from people who don't want random generation but resent those who like it getting any mechanical advantage for using it, but I don't have much time for those. Being willing to roll with the punches is a solid reason to give someone a few bennies in my view and it's not an overpowering advantage).
 
And in early old school I think the advantage to high stats was not significant long term. I'm thinking OD&D/Basic. The bonuses helped but money was a bigger decider of effectiveness. Better armor, weapons, magic items. Stats had a smaller impact so wider margins weren't as bad.

That was a significant difference between OD&D and almost anything that came later; as I noted, prior to Greyhawk stats were almost irrelevant in OD&D and even after you had to get particularly good or bad to be all that noticable. This could be in particular contrast to, say, RuneQuest or even Traveller.
 
I don’t see how a sandbox can’t work with the idea of developing a backstory over time. Typically in a sandbox there are a variety of options the group can explore and they will slowly learn about themselves and reveal bits of their backstories as they explore the various interesting things in the region.

I’m not discounting the importance of slowly weaving character specific hooks into the game at some point but that doesn’t need to happen right away when they are deciding between checking out the haunted mine, riding shotgun to protect the stage from bandits, going to the Army fort a week away to get the smallpox vaccine or hiring on as goons with the local boss.
That might be a typical sandbox in games or circles I'm not running games in. From my point of view, those are atypical. So, I'm guessing we are both looking at it from our self-selected biases.

Nothing wrong with gaming the way you described there, but players with that preference would be a bad fit with how I run games.
 
I think there are many decent character generation variants in both random and non-random flavors, and to declare one type as superior to the other either means you've only seen bad examples of one of them, or you are over-generalizing what players care about. I'll chat over a few games to further lay out my thoughts:

I've tried to go to pains to note that when I say I can't be bothered with random character gen any more, its not a figure of speech: the "I" in that sentence is there for a purpose.

That said, there's at least strong game-culture things that reinforce this; I don't think I've gamed with anyone for decades now who showed any preference for random, and a number of people who are either hostile to it, or actively won't have anything to do with it. And as noted in the prior thread, this is not a case of new players, I think the youngest person I game with these days is in their early 40's.

Anyways, yeah, I see that there are a variety of concerns people mention when it comes to chargen systems:
1.) I want to be able to play a character I have in my head! - Seems modeling is best for this, with some kind of point buy being next, random last. There are also people who enjoy "Give me a character, and watch what I do with it!"
2.) I want character balance so my character isn't worthless in play - I think different RPGs require "balance" in different degrees depending on how they are structured. RPGs that involve a lot of time spent fighting, so much that you might say the game is largely about that, should probably aim for some kind of balance in that aspect. Outside of that, what is meant by balance, and how much different players care about it all vary I think.

To be clear, the first is the primary reason I have, though I think people are too quick to write off the second; while you can end up with a mess of a character in a point build game, as long as people are willing to work with you to help, you can usually iterate toward something that serves your purposes. And its not always about combat; having the character who is avowedly an investigator but is simply worse at it in any practical way than someone who its a second-string trick for is no fun either.
 
I generally prefer random for a few reasons...
It's partially tied to the style of game I like, where combat is dangerous and death is easy to come by. That asks for a relatively quick means of character generation.

Non-random doesn't actually have to mean complicated. It often is, because people who prefer it often have very specific sorts of results they're trying for, but you can have template or array character build that's quick but non-random. Even among point build, things like The Fantasy Trip are simpler than some rolled systems.
 
How many of those who prefer character building to random generation also like extensive back story? I could see those going hand in hand while random generation goes hand in hand with playing to find out.

While I can roll either way, my default is not a tabula rasa.
 
Not a clue. I can't run a game for the "play to find out" crowd, so my pool of anecdotes is severely biased. Some randomness combined with choices and some control have worked for us, but full randomness would most likely fall very flat.

I also think "play to find out" probably work a lot better with GM's that have a planned adventure, as the finding out can be used to give motivation to work toward the goal of the adventure. While extensive backstory most likely work better with character driven sandboxy campaigns. The former would most likely not have any momentum into the sandbox, and the later would most likely clash with the GM's planed adventure.

