D&D 5E Randomness and D&D

[prabe]

I know that the last games of 1e I was involved with, we went with max hp at 1st level, but that was a heavily houseruled version of 1e, right before 3e came out. (Yes, that group skipped 2e.)

 

[Mannahnin]

It's definitely been interesting over time seeing various common house rules become official. Whether more generous ability rolling methods being tried in the 70s and made official in AD&D, 50 coins to a pound going from a Dragon article and common discussion in 1E to official in 2E, various rules for flanking and for max HP at first level we were using in 2E becoming official in 3E...

 

[UngeheuerLich]

First off, great post! Loved it! Two thumbs up!

First, the d20 is too swingy. IMO I think skill should trump randomness, so using something non-linear like 2d10 (even for attacks) would be better.

We have tried this and we tried 3d6 and noticed, that 1d20 randomness is a feature, not a bug. D&D's main attraction is, that everyone can try to take part in everything. With 3d6, a 3 point difference in skill can mean: don't bother trying.

I can imagine scenarios where you would like that, for example heist scenarios, where you might have tools to reduce DC and where you want specialists that need to do certain things. But as a general rule it feels so much disabling instead of enabling roleplaying.

 

[Mannahnin]

We have tried this and we tried 3d6 and noticed, that 1d20 randomness is a feature, not a bug. D&D's main attraction is, that everyone can try to take part in everything. With 3d6, a 3 point difference in skill can mean: don't bother trying.

I can imagine scenarios where you would like that, for example heist scenarios, where you might have tools to reduce DC and where you want specialists that need to do certain things. But as a general rule it feels so much disabling instead of enabling roleplaying.

I definitely remember observing the huge difference in feel between, say, AD&D and GURPS (which uses a 3d6 resolution system, as no doubt some of you are aware) back in the 80s and 90s.

The other big trick with moving to a bell curved multi-dice resolution system is the varying and large impact of modifiers at different points along the curve. A +2 to the roll for a person with a skill of 10 is a much different impact than to a person with a skill of, say, 14. Whereas with a simple d20 a +1 is always 5% anywhere on the scale.

I agree that the swinginess of the d20 is definitely part of D&D's charm. Certainly in combat. I could more easily see moving to a bell-curved system for skills. In fact I somewhat commonly see DMs in OD&D games use ability checks on 2d6 (easy), 3d6 (medium difficulty) or 4d6 (hard task), in part because OD&D characters with their 3d6 down the line generation method, tend to have lower stats than in other editions.

 

[Ralif Redhammer]

You make a good point that the tables (and even just the available varieties) made it far more likely that a sword+1, +3 vs. undead was going to show up over a plain footman's mace+1.

This in practice didn't matter much; the magic weapons were extremely heavily weighted towards longswords with greatswords in second (as in something like twice as many magic swords as all other magic weapons combined and 70% longswords, 25% greatswords, and 5% other - with scimitars coming under other weapons). And clerics, of course, couldn't wield edged weapons. Also longswords and greatswords were the best because of the extra damage they got against large creatures while cleric weapons got the same or sometimes less damage when attacking large or larger.

 

[James Gasik]

During the 2e era, where everyone was poring over books to try and equip their Fighters with the "best" weapon, my friends and I made a dizzying array of characters, some rather dubious (mostly because we had this insane belief that a game that let you make any character you desired would somehow make those characters good, lol- like the time I was inspired to make a Whip user after playing too much Castlevania).

What we quickly ran into, however, was that the value of these characters really depended on who was running. Some DM's had no problem with a player "finding" a powerful version of their weapon of choice. I had a tendency to flood my games with magic items because I thought they were so cool (I'd run modules, and replace "boring" items with cooler ones- like magic items from The Magister, or ones of my own design, thinking that was about the right distribution model, only to find out later that modules totally ignore the treasure type rules, lol). Another DM was super stingy and would toss you into Ravenloft and tell you how lucky you were to find a +1 knife.

But our forever DM had a compulsive gambling streak, and almost always used random treasure tables, including comprehensive ones of his own design! So we found that, in most cases, if we wanted to find a magical version of our weapon of choice, we had to go seek it out, it was probably not going to fall into our laps. Whip guy found an extremely powerful +3 mace, which I couldn't help but be grumpy about, lol.

