Chess is not an RPG: The Illusion of Game Balance

[Bedrockgames]

Umbran nailed it.

We don't want suck characters with all stats less than 7. We don't want to be forced to play a fighter because we didn't roll even a 10 for INT.

But we also don't want cookie-cutter PCs that were just using the exact same point buy as the last PC.

Celebrim's misinterpretation was that I wanted wildly random results with low numbers. I want random results in a certain desirable range.

And I think these preferences you express are largely the norm in D&D these days. I'm the one arguing for random results, but I acknowledge that isn't the norm.Most D&D players I have met share your sense of wanting some amount of randomness but also want things to fall inside a range they consider reasonable and workable. This is why many people who use 4d6 drop the lowest, do two sets. They want characters that have a certain level of competence but they do not desire that character creation be fully predictable or controllable. Basically they want some surprises but would nothing too game breaking. That is pretty reasonable in my mind.

 

[Celebrim]

We don't want suck characters with all stats less than 7. We don't want to be forced to play a fighter because we didn't roll even a 10 for INT.

Keep that in mind. It's worth coming back to.

But we also don't want cookie-cutter PCs that were just using the exact same point buy as the last PC.

Why? Why in the world would this be an outcome? What's forcing you to arrange your points in the same manner each time? Of the 15 different PC's the players have created so far in this latest campaign, I doubt any two shared the same stats. I'm fairly sure none have shared the same class or combination of classes. If you don't want cookie cutter builds, why make them?

Celebrim's misinterpretation was that I wanted wildly random results with low numbers. I want random results in a certain desirable range.

Except 4d6 doesn't give you that. It gives you wildly random results. You can still get 3, 4, 5, 6, or 7. It doesn't guarantee results fall in a certain range. It makes sucking less likely sure, but it doesn't prevent it. It just makes you think, "Well, it won't really happen to me." The problem with 4d6 take best three is sooner or later you get it producing results that look like 3d6 and then your desired results fail. I've seen it happen. I've DM'd enough players to know that look on their face when they roll up the character and they got the equivalent of 18 point buy (or worse) - which is going to happen if you've got 8 players at the table and you make them all stick to their rolls. Sure, I've played with 7's. I played Ogden Mudstump, the dwarf thief with 5 charisma.

I just rolled up 4d6 take 3 in order, it was: 10, 5, 10, 11, 6, 8. Second was 15, 12, 17, 6, 10, 7. Third was 7, 8, 13, 5, 15, 11. Fifth was 12, 13, 15, 4, 13, 13. Sixth was 15, 9, 12, 10, 13, 13. That's what 4d6 really looks like. Random. Actually freaking random. You take some random samples, and you'll get anything from like 9 to 40 point buy.

What if I told you that we were going to use point buy, but we were going to use 5d10 to determine how many points you got to spend?

Tthat guy with the 10, 5, 10, 11, 6, 8 is not going to play it. The dice didn't 'do their job' - though in 30 or 40 rolls, probably 4 or 5 will be that bad. Sooner or latter they are going to end up playing the seventh set of dice I rolled while writing this: 15, 15, 17, 13, 15, 13. That's the thrilling sort of results they were gambling on all along. If you have them roll away from the DM, you can bet on them forgetting those first few sets. If they roll in front of the DM, you can bet they'll beg or kill the character off. Random means: "Either I win the jackpot, or I'm going to enter into a complex negotiation that makes a mockery of the so called rules of character creation."

If what you really want was no numbers less than 7, why don't you do 2d6+6 straight up in order? If what you really wanted was results in a certain desirable range, why don't you use point by with 22+2d4 points?

I would argue that the things you prioritize aren't in the these alternatives, and they aren't 'don't get low numbers' or 'don't have no stat higher than a 10'. I played that way about 15 years, both as a DM and a PC. I know some of the attractions. Rolling your stats seems more real, more hard core, especially if you are used to rolling your stats. Point buy brings that gambler's thrill - maybe this time I'll get lucky. Point buy brings truly odd results - two 17's and 3 9's, an 18 and 3 8's, 2 18's and a 4, etc. - that you might never otherwise consider playing. And then there is always the interest of having a pretty decent roll in a stat you'd never otherwise place it in, which arguably is my favorite part. That randomness has some benefits, but... it just isn't worth it not the least of which is because in practice its more like character generation in Baldur's Gate (the video game). People roll until they have the numbers that they can live with.

