I wonder how many sets of stats you would have to roll to get straight 12's? Seems so statistically unlikely that is a waste of time even debating that type of character.
For those who are curious, the odds of no stats above 12 are 1.8%, and the odds of no stats above 13 are 7.2% (all stats 12 is much less likely but not a very important case).
Good! Not amazing but I got options this could work!
2. 10 15 11 11 12 17
Wow pretty good! Going to start with an 18 in my main stat can't beat that.
3. 14 13 7 9 11 5
Holy sh... This is terrible. Negative mods on three stats... But I guess good for you guys...
The fact we discuss averages and chance doesn't mean it doesn't happen and that it will never have an effect. This method just produced two characters with 21 points difference in stats....
The point being that third person will likely ask for a reroll and if it's bad then they do Point buy and still sit back with the other two with no weaknesses.
If they did the Coville method that last character is utter trash as it's 15 STR and 7 Con makes it terrible for any build that uses the best stat.
But the internet says I do! The internet can't possibly be wrong... can it?
I mean...If I wanted to play an average or below average character I would play Accountants and Apartments not Dungeons and Dragons.
Ok...? I mean, just because the rest of your party has straight 18s, does not mean you can't have fun with 16, 14, 10, 8, 8, 8.
Even then you are about 90% of the way to a decent character.
We are talking about
12 12 12 12 12 12
or
8 8 10 11 14 7
Those are not in the same ballpark as 16, 14, 10, 8, 8, 8. as you at least have a 16 in your main stat and a 14 in CON. You could make a pretty effective STR fighter with that.
Or a very good Artificer (especially if they survive to level 14 which isn't difficult, especially if you play an armorer) or an almost even better moon druid.
For those who are curious, the odds of no stats above 12 are 1.8%, and the odds of no stats above 13 are 7.2% (all stats 12 is much less likely but not a very important case).
Straight 12s is just an example used for simplicity's sake.
Good! Not amazing but I got options this could work!
2. 10 15 11 11 12 17
Wow pretty good! Going to start with an 18 in my main stat can't beat that.
3. 14 13 7 9 11 5
Holy sh... This is terrible. Negative mods on three stats... But I guess good for you guys...
The fact we discuss averages and chance doesn't mean it doesn't happen and that it will never have an effect. This method just produced two characters with 21 points difference in stats....
The point being that third person will likely ask for a reroll and if it's bad then they do Point buy and still sit back with the other two with no weaknesses.
If they did the Coville method that last character is utter trash as it's 15 STR and 7 Con makes it terrible for any build that uses the best stat.
As the DM in this group, I would not allow a re-roll, but I would put a 1 in front of that 5 making it a 15. I also don't do stats straight down the line so they could put those stats where ever they like.
But I also don't fault other people for choosing how they want to do things at their table. If you want to use point buy, standard array, draw cards, throw darts, spin a wheel or any other method you can think of, then go for it. Just don't tell me that my fun is wrong and expect me to pay any attention to you.
"As the DM in this group, I would not allow a re-roll, but I would put a 1 in front of that 5 making it a 15. I also don't do stats straight down the line so they could put those stats where ever they like."
And we come full circle as the the method cannot stand on it's own merits and needs help to function.
Also this shows that the Coville method is only good if you roll well.
However you want to go with it it's your table but it's pretty telling that you have to modify the method to make it work.
"As the DM in this group, I would not allow a re-roll, but I would put a 1 in front of that 5 making it a 15. I also don't do stats straight down the line so they could put those stats where ever they like."
And we come full circle as the the method cannot stand on it's own merits and needs help to function.
Also this shows that the Coville method is only good if you roll well.
However you want to go with it it's your table but it's pretty telling that you have to modify the method to make it work.
Every single thing about this game encourages you to modify it to work with your table and how your group wants to have fun. Why would stat generation be any different?
Because 1d20 provides a uniform distribution, with an equal probability of getting every number between 1 and 20.
The original formulation of 3D6 was used because when you add 3 dice together, there is a very low chance of getting extreme numbers, and a high chance of getting the numbers in the middle (around 10-11 total on 3D6). So for example, the chance of rolling minimum on 1d20 (1) is 5%. It's the same chance to roll a max (20). On the other hand, the chance of throwing the minimum or maximum on 3D6 is much lower (1 in 216 for a 3 or 18, or less than 1/10th as likely). On the other hand, the chance of throwing a 10 or an 11 (middle values) is much higher on 3D6 (around 13% each, or 25% total). It's only a 10% chance to throw a 10 or 11 on 1d20.
4D6-drop-1 pushes the average higher, but rolling multiple dice still reduces the chance of extreme values. You'd still have to get 3 out of 4 dice to be a 6, which is a lower probability than throwing an 18 on 1d20.