A plethora of Katana users would sigh in dismay, while the jacked-up Sylvan Elf with Two-Weapon Style Specialization happily dual-wielded longswords. To say nothing of my Gladiator (the DM allowed me to use the Dark Sun class), who found out that, not only was he not going to find magic wrist razors, they actually suck. Fortunately, Gladiators are built to deal with that sort of thing, and he quickly moved on to other weapons (the class is proficient with all weapons, and can specialize multiple times in different weapons).

Though I would be remiss if I didn't mention my Savage Fighter. I had two 18's and I put them in Dex and Con, like a ninny. I had a Strength of 12! I couldn't wear armor, and I was specialized in the Greatspear (an absurdly powerful weapon, stat-wise, but fairly rare since it's not in the PHB).

To say I struggled mightily was an understatement, but it turns out having gobs of hit points is fairly nice in it's own right. Through sheer luck, I kept rolling stupidly high on hit point rolls, so I didn't die. Then we found a Deck of Many Things, and I ended up with Bracers of Armor, a Belt of Stone Giant Strength, and a Wish, which I immediately used to give my spear the power to "pierce even the toughest hide" (I always try to make my wishes plausible in character, even if this means I'm thrown to the mercy of the DM, lol), since we'd just had a disastrous encounter with gargoyles when playing The Sentinel.

I never got the full stats of the weapon, but I quickly graduated to one of the most powerful characters in the campaign. Now, while I had a blast, I started to go on solo adventures more and more over time, because I realized that I was outshining the other characters- suddenly, they were the ones struggling to keep up, instead of me.

And I knew what that felt like. Which is really my main issue with high randomness in D&D. It can be fun or miserable for an individual player, but rarely the whole group. Someone always ends up with a better deal, and you're not guaranteed to have a long enough timeline for the pendulum to swing the other way.

One guy gets a 17, and you get a 8. One guy gets a Broom of Flying, and another finds himself with a Ring of Swimming he doesn't have attunement slots for. And there's no guarantee of it ever balancing out.

 

[CleverNickName]

I like random. Even as the DM, I enjoy being surprised by the dice and having to improv my reactions to them.

 

[DND_Reborn]

We have tried this and we tried 3d6 and noticed, that 1d20 randomness is a feature, not a bug.

It is a bad feature, IMO.

D&D's main attraction is, that everyone can try to take part in everything

Everyone still can, you just have a more realistic appreciation for skill and ability with a non-linear system.

Here's a reason why I don't like it:

A contested check between two creatures: #1 is level 17 with +5 ability, for +11 total. #2 is level 1 with +0 ability for +2 total.

#2 has a 13.75% chance of out-right winning plus a 2.75% of a tie. This would be like a person who recently learned how to play chess vs. a grand-master.

Even if you use expertise and make #1 a total of +17, there is still a 2.50% #1 will lose.

There really should not be any chance and IMO the number should show that. Yeah, yeah, "But the DM only calls for a check when the outcome is in doubt blah blah blah and such..." Phooey! That is crappy and lazy game design.

Now, use 3d6 instead of d20. Even with +17 if #1 rolled 3, for a total of 20, and #2 rolled 18 with +2 also for a total of 20, would be a stalemate. The chances of the most unlikely outcome would be barely more than 0.002% or about 1 in 46000!!!

The other option in such cases is to do a skill-challenge type thing instead of a single roll. Perhaps like death saves with 3 success indicating victory and 3 failures is a loss. IMO 5E should have more things like this. However, for the above example, it would have to be 5 or even 7 successes/fails, not just 3.

 

[UngeheuerLich]

It is a bad feature, IMO.


Everyone still can, you just have a more realistic appreciation for skill and ability with a non-linear system.

Here's a reason why I don't like it:

A contested check between two creatures: #1 is level 17 with +5 ability, for +11 total. #2 is level 1 with +0 ability for +2 total.

#2 has a 13.75% chance of out-right winning plus a 2.75% of a tie. This would be like a person who recently learned how to play chess vs. a grand-master.

Even if you use expertise and make #1 a total of +17, there is still a 2.50% #1 will lose.

There really should not be any chance and IMO the number should show that. Yeah, yeah, "But the DM only calls for a check when the outcome is in doubt blah blah blah and such..." Phooey! That is crappy and lazy game design.