If what you really want is inspiration to play something you wouldn't consider otherwise, go ahead and roll that 4d6 keep 3 six times - then use point buy to match the result as closely as possible. I'm done with using fortune rolls to determine the outcome of character generation. Save the fortune rolls for the game.

 

[Bedrockgames]

Keep that in mind. It's worth coming back to.



Why? Why in the world would this be an outcome? What's forcing you to arrange your points in the same manner each time? Of the 15 different PC's the players have created so far in this latest campaign, I doubt any two shared the same stats. I'm fairly sure none have shared the same class or combination of classes. If you don't want cookie cutter builds, why make them?



Except 4d6 doesn't give you that. It gives you wildly random results. You can still get 3, 4, 5, 6, or 7. It doesn't guarantee results fall in a certain range. It makes sucking less likely sure, but it doesn't prevent it. It just makes you think, "Well, it won't really happen to me." The problem with 4d6 take best three is sooner or later you get it producing results that look like 3d6 and then your desired results fail. I've seen it happen. I've DM'd enough players to know that look on their face when they roll up the character and they got the equivalent of 18 point buy (or worse) - which is going to happen if you've got 8 players at the table and you make them all stick to their rolls. Sure, I've played with 7's. I played Ogden Mudstump, the dwarf thief with 5 charisma.

I just rolled up 4d6 take 3 in order, it was: 10, 5, 10, 11, 6, 8. Second was 15, 12, 17, 6, 10, 7. Third was 7, 8, 13, 5, 15, 11. Fifth was 12, 13, 15, 4, 13, 13. Sixth was 15, 9, 12, 10, 13, 13. That's what 4d6 really looks like. Random. Actually freaking random. You take some random samples, and you'll get anything from like 9 to 40 point buy.

What if I told you that we were going to use point buy, but we were going to use 5d10 to determine how many points you got to spend?

Tthat guy with the 10, 5, 10, 11, 6, 8 is not going to play it. The dice didn't 'do their job' - though in 30 or 40 rolls, probably 4 or 5 will be that bad. Sooner or latter they are going to end up playing the seventh set of dice I rolled while writing this: 15, 15, 17, 13, 15, 13. That's the thrilling sort of results they were gambling on all along. If you have them roll away from the DM, you can bet on them forgetting those first few sets. If they roll in front of the DM, you can bet they'll beg or kill the character off. Random means: "Either I win the jackpot, or I'm going to enter into a complex negotiation that makes a mockery of the so called rules of character creation."

If what you really want was no numbers less than 7, why don't you do 2d6+6 straight up in order? If what you really wanted was results in a certain desirable range, why don't you use point by with 22+2d4 points?

I would argue that the things you prioritize aren't in the these alternatives, and they aren't 'don't get low numbers' or 'don't have no stat higher than a 10'. I played that way about 15 years, both as a DM and a PC. I know some of the attractions. Rolling your stats seems more real, more hard core, especially if you are used to rolling your stats. Point buy brings that gambler's thrill - maybe this time I'll get lucky. Point buy brings truly odd results - two 17's and 3 9's, an 18 and 3 8's, 2 18's and a 4, etc. - that you might never otherwise consider playing. And then there is always the interest of having a pretty decent roll in a stat you'd never otherwise place it in, which arguably is my favorite part. That randomness has some benefits, but... it just isn't worth it not the least of which is because in practice its more like character generation in Baldur's Gate (the video game). People roll until they have the numbers that they can live with.

If what you really want is inspiration to play something you wouldn't consider otherwise, go ahead and roll that 4d6 keep 3 six times - then use point buy to match the result as closely as possible. I'm done with using fortune rolls to determine the outcome of character generation. Save the fortune rolls for the game.

