Using a D20

[The Nexus]

One of the things I've always wondered is what difference would it make if you switched from 3d6 to 1d20.

 

I'm not the best in Math, but some things are pretty obvious, for example 8- roll would be easier to make, but I believe that 11- would be pretty close... Or am I off in that?

 

Has anyone ever done this? If so what were the results? If I try this what do you think I would need to change or what pitfalls am I not thinking about?

[Steve Long]

Re: Using a D20

 

6E2 280 has a chart listing the percentage chance to roll any given number or less on 3d6. Since a 1d20 roll is an easy linear calculation of percentage, you can compare the two to determine relative effectiveness.

 

For example:

8- on 3d6 = 25.93% chance

8- on 1d20 = 40% chance

 

10- on 3d6 = 50%

10- on 1d20 = 50%

 

11- on 3d6 = 62.5% chance

11- on 1d20 = 55% chance

[Tom Carman]

Re: Using a D20

 

6E2 280 has a chart listing the percentage chance to roll any given number or less on 3d6. Since a 1d20 roll is an easy linear calculation of percentage, you can compare the two to determine relative effectiveness.

 

For example:

8- on 3d6 = 25.93% chance

8- on 1d20 = 40% chance

 

10- on 3d6 = 50%

10- on 1d20 = 50%

 

11- on 3d6 = 62.5% chance

11- on 1d20 = 55% chance

 

That doesn't quite convey the real difference because the numbers are clustered around the middle range. The ends of the bell curve are where it gets extreme. A d20 is always a 5% chance (1 in 20) for any given number. By contrast, a 3 or an 18 on 3d6 is a 1:216 chance IIRC.

[Christopher]

Re: Using a D20

 

That doesn't quite convey the real difference because the numbers are clustered around the middle range. The ends of the bell curve are where it gets extreme. A d20 is always a 5% chance (1 in 20) for any given number. By contrast' date=' a 3 or an 18 on 3d6 is a 1:216 chance IIRC.[/quote']

There is a great difference. And it doesn't even ends with the propabilites. What about the difference in the range of values (3d6 only have 16 sums/results, a d20 has 25% more), that would affect how much a modifier is worth (i.e. 1/2 can have a totally different statistical impact on an 3d6 and d20 System). When you are intersted in the math, I could give you the excat propabilites for every possible result of 3d6*

 

* Well, actually anything from 1d6 to 15d6 and I have java code that can go up to 23d6. I just don't had the time or interest to go to the technical limit. I made this for a "xd6 vs. Target number or less" System (remotely similar to Star Wars d6).

[Kraven Kor]

Re: Using a D20

 

I would sooner run D20 using 3d6 than run HERO using a D20; for the very reason of how much I hate the randomness of D20. Conan clone rolls a 1, drops his sword, then is killed by the goblin who rolls a 20 with his pointy stick for x2 damage

 

But, do as you will

[The Nexus]

Re: Using a D20

 

I think that randomness is more (dare I use this word when talking about role-playing games?) realistic. I like the possibility of anyone having the chance to hit anyone. I don't typically do fumbles because of how often they'll effect the PC's versus how often NPC's get nailed with it.

[Xavier Onassiss]

Re: Using a D20

 

I think that randomness is more (dare I use this word when talking about role-playing games?) realistic. I like the possibility of anyone having the chance to hit anyone. I don't typically do fumbles because of how often they'll effect the PC's versus how often NPC's get nailed with it.

 

The 'flat' probability distribution of the d20 is definitely not realistic in some cases:

 

When a top-ranked expert in a given skill, using the best equipment available, under ideal circumstances and every possible variable in his favor, still has a 5% chance of failure, something is very wrong. The long odds of rolling an 18 on 3d6 much better reflects the nature of 'fluke' failures like this.

 

At the other end of the spectrum, the d20 player's "well I've always got a 5% chance to succeed no matter how bad I am at this" mantra results in some pretty unrealistic behavior, as well. The longshot of rolling a 3 on 3d6 discourages a lot of unrealistic stupid stunts that waste game-time.

 

As for the 'middle ground', all else being equal, I'll take the 3d6 because predictability is a feature, not a bug. It's easier to balance encounters when you have a fair idea how well you can expect your PC's and NPC's to roll, and it gives the PC's some idea what their characters should be capable of. Most heroic characters are competent enough to realistically assess their own abilities -- if they're not, they should have complications such as Overconfident or Unsure of Self, and be encouraged to role-play that way.