This is actually the opposite of my experience; filling in the backstory as you go works better for settings that are general and where the character doesn't need to be fit into the campaign carefully.
 
One obvious thing that jumped out at me in my unscientific survey was that very few games actually have completely random char-gen--I think maybe 2 or 3 of the 61 I looked at even had that as an option. Normally if a game includes random elements, like generation of basic statistics, it has other things you choose about the starting character, like class/profession/archetype etc., skills, gear, and so on. And very frequently games that have random generation of basic attributes allow you to generate the whole set of numbers first and put each one where you like--which deals with the problem Z ZDL mentioned of lacking the correct value for a prime requisite stat. Or they allow you to adjust rolled stats, raising one by lowering another. So even for figuring basic stats these systems are often a mix of randomness and choice.

Technically even OD&D didn't have complete random character gen; you picked your class after all.

The big issue with others is that as your system gets more detailed, barring a very strong archetype or class system, trying to completely random generate skills and other ancillary abilities can produce results that are simply incoherent, both in function or even in trying to figure how it came about. That's one reason random gen superhero games are in the minority; yeah, you can get your Spider-Man or Superman types where the powers are just sort of hung on there, but when you look at the genre, they're very much the exception.

Still, I think thebigh thebigh is right that random systems tend to produce starting characters of different power or ability levels, though how much so of course depends on how important the randomly-generated elements actually are for play. As Bunch Bunch pointed out, in OD&D stats meant very little--except for initial hit points. Given the balance issue, and that some people care a good deal about it, it's a little surprising that games with random char-gen procedures don't include more balancing mechanisms--for instance, giving characters with low attributes more points to spend on skills, or an extra 'hero point' or something similar. I'd guess this is rare because (a) many games with mandatory random elements are deliberately 'old-school,' (b) figuring out the conversion factor could be a difficult part of design, and (c) it's easier just to include a non-random alternative rather than a balancing mechanism.

There have been some games that did that at least in the past--DragonQuest gave high attribute total characters less experience as I recall--but my suspicion is that ones of more modern vintage either think its irrelevant or that people should just suck it up.

As to the idea that random elements can keep players from creating the characters they want to play, I guess that's true--if the part of character generation that is randomized is really that important to the character concept. But from my perspective, that's a very minor part of the constraints that limit what characters are possible. Much more important are the basic tenets of the particular game and campaign. Even if the universal design system would allow you to create it, you can't play a resurrected Alexander the Great zombie with a magic guitar in the gritty spy game set in Central Europe just before WW2. You can't play a magic-wielding character in systems or settings where there is no magic. And so on.

I'd suggest that at least some control over output is necessary for anyone who doesn't come to the table with a complete tabula rasa; a rather lot of people will have a concept that goes with a character who's strong or weak, smart or dumb, quick or slow for example.
 
That might be a typical sandbox in games or circles I'm not running games in. From my point of view, those are atypical. So, I'm guessing we are both looking at it from our self-selected biases.

Nothing wrong with gaming the way you described there, but players with that preference would be a bad fit with how I run games.
I can appreciate that although I wonder if that would be the case in practice. Now campaign length could be a factor, if we are taking a limited game that will last eight or ten sessions I could see the need to expedite things but if it is a longer term game I would think there would be the opportunity to weave in character bits as you go.

Anyhow as long as you and those you play with are happy with the result you are doing it correctly :thumbsup:
 
One thing that trolling through various games' char-gen showed me is that I really don't have a strong preference for random vs. non-random as such. Particularly on the issue of basic stats or attributes, where a lot of the dislike for random methods seems to lie.

The thing that I don't like about many non-random approaches is the amount of knowledge of the system that they demand. This can include what I'd call 'system mastery'--that is, knowledge of hidden synergies or complications within the mechanics that can make one option better than another although they appear equivalent--but it doesn't have to mean that. It can simply mean that you have to read through a large number of options and pick from them as you generate the character.