Janx said he didn't want all stats less than 7, he never objected to one or two stats in that range. You show us your individual rolls all you like but there is a difference between rolling 3d6 and rolling 4d6 drop the lowest. Obviously, you will still have the possibility of a terrible roll. It is still random. That is why many folks who play this way have a "hopeless" character rule where if the character is absolutely hopeless (and definitions on that can vary) you re-roll. But I don't get why it is a problem to understand why someone might want that safety valve but still dislike doing a point buy. With a point buy you get what numbers you wish. You have more control over individual stats. Some people like that. But lots of people don't. They want the random element. There is always a matter of degree. Janx wants less extremes than I do in my random rolls, but clearly still wants some amount of that kind of variation and unpredictability.

 

[Aura]

But too often I see folks, myself certainly included, trying to position their set of preferences as the most ideal way to approach RPGS, and they use all manner of logic and evidence to prove that point. I find it odd that "logic" and "reason" so often lead people back to their own set of preferences (just like the Wick article, he makes a good argument for his position, but you can tell he started with his conclusion and worked his way toward it, rather than the other way around).

I agree on the issue of people often starting with the conclusion in their argument. For example, you note Wick made a good argument for his position, but you could tell he started with his conclusion. This left me wondering why you said that.

Reflecting on the article, I felt he was working from the conclusion is because his arguments didn't support the conclusions. For example, he concludes balance does not matter, but seems to argue, in the context of the RPG as he sees it, that balance does not exist. Similarly, he defines a RPG as a game that rewards players for acting in character, but argues on the subject of not penalizing said activity.

In both cases, he didn't really seem to get to his conclusion, which led me to suspect he was working from the conclusion. If I may ask, what led you to also suspect as I did?

 

[Bedrockgames]

I agree on the issue of people often starting with the conclusion in their argument. For example, you note Wick made a good argument for his position, but you could tell he started with his conclusion. This left me wondering why you said that.

Reflecting on the article, I felt he was working from the conclusion is because his arguments didn't support the conclusions. For example, he concludes balance does not matter, but seems to argue, in the context of the RPG as he sees it, that balance does not exist. Similarly, he defines a RPG as a game that rewards players for acting in character, but argues on the subject of not penalizing said activity.

In both cases, he didn't really seem to get to his conclusion, which led me to suspect he was working from the conclusion. If I may ask, what led you to also suspect as I did?

I think said the argument was good because I have seen people meet with a lot of difficulty refuting his points. That doesn't mean his conclusion is supported by his argument. I definitely think there are serious issues with his conclusion and I feel like there is a bit of sophistry at work in how he gets there. I suspect if we drill down we will find serious flaws in his reasoning (for example much of it rests on his definition of RPG and I don't think that definition is an accurate one that reflects how people in the hobby use the word). I also think he starts with something everyone just kind of intuitively recognizes: chess isn't an RPG. His argument seems to be, anything you add to chess doesn't suddenly make it an RPG. He then draws a comparison saying D&D actually meets the requirements of a board game just like chess and anything we add to it, doesn't make it an RPG. I think his argument fails to acknowledge from the very beginning that D&D has a much different set of assumptions than chess. Importantly, in D&D you are given permission to try anything you want and the GM is supposed to rule on whether it works or not. Plus you have all kinds of mechanics for social things (even in the white box with attributes like Wisdom and Charisma and talks about the parameters of play...which chess doesn't do). The history isn't even really important here though, what matters is no matter how far back one wants to go and peal away at D&D to 'prove' it isn't an RPG, clearly it is one now. That is how it is used and that is what it is designed to do. Either way though, I think everyone intuitively understands D&D is an RPG in the same way they understand Chess is not. But in my mind saying D&D isn't a roleplaying game is like saying the Model T isn't a car. I am sure you can build an argument for that case, but I don't think many people would take it seriously.

 

[Celebrim]

That is why many folks who play this way have a "hopeless" character rule where if the character is absolutely hopeless (and definitions on that can vary) you re-roll.

I think that my posts have indicated lots of experience with people declaring that their character is hopeless and can they please have a do over.