Rollback Post to RevisionRollBack
WOTC lies. We know that WOTC lies. WOTC knows that we know that WOTC lies. We know that WOTC knows that we know that WOTC lies. And still they lie.
Because of the above (a paraphrase from Orwell) I no longer post to the forums -- PM me if you need help or anything.
"As the DM in this group, I would not allow a re-roll, but I would put a 1 in front of that 5 making it a 15. I also don't do stats straight down the line so they could put those stats where ever they like."
And we come full circle as the the method cannot stand on it's own merits and needs help to function.
Also this shows that the Coville method is only good if you roll well.
However you want to go with it it's your table but it's pretty telling that you have to modify the method to make it work.
Every single thing about this game encourages you to modify it to work with your table and how your group wants to have fun. Why would stat generation be any different?
Because of bounded accuracy, mods being used for number of uses for abilities, and classes that can be very stat dependant (monk) stats need to be not only relatively close across the board but also within the parameters the designs had in mind when you are trying to calculate encounters as a DM.
You have two methods that produce this. And one that just craps on that idea.
Rolling introduces additional variations to a system that played straight up already has balance issues.
An experienced DM can likely handle this and I've seen it work first hand.
However, new players and DMs this is the worst way to approach it.
It can work yes but to be honest most DMs don't have the experience to pull it off.
Because 1d20 provides a uniform distribution, with an equal probability of getting every number between 1 and 20.
The original formulation of 3D6 was used because when you add 3 dice together, there is a very low chance of getting extreme numbers, and a high chance of getting the numbers in the middle (around 10-11 total on 3D6). So for example, the chance of rolling minimum on 1d20 (1) is 5%. It's the same chance to roll a max (20). On the other hand, the chance of throwing the minimum or maximum on 3D6 is much lower (1 in 216 for a 3 or 18, or less than 1/10th as likely). On the other hand, the chance of throwing a 10 or an 11 (middle values) is much higher on 3D6 (around 13% each, or 25% total). It's only a 10% chance to throw a 10 or 11 on 1d20.
4D6-drop-1 pushes the average higher, but rolling multiple dice still reduces the chance of extreme values. You'd still have to get 3 out of 4 dice to be a 6, which is a lower probability than throwing an 18 on 1d20.
Also old editions stats really didn't matter as much as they do now.
It's a holdover from a system where it really didn't matter so why not roll?
It just goes to show that you and I have had much different experiences and expectations. You play your way and I will continue to play mine and we will both be happy.
It just goes to show that you and I have had much different experiences and expectations. You play your way and I will continue to play mine and we will both be happy.
Never said otherwise!
Just answering the question on why it's frowned on.... Especially for new groups.
Whatever works is good for you guys. I'm just saying new players need to know the potential significant CONs of using it.
I don't want to do random rolls. It adds a variance to PC power that can be substantial enough to be a disincentive for someone to continue playing. I like to do what I can to fulfill my party role as well as possible. If one players rolls an 18 in their primary stat, they can be starting the game with a max bonus in that stat (depending on race choice). This not only gives them a great power level, but it also frees them up for additional power as they level through feats instead of ASis. If I low roll, and need to use all my ASIs to bolster my primary stat, then I'm falling behind in power relative to party mates.
Would I bail on a campaign because of this? No, I've seen low stat characters do fun things through future lucky rolls. But it sure would leave a bad taste, and probably make me feel like I'm not doing as much for the party as I could have if I only didn't get unlucky with a single die roll in session 0.
With standard array or point buy, everyone gets to make decisions for their characters that matter forever. PC power still has variance based on choices made, but (to me), that feels a lot better since it is based on choices and not luck.
So I may be wrong but what irks me is the fact that people lie about their reasoning behind it a lot I think. If they were honest and said "I do it so I can potentially get great stats because I find that fun" I would be more ok with it as thats honesty at least.
I do it because it's how I learned to do it way back when.
I also like standard array to keep things fair and for beginners.
Rollback Post to RevisionRollBack
"Sooner or later, your Players are going to smash your railroad into a sandbox."
-Vedexent
"real life is a super high CR."
-OboeLauren
"............anybody got any potatoes? We could drop a potato in each hole an' see which ones get viciously mauled by horrible monsters?"
But I also don't fault other people for choosing how they want to do things at their table. If you want to use point buy, standard array, draw cards, throw darts, spin a wheel or any other method you can think of, then go for it. Just don't tell me that my fun is wrong and expect me to pay any attention to you.
Bear in mind that the topic of this thread is "Why is 4d6 so awful", so it's pretty much asking for criticism.