[CrosshairCollie]

Re: Using a D20

 

I would sooner run D20 using 3d6 than run HERO using a D20; for the very reason of how much I hate the randomness of D20. Conan clone rolls a 1, drops his sword, then is killed by the goblin who rolls a 20 with his pointy stick for x2 damage

 

But, do as you will

 

Of course, if rolling a 1 means something other than 'you missed', your GM is houseruling, and you have bigger problems.

[Xavier Onassiss]

Re: Using a D20

 

I've sometimes considered converting the d20 system to a 4d6 system, using six-sided dice numbered zero thru five (instead of one thru six); the result would be 0-20 (not 1-20), with the results nicely weighted towards the middle. (Chance of rolling a 20: 1 in 1296.)

[Darbor]

Re: Using a D20

 

I've sometimes considered converting the d20 system to a 4d6 system' date=' using six-sided dice numbered zero thru five (instead of one thru six); the result would be 0-20 (not 1-20), with the results nicely weighted towards the middle. (Chance of rolling a 20: 1 in 1296.)[/quote']

Very interesting! I prefer center weighted systems. So let me turn it around on you. Assuming D&D has been carefully crafted over years of experience to work with D20, how well does the weighted dice play in the D20 systems? Are there problems? Did you have to make adjustments to make it play well?

[Xavier Onassiss]

Re: Using a D20

 

Very interesting! I prefer center weighted systems. So let me turn it around on you. Assuming D&D has been carefully crafted over years of experience to work with D20' date=' how well does the weighted dice play in the D20 systems? Are there problems? Did you have to make adjustments to make it play well?[/quote']

 

I haven't actually done this yet -- just tinkered around with the numbers. But I'm dubious about D&D being 'carefully crafted' in any edition. WOTC made an effort to balance everything in 4th edition, but IMHO, only partially succeeded. However, considering they balanced it based on "average" die rolls, weighting the rolls towards the average would probably be an improvement. Critical hit ranges for weapons would have to be adjusted, assuming you'd want to keep the same % chance for crits. (which I wouldn't)

 

Prior to 4th ed, there wasn't any 'game balance' in D&D that I could see: AC didn't scale up with attack bonuses; spell save DC's didn't scale up with saving throw bonuses; etc. (And in spite of both these problems, fighters almost always came off 2nd to spellcasters. WTF?) None of it made any sense, from a game balance point of view. I had some fun with it, but my inner game designer had to hold his nose while I was doing it.

[Burrito Boy]

Re: Using a D20

 

The 'flat' probability distribution of the d20 is definitely not realistic in some cases:

 

When a top-ranked expert in a given skill, using the best equipment available, under ideal circumstances and every possible variable in his favor, still has a 5% chance of failure, something is very wrong. The long odds of rolling an 18 on 3d6 much better reflects the nature of 'fluke' failures like this.

 

At the other end of the spectrum, the d20 player's "well I've always got a 5% chance to succeed no matter how bad I am at this" mantra results in some pretty unrealistic behavior, as well. The longshot of rolling a 3 on 3d6 discourages a lot of unrealistic stupid stunts that waste game-time.

 

As for the 'middle ground', all else being equal, I'll take the 3d6 because predictability is a feature, not a bug. It's easier to balance encounters when you have a fair idea how well you can expect your PC's and NPC's to roll, and it gives the PC's some idea what their characters should be capable of. Most heroic characters are competent enough to realistically assess their own abilities -- if they're not, they should have complications such as Overconfident or Unsure of Self, and be encouraged to role-play that way.

 

You can sometimes Take 10 or Take 20, minimizing the possibility of failure. At least in 3.x Edition. I'm not sure about 4E since I've never played it.

[phookz]

Re: Using a D20

 

Prior to 4th ed' date=' there wasn't any 'game balance' in D&D that I could see: AC didn't scale up with attack bonuses; spell save DC's didn't scale up with saving throw bonuses; etc. (And in spite of both these problems, fighters almost always came off 2nd to spellcasters. WTF?) None of it made any sense, from a game balance point of view. I had some fun with it, but my inner game designer had to hold his nose while I was doing it.[/quote']

 

The thing that I never liked about D&D is that the AC does scale, but it scales under the assumption that your characters will acquire magic items to get it to scale. You want to run a low-magic campaign? Tough - the system isn't designed to work that way and you'll have to house rule something in to keep balance.