But I don't get why it is a problem to understand why someone might want that safety valve but still dislike doing a point buy.

Because, once again, what that safety valve indicates is that what they really want is not random, but (at least the chance) to be above average. No one ever suggests they need do overs when they randomly generate 4 16's or higher. They are like gamblers who get do overs whenever they lose. They get the thrill of thinking that they legitimately won the lottery, but none of the downside of losing it. It ends up being a form of self-delusion.

But lots of people don't. They want the random element. There is always a matter of degree. Janx wants less extremes than I do in my random rolls, but clearly still wants some amount of that kind of variation and unpredictability.

I think that Gygax's discussion of this in the 1e DMG is spot on and indicates he understood what people really wanted: "White it is possible to generate some fairly playable characters by rolling 3d6, there is often an extend period of attempts at finding one due to quirks of the dice. Furthermore, these rather marginal characters tend to have short life expectancy - which tends to discourage new players, as does having to make do with some character of a race or class which he or she really can't or won't identify with."

So Gygax's experience isn't that players are hard core and roll 3d6 straight up and play such characters successfully - as I've often heard bragged on the boards as being 'old school'. His experience is fairly similar to mine. Players find a way to keep rolling dice until they get what they want. Moreover, Gygax doesn't assert that it is true that stats don't matter and anything can be played successfully. He asserts that players probably neither want to nor should be playing marginal characters. So he provides not one but 4 methods to players which mitigate randomness in various fashions, and contrary to some assertions he doesn't set out one as default or even that only one method should be used. In fact, the clearest reading of the sentence, "Four alternatives are offered for player characters.", is go ahead and allow the player to choose what sort of randomness mitigation he's most comfortable with.

But what isn't said about those methods that perhaps should have been said, is that it tends to have the very same problem as 3d6 straight up. The first time through a method still might generate a "hopeless" character, or a character that the player can't or won't relate to, and then you'll be right back to an "extended period of attempts at finding one due to quirks of the dice". Eventually I learned that players will just keep rolling till they get what they want, promptly forget that whole extended period, and then declare how much they like randomness with a perfectly straight face.

And because what players really want that declare they like is randomness is to be above average, in the long run 'random' character generation tended to closely match starting out with one point buy and steadily increasing it. Often in my experience this was accompanied by all the sorts of drama I indicated earlier. I even played with one group where this had been taken to its logical conclusion. The DM had simply allowed everyone to basically have 17's and 18's in practically all stats, and then responded to this by having all NPCs have 18's in all stats. For that DM, that had ended the drama and the hypocrisy - the players had finally gotten what they emotionally if not logically had wanted. They could feel good about their above average characters, even though had they thought about it logically rather than emotionally, there never was a world where the PCs stats were so average.

You basically admit it yourself. You say you want random, but you want no possibility of a terrible result. If you really wanted that, you should have used method 3, rather than method 1. Method 3 gives very good odds that your scores will be at least average. However, what it doesn't do that pretending to like 4d6 take 3 does, is make it easy to perform and justify getting a do over, nor does it give you what you want instead of just something you could play. Again, if your goals were what you say they are, and if you'd reflected on it at all, you'd do something different than what you do.

 

[Bedrockgames]

You basically admit it yourself. You say you want random, but you want no possibility of a terrible result. If you really wanted that, you should have used method 3, rather than method 1. Method 3 gives very good odds that your scores will be at least average. However, what it doesn't do that pretending to like 4d6 take 3 does, is make it easy to perform and justify getting a do over, nor does it give you what you want instead of just something you could play. Again, if your goals were what you say they are, and if you'd reflected on it at all, you'd do something different than what you do.

I never said I want no possibility of terrible result. I said I do. I like 3d6 down the line. I just recognize most people don't want that and said 4d6 drop the lowest is the norm for that reason. But I'm absolutely fine with terrible results or I wouldn't enjoy rolling 3d6 or having things like random powers and flaws in character creation.