Now, use 3d6 instead of d20. Even with +17 if #1 rolled 3, for a total of 20, and #2 rolled 18 with +2 also for a total of 20, would be a stalemate. The chances of the most unlikely outcome would be barely more than 0.002% or about 1 in 46000!!!

The other option in such cases is to do a skill-challenge type thing instead of a single roll. Perhaps like death saves with 3 success indicating victory and 3 failures is a loss. IMO 5E should have more things like this. However, for the above example, it would have to be 5 or even 7 successes/fails, not just 3.

Why exactly do you play DnD?
You seem to hate it.

For your chess example: maybe it is not one contested roll for the whole game, but it is one for the opening, one for the mid game and one for the end game.

Maybe you don't have to do contested rolls (note, how OneDnD seems to avoid contested rolls), but you need to do rolls against DC15 until you accumulate 3 successes or failures... And suddenly a +5 advantage will help.
Then you coud assume that the grand master has expertise, not just proficiency.
Maybe proficiency in chess game and expertise in investigation and deception and insight.

 

[DND_Reborn]

Why exactly do you play DnD?

D&D isn't the problem. 5E is IMO. And Bounded Accuracy in particular. It is a useless mechanic and didn't solve any problem in the game, simply shifted where the problem was.

I love D&D and have for over 40 years. It is why I have stuck with it as long as I have. I very rarely ever play any other RPG. Unfortunately, the groups I play and run all started with 5E and are much younger players who are used to the "video game effect" of instant reward/satisfaction.

For your chess example: maybe it is not one contested roll for the whole game, but it is one for the opening, one for the mid game and one for the end game.

Yeah, I know, I suggested a series of contested rolls.

Maybe you don't have to do contested rolls (note, how OneDnD seems to avoid contested rolls), but you need to do rolls against DC15 until you accumulate 3 successes or failures... And suddenly a +5 advantage will help.

OneDnD. Ugh. Only making matters worse....

Then you coud assume that the grand master has expertise, not just proficiency.

Yeah, I did that, too.

Even if you use expertise and make #1 a total of +17, there is still a 2.50% #1 will lose.

Maybe proficiency in chess game and expertise in investigation and deception and insight.

Why? Implying advantage would help, but the other player could have similar skills (albeit without as much benefit with having a much lower bonus).

The die should not outstrip experience, but that is precisely what bounded accuracy has done and leads to some ludicrous results.

Two rogues are attempting to pick a lock with DC 15. R1 has +11, R2 has +2.

R1 has a 15% chance of failing. R2 has a 40% chance of success. The probability of both those events happening is 6% (more than 1 in 20).

I'm sorry, but there is no way a 17th level PC with DEX 20 should have a 6% chance of failing when a 1st level PC with DEX 10 can succeed.

Sure, there are optional rules for auto-success as well, but all these options exist in the game because the designers know the numbers don't work to take care of such things on their own.

 

[UngeheuerLich]

D&D isn't the problem. 5E is IMO. And Bounded Accuracy in particular. It is a useless mechanic and didn't solve any problem in the game, simply shifted where the problem was.

I love D&D and have for over 40 years. It is why I have stuck with it as long as I have. I very rarely ever play any other RPG. Unfortunately, the groups I play and run all started with 5E and are much younger players who are used to the "video game effect" of instant reward/satisfaction.

A bit sad, but even if I started with ADnD, I would never go back, as failing a save and die is just no fun.
Back then we somehow found ways to game the system and not constantly die. Playing with so many house rules, that the game worked.
Fun back then, but not so much anymore.

Yeah, I know, I suggested a series of contested rolls.

OneDnD. Ugh. Only making matters worse....

Yeah, I did that, too.

Why? Implying advantage would help, but the other player could have similar skills (albeit without as much benefit with having a much lower bonus).

The die should not outstrip experience, but that is precisely what bounded accuracy has done and leads to some ludicrous results.

Two rogues are attempting to pick a lock with DC 15. R1 has +11, R2 has +2.

R1 has a 15% chance of failing. R2 has a 40% chance of success. The probability of both those events happening is 6% (more than 1 in 20).