For example, in All for One: Regime Diabolique, in addition to generating statistics and skills, you need to choose resources, talents, and flaws (you don't have to pick all 3, but you need at least 1 and I'd guess most people pick 2 or 3). Deciding on a resource isn't that bad: there are only 11 different ones, and they are fairly self-explanatory. But there are something like 89 talents to choose from, and 70 flaws. Some are straightforward, others not; some can only be taken at char-gen while others can be added later as your character gains experience. It's exception-based design with a vengeance. You might think 'well, doubtless many of these talents only apply to a particular type or class of character, so in fact there are not so many to pick from.' Alas, no--all characters in the game are musketeers in France during the 1620s.

Another example is Paleomythic, a game I've been reading through lately. It has a very simple system overall and char-gen is quite straightforward too. But as part of it you can select anywhere from 0 to 3 talents. These 'talents' are really more like archetypes, occupations, or classes--each one has its own capabilities, provides its own advantages, and even its own starting gear. Their description takes up 67 pages in the Kindle e-book of the rules. That's a lot to wade through before even beginning to play what is a pretty simple game. You can of course avoid this by taking no talents, but if so you are likely nerfing your character somewhat. Or you can roll randomly for talents, but that is more randomness than I want--it would be rather like rolling for character class.

IME, systems for picking skills don't suffer as much from this kind of information bloat, because the skills' meanings are usually fairly transparent and there is often a unified system for what each level of skill means in mechanical terms (if the game has skill levels).
 
Last edited:
That seems more a generic objection to feat/talent systems than anything to do with build systems, per se. After all, if there's a problem there, its not like rolling them would be better.
 
Non-random doesn't actually have to mean complicated.
That's why I wrote 'generally'... that word is there for a reason...
What Lofgeornost describes are the sort of 'build' games that really do stall my desire to learn/play them.
I'm fine if a random chargen game hands me a 'gimped' character, but going through a complex build system only to find that 'feat X' doesn't do what I thought it did because I also need 'feat Q' and 'skill 3'... that's annoying, and may not show itself till well into the campaign/adventure.
It's fair to say that I also, generally, prefer the sorts of players who like random vs. the nitpicky sort of building.

Still, it's not like I REFUSE to play character build games.
 
This is actually the opposite of my experience; filling in the backstory as you go works better for settings that are general and where the character doesn't need to be fit into the campaign carefully.
Interesting. Maybe I just prefer more fleshed out characters from the start, and found that I best fit that into the sandboxy type campaigns that I run, or there are other variables involved that would require a more in-depth discussion than what would be suitable for this thread.

I can appreciate that although I wonder if that would be the case in practice. Now campaign length could be a factor, if we are taking a limited game that will last eight or ten sessions I could see the need to expedite things but if it is a longer term game I would think there would be the opportunity to weave in character bits as you go.
Well, I generally don't run games that are limited to such a limited number of sessions. So that's not the case for me.

However, even if hashing out a background at char gen, there's still a lot being weaved in later as well. It's not that we have figured out everything from the characters say thirty previous years before the game starts.
Anyhow as long as you and those you play with are happy with the result you are doing it correctly :thumbsup:
Yup. :thumbsup: But making sure everyone's gaming style at a table is compatible can be tricky.
 
That's why I wrote 'generally'... that word is there for a reason...
What Lofgeornost describes are the sort of 'build' games that really do stall my desire to learn/play them.
I'm fine if a random chargen game hands me a 'gimped' character, but going through a complex build system only to find that 'feat X' doesn't do what I thought it did because I also need 'feat Q' and 'skill 3'... that's annoying, and may not show itself till well into the campaign/adventure.
It's fair to say that I also, generally, prefer the sorts of players who like random vs. the nitpicky sort of building.

Still, it's not like I REFUSE to play character build games.
I hate the "char op" mini-game that late 2nd ed/3rd ed D&D brought to the table. Second edition Mutants and Masterminds was kind of a sweet spot, but for third edition the loaf was sliced so finely i was getting out-manouvered by my players. That's why I like BW and SoM. With BW, it's easy to make a character, because all skills are useful, and any real character creation doesn't start until after the game starts. With SoM, it's hard to gimp a build because again, everything is useful. And the fact that the point-boy char-gen is hidden behind life-paths makes it harder to min-max out of the gate.
 