You make a very long post and I really don't have time or interest in addressing it at length. All I can say is my interests are different than your interests. I like random in the game and during character creation. I am fine with a bad result. That isn't a brag, I don't think it makes me better than anyone else, I just think it makes the game fun for me. You can either believe what I am telling you or not. That is up to you and it isn't going to change how I look at games or how I play them. You obviously prefer something more like a point buy and that clearly gives you what you want. That is great. Keep doing it. I am glad you and your group have a method that works for. Don't understand why you have trouble believing that the method I like works for me (especially when I fully acknowledge most people don't like it).

Also I think there is something of a false choice at work in your reasoning. You keep trying to make it about choosing between all stats in an average range or the possibility of all stats being terrible. I think guys like Janx are totally fine with one or two really bad scores on occasion, they just don't want the hopeless character (one that isn't good at anything). I can't speak for him of course, but that is my experience with may folks who like the 4d6 method. Their okay having 6s and 7s once in a while, they just don't want a character they can't work with.

 

[Celebrim]

Also I think there is something of a false choice at work in your reasoning. You keep trying to make it about choosing between all stats in an average range or the possibility of all stats being terrible. I think guys like Janx are totally fine with one or two really bad scores on occasion, they just don't want the hopeless character (one that isn't good at anything). I can't speak for him of course, but that is my experience with may folks who like the 4d6 method. Their okay having 6s and 7s once in a while, they just don't want a character they can't work with.

I quite understand that. I also am fine with having 6's and 7's on occasion. I'd even say its one of the things I tend to miss about random generation is you tended to see and play more weird characters. But again, if you aren't fine with a hopeless character, if you think that there are characters you can't work with then you need to admit that you don't like randomness. What you really want is a character generation system that delivers to you a character you can work with, something random generation is never guaranteed to do. Even if you do 9d6 take best 3, there is a good chance that the score will be below 16 - which defeats the whole intention of the method which is to ensure that you get a 16+ in your prime requisite. If that is what the real intention is, stop pretending to like random.

I think you actually understand how central all that is to my argument, because you do actually say what you must say if you honestly like random:

I never said I want no possibility of terrible result. I said I do. I like 3d6 down the line...I like random in the game and during character creation. I am fine with a bad result.

Ok. If you like bad results, then I suggest that you do 3d4 straight up down the line, while the rest of the table do method III (36 3d6, in groups of 6, take best result from each group straight up). And you can use your 3d4 straight up for every new character as well.

 

[Bedrockgames]

Ok. If you like bad results, then I suggest that you do 3d4 straight up down the line, while the rest of the table do method III (36 3d6, in groups of 6, take best result from each group straight up). And you can use your 3d4 straight up for every new character as well.

That is a pretty extreme conclusion to take this to. I do like everyone using the same method so that there is fairness and equal possibility of good and bad outcomes. I also don't want every stat to be low. I want a range of stats possible. For me half the fun is hoping for an 18 but knowing it could be any result between 3-18. So 3d4 straight down fails to achieve what i am after. Mixing it up so I am using one method and everyone else another also doesn't quite achieve what I want. I guess I just don't know why it is so hard to accept someone might like this and they are not trying to pull a fast one over you.

 

[Bedrockgames]

I quite understand that. I also am fine with having 6's and 7's on occasion. I'd even say its one of the things I tend to miss about random generation is you tended to see and play more weird characters. But again, if you aren't fine with a hopeless character, if you think that there are characters you can't work with then you need to admit that you don't like randomness. What you really want is a character generation system that delivers to you a character you can work with, something random generation is never guaranteed to do. Even if you do 9d6 take best 3, there is a good chance that the score will be below 16 - which defeats the whole intention of the method which is to ensure that you get a 16+ in your prime requisite. If that is what the real intention is, stop pretending to like random.
.

They are not pretending to like random. They want enough randomness to make it unpredictable and exciting, but not so random they get lots of bad characters. Again I can't speak for others, but based on what they are saying, it seems they are actually telling you the truth when they say they like random results and find 4d6 drop the lowest gives them the kind of ranges they prefer. Then you take it to rhetorical extremes saying they don't want any bad results at all, or because they chuck characters that are completely hopeless they can't enjoy randomness at all.