I'm sorry, but there is no way a 17th level PC with DEX 20 should have a 6% chance of failing when a 1st level PC with DEX 10 can succeed.

Sure, there are optional rules for auto-success as well, but all these options exist in the game because the designers know the numbers don't work to take care of such things on their own.

You know, better players in chess can lose vs worse players. Since I am not bad at playing chess, playing in a league and knowing hiw elo rating works, I can say, that even a grandmaster can lose against a player who is quite a bit worse. In blitz chess it is a lot more likely than in a long game. I don't want to bore you with details.
You can easily model it, if you think this is fun in your D&D game with a roll every minute of the game and you will see, that the statistical advantage of a +2 bonus will eventually inevitably (99.9% chance) lead to the better player winning (if they are better enough).

So, back to the actual problem of a level 17 character and a level 0 character succeeding at the same task. This is the feature, not a bug.
Why should someone with no wizard training be less good in juggling or balancing? What does having being a level 17 wizard to do with hiding?
This is just bollocks.

 

[DND_Reborn]

A bit sad, but even if I started with ADnD, I would never go back, as failing a save and die is just no fun.
Back then we somehow found ways to game the system and not constantly die. Playing with so many house rules, that the game worked.
Fun back then, but not so much anymore.

The threat of those dangers is part of what made the game exciting. When you faced a creature or spell with SoD effect or level draining, you knew it was a serious threat and had to act/plan accordingly.

Too much in 5E I see players just blithely running into the fray because there are no such threats anymore and most encounters are not dangerous enough to really pose much of a threat. Now, I know the DM can make things as hard as they want, etc. but I am talking about the design parameters of the game, not individual tables.

You know, better players in chess can lose vs worse players. Since I am not bad at playing chess, playing in a league and knowing hiw elo rating works, I can say, that even a grandmaster can lose against a player who is quite a bit worse. In blitz chess it is a lot more likely than in a long game. I don't want to bore you with details.
You can easily model it, if you think this is fun in your D&D game with a roll every minute of the game and you will see, that the statistical advantage of a +2 bonus will eventually inevitably (99.9% chance) lead to the better player winning (if they are better enough).

I am not just talking about "better players vs. worse players". I am talking about the BEST you can possibly be against someone who just knows the rules and how to play (basic proficiency without any ability modifier). There is a big difference.

I know I can always do more rolls, etc. but that isn't the way 5E is designed. I would prefer the bonuses and die sizes be sufficient to resolve realistically in a single roll. Others have even commented in other threads about trying to simplify combat to this level because for their game combat is not the "funner" part of the game.

So, back to the actual problem of a level 17 character and a level 0 character succeeding at the same task. This is the feature, not a bug.
Why should someone with no wizard training be less good in juggling or balancing? What does having being a level 17 wizard to do with hiding?
This is just bollocks.

Ok, back to it!

It isn't a feature, it is a bug. IF both PCs have proficiency, the assumption is the 17th level has been using/practicing that skill while adventuring, increasing their bonus from +2 to +6. It has nothing to do with class, it is about proficiency and ability scores.

Anyway, FWIW, I have no problem with both succeeding at the same task, I take issue with it when the higher bonus fails but the lower succeeds, particularly when the bonuses are so extremely different. With 5E this is because the d20 is so much larger than the bonuses. I've actually worked on a d12 or 2d6 system (which I mentioned upthread IIRC) to keep the bonuses within range of the dice. I certainly am not a big fan of bonuses really outstripping the die size, but at least roughly equal would be better IMO.

A level 1 PC with DEX 20 has a +7, which is better than a level 17 PC with DEX 10 at +6. Even without proficiency the DEX 20's +5 is nearly as good as the level 17's +6. I am not fond of this because IMO experience should definitely trump ability score (which represents potential training, natural talent, etc. in 5E). Instead of capping at +5 and +6 for ability / proficiency, I would prefer +4 and +8 respectively; this way experience is potentially up to twice as valuable as ability alone.

So, using your examples, wizard training has nothing to do with juggling or balancing (in fact, neither sleight of hand? nor acrobatics are wizard skills, so a wizard PC would get those from race or background or a feat maybe).

A level 17 wizard (don't know why you are singling out wizards... ) has nothing to do with hiding either (as stealth is also not a wizard skill), and being level 17 doesn't mean a thing for it unless the PC somehow has proficiency.