That seems more a generic objection to feat/talent systems than anything to do with build systems, per se. After all, if there's a problem there, its not like rolling them would be better.

Yes; as I said, I don't object to non-random char-gen on principle or anything and a lot of games I enjoy include a lot of non-random elements in their char-gen procedures. But complex or lengthy feat/talent systems are normally non-random (with some exceptions) and as you note don't work well with randomization. So it seems reasonable to me to say that one thing I don't like about non-random char-gen is its tendency to that kind of thing. It's not that non-random systems have to include bloated feat or talent lists, but they are much more likely to do so than more random ones.
 
Not a clue. I can't run a game for the "play to find out" crowd, so my pool of anecdotes is severely biased. Some randomness combined with choices and some control have worked for us, but full randomness would most likely fall very flat.

I also think "play to find out" probably work a lot better with GM's that have a planned adventure, as the finding out can be used to give motivation to work toward the goal of the adventure. While extensive backstory most likely work better with character driven sandboxy campaigns. The former would most likely not have any momentum into the sandbox, and the later would most likely clash with the GM's planed adventure.
The phrase "play to find out" comes from some games, Burning Wheel among them, that discourage deep backstory while also encouraging sandbox play. The play to find out is both about the character AND the campaign.

I use modules, but I try to be very open to how they play out, and I've got a lot more tolerance for players walking away from them (we bailed out on Street of Gems from The Companions because the players just weren't interested in the situation in the town. In my Glorantha campaign, the players decided to talk to and free the white wyrm of Lair of the White Wyrm which resulted in that early PC with the worst attributes being labeled Wyrm Friend. The module assumes the wyrm is just another monster to be killed for it's treasure (the PCs still got the treasure, the wyrm brought it to them after being released...). The wyrm freeing was accomplished by the dwarves they met in the dungeon (led by a former PC dwarf after that player rage quit the game).
 
Looking through various games' char-gen also strengthened my conclusion that, on the whole, it's easier to convert a system that has random elements in it to entirely non-random than vice-versa.

In the systems I looked at, the most common significant randomized element was rolling for basic statistics, attributes, what have you. It's not very hard to convert this to an approach where players simply have a set number of points to allot to statistics, with appropriate caps and minima. If the system has a lot of 'break points' built into it, this can lead to characters having similar stats, but I don't see this as a big problem for play. Starting cash is also often handled randomly, but that's trivial to turn into a set figure.

Lifepath elements that require rolling on tables may be a bigger problem; it depends if some outcomes on the table are clearly better than others. You can simply rule that players pick whichever option from the table they want--Artesia allows this, for example--but that's open to abuse. So conversion might require rejiggering things so that all options are equal, or requiring a player to take equal numbers of good and bad results of their choice.

On the other hand, non-random elements of char-gen can be hard to make random. Skills, talents, feats, etc. are difficult to select randomly, unless the players are willing to accept very bizarre characters. More narrative systems like Fate and Heroquest, where players create unique keywords, aspects, etc. for their characters as a key part of describing them mechanically are also really resistant to randomization.
 
It's not very hard to convert this to an approach where players simply have a set number of points to allot to statistics, with appropriate caps and minima.

Exactly. It can be as easy as instead of physically rolling the dice, just pretend you rolled this, this, and this.
 
That seems more a generic objection to feat/talent systems than anything to do with build systems, per se. After all, if there's a problem there, its not like rolling them would be better.
Except it kind of is in a random chargen. So both V&V and Golden Hetoes are supers RPGs where the random portion of chargen reduces the need for any system mastery. You get what you get and then try to make it work.
I love Champions but a lot of that had to do with getting in on it right as 2nd edition came out. The number of powers, effect, limitations and disadvantages we're smaller then they ultimately would become so it was easier and less intimidating to get started. I took one look at many peoples beloved 4th edition Hero Games Champions and said "Oh hell no!" I didn't even want to look through that much just to figure out what had or had not changed. Some of it is just psychological. Big books of lists are intimidating to try to master. The bigger the worse.
I mean in my mind I look at a lot of the new stuff coming out and think man that's too much just to get into a game.
 