The basic problem is that random stat generation vs standard array or point build is a tradeoff. Randomization does make character creation more exciting (and if you don't allow reordering stats, can produce unexpected characters), at a cost of characters being potentially significantly unbalanced relative to one another. For one-shots or campaigns with high levels of character churn that can be a net gain, but if you're going to be playing the same character for twenty sessions, having "how good is this character going to be" defined by five minutes of die rolling places an unduly high weight on those rolls.
But I also don't fault other people for choosing how they want to do things at their table. If you want to use point buy, standard array, draw cards, throw darts, spin a wheel or any other method you can think of, then go for it. Just don't tell me that my fun is wrong and expect me to pay any attention to you.
Bear in mind that the topic of this thread is "Why is 4d6 so awful", so it's pretty much asking for criticism.
The basic problem is that random stat generation vs standard array or point build is a tradeoff. Randomization does make character creation more exciting (and if you don't allow reordering stats, can produce unexpected characters), at a cost of characters being potentially significantly unbalanced relative to one another. For one-shots or campaigns with high levels of character churn that can be a net gain, but if you're going to be playing the same character for twenty sessions, having "how good is this character going to be" defined by five minutes of die rolling places an unduly high weight on those rolls.
4D6-drop-1 pushes the average higher, but rolling multiple dice still reduces the chance of extreme values. You'd still have to get 3 out of 4 dice to be a 6, which is a lower probability than throwing an 18 on 1d20.
Rolling an 18 or higher is more likely on a d20 (15%) yes. However, the more dice you roll will increase the average as long as you drop the lowest. Roll 4d6, drop lowest give about a 9.3% chance of rolling an 18. If you roll 5d6, drop the lowest two, that will increase the chance even more. The extremes will thus be biased towards higher numbers.
4D6-drop-1 pushes the average higher, but rolling multiple dice still reduces the chance of extreme values. You'd still have to get 3 out of 4 dice to be a 6, which is a lower probability than throwing an 18 on 1d20.
Rolling an 18 or higher is more likely on a d20 (15%) yes. However, the more dice you roll will increase the average as long as you drop the lowest. Roll 4d6, drop lowest give about a 9.3% chance of rolling an 18. If you roll 5d6, drop the lowest two, that will increase the chance even more. The extremes will thus be biased towards higher numbers.
That’s a relatively high percentage chance. In theory, your odds are slightly better than half of rolling at least one 18 on 4d6 drop the lowest. (I think that’s right. I freely admit I’m terrible at math.)
Rollback Post to RevisionRollBack
To post a comment, please login or register a new account.
I wonder how many sets of stats you would have to roll to get straight 12's? Seems so statistically unlikely that is a waste of time even debating that type of character.
She/Her Player and Dungeon Master
For those who are curious, the odds of no stats above 12 are 1.8%, and the odds of no stats above 13 are 7.2% (all stats 12 is much less likely but not a very important case).
Ok so what's the odds across a whole party?
I just rolled 3 sets of stats:
1. 14 11 13 12 14 14
Good! Not amazing but I got options this could work!
2. 10 15 11 11 12 17
Wow pretty good! Going to start with an 18 in my main stat can't beat that.
3. 14 13 7 9 11 5
Holy sh... This is terrible. Negative mods on three stats... But I guess good for you guys...
The fact we discuss averages and chance doesn't mean it doesn't happen and that it will never have an effect. This method just produced two characters with 21 points difference in stats....
The point being that third person will likely ask for a reroll and if it's bad then they do Point buy and still sit back with the other two with no weaknesses.
If they did the Coville method that last character is utter trash as it's 15 STR and 7 Con makes it terrible for any build that uses the best stat.
Or a very good Artificer (especially if they survive to level 14 which isn't difficult, especially if you play an armorer) or an almost even better moon druid.
Straight 12s is just an example used for simplicity's sake.
As the DM in this group, I would not allow a re-roll, but I would put a 1 in front of that 5 making it a 15. I also don't do stats straight down the line so they could put those stats where ever they like.
But I also don't fault other people for choosing how they want to do things at their table. If you want to use point buy, standard array, draw cards, throw darts, spin a wheel or any other method you can think of, then go for it. Just don't tell me that my fun is wrong and expect me to pay any attention to you.
She/Her Player and Dungeon Master
"As the DM in this group, I would not allow a re-roll, but I would put a 1 in front of that 5 making it a 15. I also don't do stats straight down the line so they could put those stats where ever they like."
And we come full circle as the the method cannot stand on it's own merits and needs help to function.
Also this shows that the Coville method is only good if you roll well.
However you want to go with it it's your table but it's pretty telling that you have to modify the method to make it work.
Every single thing about this game encourages you to modify it to work with your table and how your group wants to have fun. Why would stat generation be any different?
She/Her Player and Dungeon Master
Uh dumb question, why not instead of 4d6 use 1d20
: Systems Online : Nikoli_Goodfellow Homebrew : My WIP Homebrew Class :
(\_/)
( u u)
o/ \🥛🍪 Hey, take care of yourself alright?