If either of those were true, as you suggest, I would agree it would be bollocks.

 

[UngeheuerLich]

Snip.

Regarding chess:
1. Basic proficiency is more than just knowing how to move the pieces.

2. As a chess player, i can say that it is not a single roll to determine the outcome. Every move you make has a potential to win or lose you the game. An average game takes 30 to 40 moves, so you need to roll 30 to 40 times? To make the best move in a position, it sometimes takes a DC 10, sometimes a DC 25 check.
Someone with basic proficiency will never find the best move in a certain situation because of their skill, but on a 20 might accidently still do it.
It is unlikely that they will make the best moves at each move.
Even a grand master won't always find the best move, but won't accidently make too obvious blunders.

I am not sure, if DnD is the best game to model a chess game, but if I play vs magnus Carlson, i will only make good moves, but after 30 moves, he somehow has improved his position so much, that I will lose.

Speaking in game terms, we for a regular game we will take 30 turns in 2 hours of time per person. If we use the play chess action every 6 seconds, we will have made 10 rolls per minute each, so 2400 rolls. Even 30 rolls would be sufficient to make a +2 bonus difference matter.

But as I said, I am not sure, that DnD wants to model playing chess.
But a single roll in no way reflects how the game is played. So you might to think about your preferences.

Regarding the rest:
You call it a bug. I call it a feature.
You think save or die is fun. I don't think so. You think it is tension, that every roll can kill you instantly, and preparing accordingly is the only way to do it right.
I say: scry buf teleport is the most stupid kind of gameplay ever. And if I as a DM use it against players, they would cry foul.

Oh. And regarding the wizard who does not have stalth on their class list. We have backgrounds for 8 years now, which grant half of the wizard's skill. So with their urchin or criminal background, of course they have stealth.

 

[Lanefan]

Everyone still can, you just have a more realistic appreciation for skill and ability with a non-linear system.

Here's a reason why I don't like it:

A contested check between two creatures: #1 is level 17 with +5 ability, for +11 total. #2 is level 1 with +0 ability for +2 total.

#2 has a 13.75% chance of out-right winning plus a 2.75% of a tie. This would be like a person who recently learned how to play chess vs. a grand-master.

Even if you use expertise and make #1 a total of +17, there is still a 2.50% #1 will lose.

There really should not be any chance and IMO the number should show that.

Reading this, I would posit you're giving too much weight to skills and asking/expecting them to produce a steeper power curve than is really desirable. One of 5e's best traits is that it flattened the overall power curve from what 4e and - even more so - 3e had, back to more like what 0-1-2e ran with. You seem to want to fight this trend.

It's not a chess newbie vs a grandmaster, it's a chess-club regular vs the top player in the club. Taking a skill in something makes you better at it, sure, but doesn't and shouldn't make you perfect.

That, and one of the pure joys of D&D for me is the chance of the upset result. Take that away and it gets boring pretty fast.

Yeah, yeah, "But the DM only calls for a check when the outcome is in doubt blah blah blah and such..." Phooey! That is crappy and lazy game design.

Now, use 3d6 instead of d20. Even with +17 if #1 rolled 3, for a total of 20, and #2 rolled 18 with +2 also for a total of 20, would be a stalemate. The chances of the most unlikely outcome would be barely more than 0.002% or about 1 in 46000!!!

Which is deep into "why bother?" territory.

 

[Lanefan]

You think save or die is fun. [...] You think it is tension, that every roll can kill you instantly, and preparing accordingly is the only way to do it right.
I say: scry buf teleport is the most stupid kind of gameplay ever. And if I as a DM use it against players, they would cry foul.

You're both right.

 

[W'rkncacnter]

It's not a chess newbie vs a grandmaster, it's a chess-club regular vs the top player in the club. Taking a skill in something makes you better at it, sure, but doesn't and shouldn't make you perfect.

taking into consideration that almost nobody IRL is considered level 5 or higher, and that an ability score of 20 is typically considered outright superhuman, i'd say a +11 is beyond a grandmaster, even when factoring for expertise.