I prefer to meet my character during character generation, not assemble him on a table like Dr. Frankenstein. So I like random stats, rather than point buy. Obviously it's just a matter of taste. Another thing I like about games with random char gen is that it is generally easy to override it with house rules, and most games explicity suggest methods in a side-bar. I have not experienced the converse in games with point buy systems. Flexibility is always good.
 
Another thing I like about games with random char gen is that it is generally easy to override it with house rules, and most games explicity suggest methods in a side-bar.
Or later on in-game... ways to change your stats through training, ways to learn/train new skills. DCC has the whole 'quest for it' mantra which suggests some flaccid man-rat character might find his way to being Prince Charming if he really strove to do so... "I heard there's a guy in Riften who can change your face..."
 
I also think "play to find out" probably work a lot better with GM's that have a planned adventure, as the finding out can be used to give motivation to work toward the goal of the adventure. While extensive backstory most likely work better with character driven sandboxy campaigns. The former would most likely not have any momentum into the sandbox, and the later would most likely clash with the GM's planed adventure.
"Play to find out," isn't just for players though. As a GM, I want that too, which means a sandbox is generally going to work better than pre-planned plots.

A PC doesn't need a backstory in order to go adventuring, they just need to want to go adventuring. Why? We can play to find out.
 
Last edited:
That's why I wrote 'generally'... that word is there for a reason...

I think even as a generalization its a problem; I think it conflates non-random with point build, and there are a lot of non-point-build non-random character gen systems. There may be a limited number of simple point-build games (like I said, the people attracted to those in the first place often want more nuance) but archetype systems are hardly time consuming, and they aren't thin on the ground.
 
Looking through various games' char-gen also strengthened my conclusion that, on the whole, it's easier to convert a system that has random elements in it to entirely non-random than vice-versa.

I dunno, man; it still seems like random-gen systems are much more prone (at least these days, I think someone brought up Hero which used to be a strong counter-example and is still a moderate one) to having strong breakpointing that will create problems when made non-random. At the very least you're going to want to use arrays or some such.

In the systems I looked at, the most common significant randomized element was rolling for basic statistics, attributes, what have you. It's not very hard to convert this to an approach where players simply have a set number of points to allot to statistics, with appropriate caps and minima. If the system has a lot of 'break points' built into it, this can lead to characters having similar stats, but I don't see this as a big problem for play. Starting cash is also often handled randomly, but that's trivial to turn into a set figure.

See, I kind of see overly strong hammering of breakpoints with the accompanying similarity of stats to be a bit of a problem. If most of a stat range isn't going to be used, it might as well not be there (its the biggest single complaint I used to have with Hero).

Lifepath elements that require rolling on tables may be a bigger problem; it depends if some outcomes on the table are clearly better than others. You can simply rule that players pick whichever option from the table they want--Artesia allows this, for example--but that's open to abuse. So conversion might require rejiggering things so that all options are equal, or requiring a player to take equal numbers of good and bad results of their choice.

On the other hand, non-random elements of char-gen can be hard to make random. Skills, talents, feats, etc. are difficult to select randomly, unless the players are willing to accept very bizarre characters. More narrative systems like Fate and Heroquest, where players create unique keywords, aspects, etc. for their characters as a key part of describing them mechanically are also really resistant to randomization.

Well, that's more of an issue with relatively few systems that use powers, talents and the like being random in the first place. If you look at the simpler ones, they can be made random relatively easily (I know there's a random generator for Fantasy Trip characters for example; it just uses a table so the skill/talents cluster up properly).

Now if you want to argue its hard to make systems with a lot of moving parts purely random in the first place, I wouldn't argue with you.
 
Except it kind of is in a random chargen. So both V&V and Golden Hetoes are supers RPGs where the random portion of chargen reduces the need for any system mastery. You get what you get and then try to make it work.

I'm extremely familiar with V&V, and I think I'd qualify that as "try to make it work." Superhero games are one of the places random gen can most easy fail, just because they can all too easily produce a character who's neither coherent or even, in any meaningful way, functional. And that's not even getting into the power swings.