Because unmodified stats fall between 3 and 18.
However if that is what your group wants to do then go for it.
She/Her Player and Dungeon Master
Because 1d20 provides a uniform distribution, with an equal probability of getting every number between 1 and 20.
The original formulation of 3D6 was used because when you add 3 dice together, there is a very low chance of getting extreme numbers, and a high chance of getting the numbers in the middle (around 10-11 total on 3D6). So for example, the chance of rolling minimum on 1d20 (1) is 5%. It's the same chance to roll a max (20). On the other hand, the chance of throwing the minimum or maximum on 3D6 is much lower (1 in 216 for a 3 or 18, or less than 1/10th as likely). On the other hand, the chance of throwing a 10 or an 11 (middle values) is much higher on 3D6 (around 13% each, or 25% total). It's only a 10% chance to throw a 10 or 11 on 1d20.
4D6-drop-1 pushes the average higher, but rolling multiple dice still reduces the chance of extreme values. You'd still have to get 3 out of 4 dice to be a 6, which is a lower probability than throwing an 18 on 1d20.
WOTC lies. We know that WOTC lies. WOTC knows that we know that WOTC lies. We know that WOTC knows that we know that WOTC lies. And still they lie.
Because of the above (a paraphrase from Orwell) I no longer post to the forums -- PM me if you need help or anything.
Because of bounded accuracy, mods being used for number of uses for abilities, and classes that can be very stat dependant (monk) stats need to be not only relatively close across the board but also within the parameters the designs had in mind when you are trying to calculate encounters as a DM.
You have two methods that produce this. And one that just craps on that idea.
Rolling introduces additional variations to a system that played straight up already has balance issues.
An experienced DM can likely handle this and I've seen it work first hand.
However, new players and DMs this is the worst way to approach it.
It can work yes but to be honest most DMs don't have the experience to pull it off.
Also old editions stats really didn't matter as much as they do now.
It's a holdover from a system where it really didn't matter so why not roll?
Now it does matter and it creates uneven play
It just goes to show that you and I have had much different experiences and expectations. You play your way and I will continue to play mine and we will both be happy.
She/Her Player and Dungeon Master
Never said otherwise!
Just answering the question on why it's frowned on.... Especially for new groups.
Whatever works is good for you guys. I'm just saying new players need to know the potential significant CONs of using it.
I don't want to do random rolls. It adds a variance to PC power that can be substantial enough to be a disincentive for someone to continue playing. I like to do what I can to fulfill my party role as well as possible. If one players rolls an 18 in their primary stat, they can be starting the game with a max bonus in that stat (depending on race choice). This not only gives them a great power level, but it also frees them up for additional power as they level through feats instead of ASis. If I low roll, and need to use all my ASIs to bolster my primary stat, then I'm falling behind in power relative to party mates.
Would I bail on a campaign because of this? No, I've seen low stat characters do fun things through future lucky rolls. But it sure would leave a bad taste, and probably make me feel like I'm not doing as much for the party as I could have if I only didn't get unlucky with a single die roll in session 0.
With standard array or point buy, everyone gets to make decisions for their characters that matter forever. PC power still has variance based on choices made, but (to me), that feels a lot better since it is based on choices and not luck.
I do it because it's how I learned to do it way back when.
I also like standard array to keep things fair and for beginners.
"Sooner or later, your Players are going to smash your railroad into a sandbox."
-Vedexent
"real life is a super high CR."
-OboeLauren
"............anybody got any potatoes? We could drop a potato in each hole an' see which ones get viciously mauled by horrible monsters?"
-Ilyara Thundertale
Bear in mind that the topic of this thread is "Why is 4d6 so awful", so it's pretty much asking for criticism.
The basic problem is that random stat generation vs standard array or point build is a tradeoff. Randomization does make character creation more exciting (and if you don't allow reordering stats, can produce unexpected characters), at a cost of characters being potentially significantly unbalanced relative to one another. For one-shots or campaigns with high levels of character churn that can be a net gain, but if you're going to be playing the same character for twenty sessions, having "how good is this character going to be" defined by five minutes of die rolling places an unduly high weight on those rolls.
I wish I could like this comment 1000 times.
Rolling an 18 or higher is more likely on a d20 (15%) yes. However, the more dice you roll will increase the average as long as you drop the lowest. Roll 4d6, drop lowest give about a 9.3% chance of rolling an 18. If you roll 5d6, drop the lowest two, that will increase the chance even more. The extremes will thus be biased towards higher numbers.
That’s a relatively high percentage chance. In theory, your odds are slightly better than half of rolling at least one 18 on 4d6 drop the lowest. (I think that’s right. I freely admit I’m terrible at math.)