 

[UngeheuerLich]

Reading this, I would posit you're giving too much weight to skills and asking/expecting them to produce a steeper power curve than is really desirable. One of 5e's best traits is that it flattened the overall power curve from what 4e and - even more so - 3e had, back to more like what 0-1-2e ran with. You seem to want to fight this trend.

It's not a chess newbie vs a grandmaster, it's a chess-club regular vs the top player in the club. Taking a skill in something makes you better at it, sure, but doesn't and shouldn't make you perfect.

That, and one of the pure joys of D&D for me is the chance of the upset result. Take that away and it gets boring pretty fast.

Which is deep into "why bother?" territory.

Also. A 1 in 20 chance for a draw seems a bad way to model chess. So a straight contested roll to decide who wins seems dumb.

 

[Lanefan]

Also. A 1 in 20 chance for a draw seems a bad way to model chess. So a straight contested roll to decide who wins seems dumb.

It's a long time since I played much chess but I seem to remember the number of draws was somewhat higher than 1-in-20.

But, add in the American dislike of ties - there always has to be a winner and loser - and maybe 1/20 is a good compromise?

 

[DND_Reborn]

1. Basic proficiency is more than just knowing how to move the pieces.

Correct, I said...

someone who just knows the rules and how to play (basic proficiency without any ability modifier)

Proficiency means you have played, learned, done whatever and are competent. You can "play chess" by learning the moves and just using INT modifier (no "proficiency" for games: chess). Proficiency is more than that, but not really a lot more since the base bonus is +2.

I am not sure, if DnD is the best game to model a chess game, but if I play vs magnus Carlson, i will only make good moves, but after 30 moves, he somehow has improved his position so much, that I will lose.

Speaking in game terms, we for a regular game we will take 30 turns in 2 hours of time per person. If we use the play chess action every 6 seconds, we will have made 10 rolls per minute each, so 2400 rolls. Even 30 rolls would be sufficient to make a +2 bonus difference matter.

But as I said, I am not sure, that DnD wants to model playing chess.
But a single roll in no way reflects how the game is played. So you might to think about your preferences.

You said it: DnD is not the best game to model chess... or really much else. It isn't a simulation game (especially they way 5E goes...).

I've already said (repeatedly, though you seem to ignore it...) that you could do a series of rolls, etc. OR you could design the game so a single roll (such as a contested check) does the job. Combat takes long enough with all the rolls, I would hate to make dozens of rolls for contested check because the system requires it for accuracy. Especially since there is really nothing more to it than rolling--sort of boring IMO.

Regarding the rest:
You call it a bug. I call it a feature.
You think save or die is fun. I don't think so. You think it is tension, that every roll can kill you instantly, and preparing accordingly is the only way to do it right.

Stop putting words into my mouth, ok? (or, more precisely, thoughts into my head? ).

The bolded part is nothing I ever wrote or implied, especially regarding SoD effects. I said I like, it is my preference.

Oh. And regarding the wizard who does not have stalth on their class list. We have backgrounds for 8 years now, which grant half of the wizard's skill. So with their urchin or criminal background, of course they have stealth.

Are you even reading my posts? I SAID ALL THAT ALREADY:

So, using your examples, wizard training has nothing to do with juggling or balancing (in fact, neither sleight of hand? nor acrobatics are wizard skills, so a wizard PC would get those from race or background or a feat maybe).

Anyway, as you said: I call it a bug, you call it a feature. We will not agree, so what is the point of continuing this...? Agree to disagree?

 

[OB1]

I think of randomness as one of the 3 pillars of D&D, along with DM generated Fate, and Player generated Choice.

• DMs (Fate) determine the world and everything in it, whether running a home brew or a pre-gen.
• Players (Choice) determine what their PCs do in response to the Fate before them.
• Dice (Chace) resolve conflict from uncertainty between Fate and Choice. The story is no longer in the hands of the DM or the Players when Dice come into play.

DMs can also choose to lean into chance by using random encounter tables, random treasure tables and rolling random damage for monsters. They also get to decide how, when and how often Chance enters the game. Because these are Fate related, the DM should decide when to lean into this, as fits the style of play they want to present and their players want to play.

PCs can lean into chance by rolling stats, rolling hp, etc. Because these are Choice related, players should decide when they want to apply these pieces.

There is a huge range in 5e from how much Random you and your group want in the game, and that, I believe, is a great thing.