D&D 5E How difficult should Difficulty be?

[DND_Reborn]

I think this, and the bolded portion in particular, is our biggest point of disagreement. I don’t see how one could possibly describe a task than any rando with no training whatsoever can just luck into 5% of the time as “nearly impossible.” 5% is a platinum trophy in Dark Souls. That’s not “nearly impossible,” that’s just hard. Winning Olympic gold, that’s nearly impossible. Only the best athletes in the world even have a chance at that, and only a very few of them succeed at it.

1 in 20 isn't nearly impossible. Winning the lottery is. 1 in 20 is just very hard.

Edit: Thinking further, I roll a natural 20 pretty much every week, usually multiples. If a 5% chance is nearly impossible, I achieve the nearly impossible on a weekly basis, usually multiple times.

This is the issue with using a linear d20 and then labelling a task "nearly impossible"

If anyone can attempt the task which is nearly impossible, and the mechanic used is a d20, you must have a chance. The lowest chance this mechanic offers in 1 in 20 odds.

Change the mechanic to 4d6-4 and now nearly impossible is less than 1 in 1000.

Now, you can make it higher than DC 20, and figure out the odds on what % of creatures would have the appropriate modifier could make that higher DC. I mean, what % of creatures get a +10 or better for instance? 1 in a million? More?

Nearly Impossible is a wide range of probabilities and each person has to figure out what that cut-off is for themselves.

So, let's say +10 is 1 in a million. Even with a 35% chance, that becomes 1 in 2,850,000 roughly. To me, those are "nearly impossible" odds.

 

[Horwath]

Change the mechanic to 4d6-4 and now nearly impossible is less than 1 in 1000.

Any special reason for preferring 4d6-4 over 3d6?

3d6 gives same average as d20 and one less die to sum up every time.

Or you really want that higher bell curve?

 

[DND_Reborn]

Any special reason for preferring 4d6-4 over 3d6?

3d6 gives same average as d20 and one less die to sum up every time.

Or you really want that higher bell curve?

4d6-4 (IOW treat 6's as 0's) gives you 0-20, average 10, and a more normal (as you say "higher") curve.

I discovered it years ago and have always loved the concept, but I know to most people the simplicity of the d20 outweighs the other benefits.

 

[Horwath]

4d6-4 (IOW treat 6's as 0's) gives you 0-20, average 10, and a more normal (as you say "higher") curve.

I discovered it years ago and have always loved the concept, but I know to most people the simplicity of the d20 outweighs the other benefits.

max or min roll is 1/1296 vs 1/216 for 3d6 vs 1/20 for d20.

maybe I like the middle ground of 3d6 more

also spread of results is only 16 pts, from 3 to 18 vs 21 pts(0 to 20)

 

[Stalker0]

One way to deal with the 5% probability is to impose disadvantage in some cases. For example you might not make the DC as high, but you decide due to the complexity or circumstances of the task, that the PC has disadvantage.

Now a 1 in 20 becomes 1 in 400 (.25%), which I think is a bit closer to the "level of impossible" some people would expect.

 

[Maxperson]

This is the issue with using a linear d20 and then labelling a task "nearly impossible"

If anyone can attempt the task which is nearly impossible, and the mechanic used is a d20, you must have a chance. The lowest chance this mechanic offers in 1 in 20 odds.

Change the mechanic to 4d6-4 and now nearly impossible is less than 1 in 1000.

Now, you can make it higher than DC 20, and figure out the odds on what % of creatures would have the appropriate modifier could make that higher DC. I mean, what % of creatures get a +10 or better for instance? 1 in a million? More?

Nearly Impossible is a wide range of probabilities and each person has to figure out what that cut-off is for themselves.

So, let's say +10 is 1 in a million. Even with a 35% chance, that becomes 1 in 2,850,000 roughly. To me, those are "nearly impossible" odds.

Why is the bold true? Clearly the game doesn't intend just anyone to be able to attempt and succeed at a nearly impossible task, so they model that by making the DC high enough to be out of range of anyone but mid to high level adventurers or the very rare low level adventurer with just the right set-up AND magic.

Lowering the DC to 25 makes it very hard for many(1 in 200) completely untrained commoners and for most 1st level adventurers. Literally nothing is nearly impossible any longer.

Nearly impossible is not an individual thing that differs for each person. It's a blanket number that makes a task nearly impossible and an individual with exceptional ability, say a 20th level PC with expertise and 20 in a stat, would have an amazingly high 10% chance to succeed.

 

[Charlaquin]

This is the issue with using a linear d20 and then labelling a task "nearly impossible"

If anyone can attempt the task which is nearly impossible, and the mechanic used is a d20, you must have a chance. The lowest chance this mechanic offers in 1 in 20 odds.