That said, there are a lot simpler versions of superhero build systems than Champions. Supers! RED comes to mind. Comparing every build system to Champions is like comparing every random generation system to Powers and Perils (except P&P doesn't even have the strengths people place on random systems in the first place, but it has some of the same "You have no idea how the character will come out because there's so much swing in some places").

Like I've said, if we're going to use some of the simplest roll systems as models then we ought to be comparing them to some of the simplest build systems, not the most complex.
 
I'm extremely familiar with V&V, and I think I'd qualify that as "try to make it work." Superhero games are one of the places random gen can most easy fail, just because they can all too easily produce a character who's neither coherent or even, in any meaningful way, functional. And that's not even getting into the power swings.

That said, there are a lot simpler versions of superhero build systems than Champions. Supers! RED comes to mind. Comparing every build system to Champions is like comparing every random generation system to Powers and Perils (except P&P doesn't even have the strengths people place on random systems in the first place, but it has some of the same "You have no idea how the character will come out because there's so much swing in some places").

Like I've said, if we're going to use some of the simplest roll systems as models then we ought to be comparing them to some of the simplest build systems, not the most complex.
Hmm I think comparing V&V to Champions is a reasonable comparison as far as complexity. Golden Heroes maybe not as much. But that wasn't what I was trying to compare. My goal was to show that games that have large amount of selections can work with random chargen which was what I replaying to. Both GH and V&V do that. Whether the games are good games it does pull a lot of the system mastery question out of the equation.
 
I think even as a generalization its a problem; I think it conflates non-random with point build, and there are a lot of non-point-build non-random character gen systems.
Sorry, I'm not fluent in all the RPGs ever made... so my generalizations can only speak to the ones I'm aware of. Even when it's not 'points' specifically, there's often a menu of choices... pick 2 from column A, 3 from column B... etc. Sometimes the choices are minimal and clear from their names... 'eggs and ham'... other times there are lots of them, and they have names that like 'moon or my hammy' which obfuscates stuff.
A lot of times I just don't care enough to try to figure it out... so give me a random roll.
(I feel like someone might any moment accuse me of being a 'filthy casual' because I don't enjoy the char-op/deck-building stuff)
 
Last edited:
Random mechanical elements are fine where they don't really matter or characters are basically disposable, but the more important they are to play, then the more important it is that they get set "fairly" so that everyone's character is of roughly equal potential; and while some players like to play PC's who are weaker than the rest of a party, that's a choice that they should get to make for themselves, rather than one that the whims of fate or the GM should be making for them ("I want to play a beggar in a party of nobles" is not the same as "you have to play a beggar in a party of nobles").

That said, I don't like "compensation for bad rolls" as a concept. If you're needing to apply fixes like that to your random character creation system, then you are acknowledging that your system is inherently flawed and not providing the sorts of results you actually want; may as well fix your actual problem by changing or removing the randomness.

(The one exception is WFRP, where doing anything but playing with the whims of fate feels deeply wrong.)
 
I'll repeat what I said in the other thread. Point-buy IMO is superior to random.

With purely rolled characters you always get a big difference in ability between the people who rolled well and the people who rolled badly. For the unlucky players, it is not much fun being Blop the Potato Peasant among a party of capable adventurers. Neither is it necessarily fun, if you rolled well, to be a shining demigod in a party of average joes- unless you have a serious case of main character syndrome. And the GM needs to do extra work to balance encounters for a party with a wide range of abilities. It's just suckitude all round.

I suppose more old-school games where life is cheap, scenarios brutal, and narrative comparatively less important, rolling for stats makes more sense.
Or games where what you roll doesn't matter much mechanically (like OD&D).
 
Or games where what you roll doesn't matter much mechanically (like OD&D).
I guess I just don’t understand the need for all the PCs to be balanced or why someone can’t have fun with a mechanically inferior character. Have you guys never had a mid to high level party where someone died or a new player joined resulting in a first level character joining the party?