 

And yeah, spellcasters have never scaled in that system. It does appear that 4e has improved on that to some degree, but I like 4e less than I did the earlier versions. If I want to play an MMO I won't do it with pen and paper.

 

To get back on topic - I agree with others that the bell curve is a feature. It's still random, it's just that the distribution is not linear and flat. It's the bell curve that means a +3 is a significant bonus to skill rolls and +5 is almost assured success. Without the bell curve these values have a lot less meaning. Switch to a d20 and you'll have a very different game experience.

 

For example, consider an 11- skill roll. Out of the box you have a 62.5% chance of success with 3D6. With a D20 you have a 55% chance. Arguable close enough.

 

Now add +5 to the roll. An 11- skill roll becomes a 16-, which has a 98.1% chance of success with 3D6. However, on a D20 only has a 80% chance of success.

[dmjalund]

Re: Using a D20

 

in 4e, i've encountered 'taking 10' but not 'taking 20'

[The Nexus]

Re: Using a D20

 

I hadn't really considered that the game was designed around the Bell Curve when they determined what the bonuses should be. That would be a much larger change than I was looking for, thanks for the help all.

[Orion]

Re: Using a D20

 

I've never liked the d6 systems for skills and combat. It's the thing I dislike most about the Hero system. I would prefer a percentile-based system any day. As I see it, I want a +1 to the roll to always mean the same thing. If two characters each get a +1 to their roll, but their bases are at 11 and 14, it has a very different affect for each. I want to be able to say this gives you a 5% or 8% or whatever better chance - and everyone should get the same bonus. It should also be the same affect every single time, whereas now it can be quite different depending if other penalties and bonuses are in action. I guess I'd have a slight preference for d6 over d20, but see both as a poor substitute for d100.

[Markdoc]

Re: Using a D20

 

Very interesting! I prefer center weighted systems. So let me turn it around on you. Assuming D&D has been carefully crafted over years of experience to work with D20' date=' how well does the weighted dice play in the D20 systems? Are there problems? Did you have to make adjustments to make it play well?[/quote']

 

I second Xaviers' comment about "well-crafted", but to answer the question D&D is designed to work with the linear-scale D20. It's relatively easy to stack on the bonuses - indeed, the system is designed for you to be able to stack on the bonuses. If you used the D20 system as-is apart from dice weighted to give a bell curve, it'd be relatively easy to reach "almost never fail" levels. You'd need to rethink the whole skill system, and bonuses from spells & characteristics to make it work. In addition, combat assumes that you can hit the higher rolls regularly: you'd need to redo criticals, plus look at dropping AC bonuses: otherwise a character that can exceed AC 30 (not that hard to do even sans fromage) is going to be untouchable by anything other than high-level characters.

 

Of course, you could ignore all of that and just make like you're playing Exalted!

 

cheers, Mark

[Christopher]

Re: Using a D20

 

I've never liked the d6 systems for skills and combat. It's the thing I dislike most about the Hero system. I would prefer a percentile-based system any day. As I see it' date=' I want a +1 to the roll to always mean the same thing. If two characters each get a +1 to their roll, but their bases are at 11 and 14, it has a very different affect for each.[/quote']

"It's not a bug, it's a feature" - inofficial Microsoft Slogan.

 

But seriously: I think it is an advantage that a weak character(11-) get's a better effective bonus from +1 than a good one(14-). One the other hand, a good one has less to worry about negative modifiers. So the game is more about base-values and internal bonusses than about stacking circumstance bonusses.

 

About D&D: I seriously dislike the entire ac-system. It's just one value, and it determines if you are hit or not. There is no "partial hit". And on the higher levels, fighting is more about HP-atrittion than about tatics. Also, it doesn't fared well with modern weapons (at least the ones in the gm-book). Just think about the fact that a Full plate offers the same bonus against a sword and musket. That's not how it worked, armors were way weaker against firearms. Also, there is no easy way to play an unarmored fighter (there are the monk and some prestige classes, but they ain't easy and compatible with half the concepts I could cook up).

Also, they have no defense skill (or no attack roll, in the chase of spells). It way always just a "Roll vs. target number", but fighting should be more like skill-contest:

"Roll-vs-Roll".