Why must you have a chance? Again, I don’t have a chance at winning Olympic gold, at least not without many years of training. Why must any random character have a chance to succeed at any task?

Heck, even if you did lower the DC to 20, anyone with less than a 10 in the relevant ability score still don’t have a chance.

Also, wait, who says anyone can attempt any task? The DM is perfectly free to rule any action a failure without calling for a roll.

 

[James Gasik]

Because it's not fun to say "no"? And we're not talking about "random characters" here either. Take Maxperson's example of a farmer with a 20 ability score. That is an Olympic-level character, the absolute peak of physical ability. There's not a lot of those kinds of people running around!

And history is full of people who have pushed the boundaries of what's believed to be possible. There was a time when experts said that a human couldn't possibly run a 4 minute mile. In the 1940’s, the mile record was pushed to 4:01, where it stood for nine years, as runners struggled with the idea that, just maybe, the experts had it right. Perhaps the human body had reached its limit.

Then, on May 6, 1954, Roger Bannister broke the 4-minute barrier, running the distance in 3:59.4. As part of his training, he relentlessly visualized the achievement in order to create a sense of certainty in his mind and body.

Barely a year after Bannister’s accomplishment, someone else ran a mile in under 4 minutes. Then some more runners did. Now, it’s almost routine. Even strong high-schoolers today run 4-minute miles.

So why not let player characters kick reason to the curb and do the impossible? Why do we need to throw high DC checks at them without good cause? Like Charlaquin said, the DM can always say "no" if it suspends disbelief too far.

 

[DND_Reborn]

Nearly impossible is not an individual thing that differs for each person. It's a blanket number that makes a task nearly impossible and an individual with exceptional ability, say a 20th level PC with expertise and 20 in a stat, would have an amazingly high 10% chance to succeed.

Correct, and that is exactly what I am saying. You, however, seem to be arguing that very thing.

Again, I don’t have a chance at winning Olympic gold, at least not without many years of training. Why must any random character have a chance to succeed at any task?

Technically, you do. It is extremely unlikely, but there is a chance...

Ok, last attempt:

I think we all agree that a task's difficulty is based on the entire set of creatures that can attempt it. A nearly impossible task will only be possible by a remote few.

If you don't agree with that, stop reading and the discussion is over as far as I am concerned.

There are 18 skills, and by default for backgrounds only since not all creatures are PCs, I think it is a reasonable assumption that most creatures will have 1, maybe 2 skills in which they are proficient. There will be overlap for those with two skills which duplicate those with just one. So, let's say there is a 1 in 12 chance a creature will have a given skill.

That grants that creature +2 proficiency. Now, what about creatures who have begun to excel at that skill? Using the same odds, 1 in 12, only 1 in 144 would be +3. Following the same progress, about 1:250,000 would be +6 in a skill--maximum proficiency.

Now, assuming an ability mod of +5 puts a creature in the 3 sd range (likely much higher actually), the odds are 1:740 roughly.

That makes the odds of someone having +6 in a skill and +5 in the appropriate ability about 1 in 185,000,00. There will be some synergy between high ability and high skill, so, let's drop that to a nice even 1 in 100,000,000.

Such a creature, given a DC 25 "nearly impossible" would still need to roll 14 or higher, so just 35% of success.

Now, that means a random creature attempting that "nearly impossible" task will have just 1 in 285,000,000 odds of successfully completing a DC 25 task. I don't know about YOU, but that sounds like something that is "nearly impossible".

Jumping back to the beginning, let's use a creature with base proficiency and +3 ability mod. 1:12 for proficiency, and let's drop the ability odds to 1:200 (2 ds roughly IIRC). Since such a creature needs a nat 20 to make the check, the odds of that even are 1:48,000.

t is much lower than the above odds, but even 1:48,000 works for me for "nearly impossible".

In summary, the "TASK" is nearly impossible, not just "nearly impossible" for the tier 4 PC with +10 or 11 to the check. It is set by the world and the likelihood creatures of that world could do it. If practically every creature in the world could do it, it would be "very easy".

So, I'm done with discussing this with both of you. You can keep the DCs the same, use magic, etc. to hit them, or whatever you want. I'm not here to try to persuade you to change things in YOUR game, just asking (as in the OP) if anyone else feels the DCs (especially at higher tasks) are too high, and also if they misrepresent the task names.

May you only roll 20s!

 

[Charlaquin]

You said you were done discussing this, so
I’m going to drop it, but I just wanted to say quickly:

If you don't agree with that, stop reading and the discussion is over as far as I am concerned.

I really, really don’t. It’s not just a matter of chance, I absolutely lack the physical ability to even compete in any sporting event at anywhere near an Olympic level, let alone win against athletes at that level. It would (and does) take years of highly dedicated training to even have a chance, which is exactly why it’s a nearly impossible feat.