Games like AD&D 1E and B/X are designed so those characters will quickly catch up to the party in level, at least quickly advance to one level below everyone else, so it shouldn’t be a big deal but I get the idea it might be with some folk.
 
I guess I just don’t understand the need for all the PCs to be balanced or why someone can’t have fun with a mechanically inferior character. Have you guys never had a mid to high level party where someone died or a new player joined resulting in a first level character joining the party?
Yes. I'm also skeptical of any discussion of balance that doesn't come with some very clear ideas about the specific game and playstyle being balanced. To start with, the term balance these days tends to be used to mean, "equally useful in combat". But what if the game isn't combat focused? How do you "balance" a hacker, a cop and a nurse? So much depends on the nature of an individual table or even session.
 
I guess I just don’t understand the need for all the PCs to be balanced or why someone can’t have fun with a mechanically inferior character. Have you guys never had a mid to high level party where someone died or a new player joined resulting in a first level character joining the party?

Games like AD&D 1E and B/X are designed so those characters will quickly catch up to the party in level, at least quickly advance to one level below everyone else, so it shouldn’t be a big deal but I get the idea it might be with some folk.
Sometimes it matters, sometimes it doesn't. There are point-buy systems out there that start you off with a random amount of points.

But a game that'll work when played by my group of old, long-time friends, is not the same as one for my current weekly game group, where there are players who are consistently lucky in chargen and who will use extra character power if they have it to mess things up.
 
Random mechanical elements are fine where they don't really matter or characters are basically disposable, but the more important they are to play, then the more important it is that they get set "fairly" so that everyone's character is of roughly equal potential; and while some players like to play PC's who are weaker than the rest of a party, that's a choice that they should get to make for themselves, rather than one that the whims of fate or the GM should be making for them ("I want to play a beggar in a party of nobles" is not the same as "you have to play a beggar in a party of nobles").
I'd phrase the question somewhat differently.

"Are you willing to risk playing a beggar in exchange for the possibility of playing a noble?"

This is where the game part becomes important for me; I don't necessarily think there's anything wrong with the idea that random generation should provide a high risk/reward ratio. I'm fine with both random and point buy being an option, but if a player chooses random I expect them to roll with the punches and if they choose point buy I expect them to be ok with the fact their character will be average, not exceptional (whatever that means in the confines of the system).
 
Sorry, I'm not fluent in all the RPGs ever made... so my generalizations can only speak to the ones I'm aware of.

Which is fair, but it still made the statement overly broad.

Even when it's not 'points' specifically, there's often a menu of choices... pick 2 from column A, 3 from column B... etc. Sometimes the choices are minimal and clear from their names... 'eggs and ham'... other times there are lots of them, and they have names that like 'moon or my hammy' which obfuscates stuff.

At which point in games of the same complexity you'll likely end up with something just as incoherent/unuseful.
 
Hmm I think comparing V&V to Champions is a reasonable comparison as far as complexity. Golden Heroes maybe not as much. But that wasn't what I was trying to compare. My goal was to show that games that have large amount of selections can work with random chargen which was what I replaying to. Both GH and V&V do that. Whether the games are good games it does pull a lot of the system mastery question out of the equation.

That's fair. I was just noting that, unlike the usual consequences with OD&D, it was actually pretty easy for V&V to throw you a character who was, while probably technically playable by some standards, not going to really work in the context of the other characters and opponents in a campaign. I know quite well because I had one I tried with at one time, and lets just say that by themselves a super sense and heightened attack were not going to cut it.

Though whether V&V (I'm guessing 2e) and Champs had the same complexity is, I suspect, one of those things people would argue about. Certainly the former packaged up a lot of things in bigger chunks, and had less meaningful choices in play.
 
I guess I just don’t understand the need for all the PCs to be balanced or why someone can’t have fun with a mechanically inferior character. Have you guys never had a mid to high level party where someone died or a new player joined resulting in a first level character joining the party?

I have. I also watched them mostly hide while they waited for the experience system to catch them up enough to where they felt like they could contribute without just dying again.
 
Banner: The best cosmic horror & Cthulhu Mythos @ DriveThruRPG.com
Back
Top