[Xavier Onassiss]

Re: Using a D20

 

I've never liked the d6 systems for skills and combat. It's the thing I dislike most about the Hero system. I would prefer a percentile-based system any day. As I see it' date=' I want a +1 to the roll to always mean the same thing. If two characters each get a +1 to their roll, but their bases are at 11 and 14, it has a very different affect for each. I want to be able to say this gives you a 5% or 8% or whatever better chance - and everyone should get the same bonus. It should also be the same affect every single time, whereas now it can be quite different depending if other penalties and bonuses are in action. I guess I'd have a slight preference for d6 over d20, but see both as a poor substitute for d100.[/quote']

 

Actually, a +1 bonus isn't the same for everyone using the d20, either. If I need a 20 to hit, my chances are 1 in 20. With a +1 it goes up to 2 in 20: from my perspective that's not a 5% improvement, it's twice as good: 100% improvement.

 

For the guy who only needs an 11 to hit, he starts off with 10 in 20 chances to hit, and the +1 gives him 11 in 20. For him, it's only a 10% improvement.

[Darbor]

Re: Using a D20

 

"It's not a bug, it's a feature" - inofficial Microsoft Slogan.

 

But seriously: I think it is an advantage that a weak character(11-) get's a better effective bonus from +1 than a good one(14-). One the other hand, a good one has less to worry about negative modifiers. So the game is more about base-values and internal bonusses than about stacking circumstance bonusses.

 

I hadn't really thought about the different ways pluses would scale. In our game we deal with minuses much more often than pluses, and anyone actually planning to use a skill has at least 11-, so it works out that those with the higher skill have an increasing chance of success. But we also play in shades of gray, often times how much you make the roll by can make a difference in the effectiveness of the use of the skill. In this case center weighting is a very good thing as it means that the PC's that have bought up the skill have a much better chance of making the roll and also increase the chance that they will make it by a wider margin and thus get a better result.

[Fearghus]

Re: Using a D20

 

It occurs to me that the bell curve on 3d6 rolls and the cost structure associated with CLs and Characteristics are interconnected with each other. The concept of the better you are at something the more difficult it is to get even better is a common theme in many RPGs. Most RPGs create an experience or character point bell curve... hero uses the natural bell curve in the dice. So rather than making each bonus more expensive than the previous bonus... each bonus becomes less effective, but they all cost the same.

 

I for one would fee it necessary to adjust the cost structure of at least CLs if I were to attempt to convert to a D20 or a percentile system. I always look at it this way. the more you mess with the system... the more you have to document your own house rules... It is important that all players know what the rules actually are after all... So I house rule some things, but not many.

[Christopher]

Re: Using a D20

 

You have a important point there, Fearghus:

d&d put a lot of effort into limiting the "bonus cramming", with the varius types of bonusses (enchantment, sacred...). Or the Skill limits by basing them on level.

And magic items prices/levels are based on effect^2, to create an exponential growth curve.

 

So it seems the hero system avoided all this, just with using a different dice system.

[Burrito Boy]

Re: Using a D20

 

The thing that I never liked about D&D is that the AC does scale, but it scales under the assumption that your characters will acquire magic items to get it to scale. You want to run a low-magic campaign? Tough - the system isn't designed to work that way and you'll have to house rule something in to keep balance.

 

And yeah, spellcasters have never scaled in that system. It does appear that 4e has improved on that to some degree, but I like 4e less than I did the earlier versions. If I want to play an MMO I won't do it with pen and paper.

 

To get back on topic - I agree with others that the bell curve is a feature. It's still random, it's just that the distribution is not linear and flat. It's the bell curve that means a +3 is a significant bonus to skill rolls and +5 is almost assured success. Without the bell curve these values have a lot less meaning. Switch to a d20 and you'll have a very different game experience.

 

For example, consider an 11- skill roll. Out of the box you have a 62.5% chance of success with 3D6. With a D20 you have a 55% chance. Arguable close enough.

 

Now add +5 to the roll. An 11- skill roll becomes a 16-, which has a 98.1% chance of success with 3D6. However, on a D20 only has a 80% chance of success.

 

I'm a math hater so I have to ask what might be a stupid question. If 16- has a 98.1% chance of success, is their any reason to spend the character points to raise a skill to 17- or 18-?

[dmjalund]

Re: Using a D20

 

I'm a math hater so I have to ask what might be a stupid question. If 16- has a 98.1% chance of success' date=' is their any reason to spend the character points to raise a skill to 17- or 18-?[/quote']Difficulty penalties

[Burrito Boy]

Re: Using a D20

 

difficulty penalties

 

D'oh!!! I said it was a stupid question.