D&D 5E Stupid math stuff that vaguely pertains to 5e.

[B.T.]

This may need to be moved, but here's my question: as a baseline measure that ignores all situational modifiers and presupposes an "average" adventurer, what should the chance of success with an action?

That is, how often should the average fighter hit with an attack? What about the average rogue? How often should a saving throw succeed? How often should characters succeed on skill checks that they are good at? What about skill checks that they're not good at?

As a side note that vaguely pertains to this vaguely-pertaining question: should characters be differentiated more by their chance of success or what they can do upon succeeding? For instance: should a fighter and rogue have similar hit chances but the fighter can do some sort kind of special maneuver that adds rider effects? Or should the game be simplified so that the fighter hits more often?

 

[I'm A Banana]

It's a simple question with a pretty intricate answer, so it'll be interesting to see the responses!

The most basic of chances that allows for character variation would give you a three-point system (25%, 50%, and 75%), but maybe you want to cluster them closer to 50% (40%, 50%, and 60%) to keep the dice roll significant. You might also make a five-point system to get more detail (40%, 45%, 50%, 55%, 60%), or more extreme (30%, 40%, 50%, 60%, 70%).

I'm kind of fond of that last one, actually....

And then, do you weight the game in favor of the party or not? Do you let a coin flip determine their fate? Or do you give them, say, a 66% chance of success out the gate because "they're heroes" and your game wants to encourage the party to win more often? Or do you maybe give them more like a 33% chance to start out with, they'll probably fail with die rolls, and then let them tweak the odds with things like Advantage?

The odds of success have varied pretty significantly over the course of the editions, from 4e's "assume the PC's are victorious, basically," to 1e tournament play grindfests like the Tomb of Horrors. Controlling that feel -- the feel of deadliness, or of challenging play -- will be a pretty important tool for 5e to have.

 

[Giltonio_Santos]

Taking into account an enemy of equal power level, I believe the following:

- The fighter should have 80-75% chance of hitting on a swing of the sword.
- The thief should have about 60-55% and increase that to same as fighter when using any backstab-like class feature.
- Enemies should fail 80-75% of the saving throws against a wizard's strongest spells.
- Cleric should have about the same as thief, improved by spells instead of combat circumstances. Enemies should fail 60-55% of the saving throws against their strongest spells.

I don't like the idea of "balanced" skill checks. I hate when 20th-level characters keep facing rivers that no peasant could swim and walls that no 1st-level thief would be able to climb, for the sake of keeping increasing skill modifiers relevant.

Cheers,

 

[Savage Wombat]

Those sound too high to me. Do we want a 1st level sword fight vs. an orc to come down to who wins initiative?

 

[Kynn]

Those sound too high to me. Do we want a 1st level sword fight vs. an orc to come down to who wins initiative?

It's hard to get the right balance between "the guy who goes first, hits first" and "combat is a whiff-fest".

In a typical AD&D fight at first level, you've got a THAC0 of like 20 and a +1 or +2 bonus, and you're fighting something with an AC of 5 or so. That means you hit about 1 attack in 3 (and the orc likewise has the same chance to hit or miss). That seems a little too whiffy for me.

In 4e D&D, you've probably got a +8 or +9 bonus to hit (+4 ability score, +3 proficiency bonus, +1 feat bonus) vs. an AC of 15 (14 + level, assuming a skirmisher opponent) which means you end up hitting 2/3 of the time.

My ideal would be somewhere between these two extremes, meaning that you'd hit like 50% of the time.

BUT!

I am okay with starting out at a low chance to hit against level-appropriate AC and then having BAB scale faster than AC does. I am okay with the formula looking something like this (assuming level-appropriate AC):

level 1 ... hit 40% of the time
level 4 ... hit 45% of the time
level 7 ... hit 50% of the time
level 10 ... hit 55% of the time
level 13 ... hit 60% of the time
level 16 ... hit 65% of the time
level 19 ... hit 70% of the time
etc

I want higher level characters to hit more often than low level against appropriate foes, and I want this math to be baked into the system and balanced appropriately.

4e assumed a constant hit chance (such that your chance to hit an appropriate foe at any level was roughly the same) and I think that was a mistake. Earlier editions like AD&D didn't assume this, but didn't have very tight math. I want tight math WITH this assumption, and monsters built accordingly.

 

[BobTheNob]

Those sound too high to me. Do we want a 1st level sword fight vs. an orc to come down to who wins initiative?

Its a ibt of a bummer that. The more you up hit probability and damage, the more you make initiative the dominant factor. The more you decrease hit probability and damage the more combat drags on.

Its such a fine balancing Act. I have always considered 50-50 all things being equal

 

[Ratskinner]

That is, how often should the average fighter hit with an attack? What about the average rogue? How often should a saving throw succeed? How often should characters succeed on skill checks that they are good at? What about skill checks that they're not good at?

As a DM, when I have to eyeball it, I generally follow a 90-70-50-30-10 rule (for d20s: 2-6-10-14-18). Corresponding to Very Easy - Easy - Typical - Hard - Very Hard. Failure has to have some very entertaining consequences for me to ask for Very Easy and sometimes Easy rolls.

As a side note that vaguely pertains to this vaguely-pertaining question: should characters be differentiated more by their chance of success or what they can do upon succeeding? For instance: should a fighter and rogue have similar hit chances but the fighter can do some sort kind of special maneuver that adds rider effects? Or should the game be simplified so that the fighter hits more often?

I prefer the latter. Highly variable "chance of success" would (I'm guessing) mean players grubbing for every "+1" they could find, and the effective power curve being very steep. Adding more options and broadening the results of the success means less of that, and I'm in favor of less of that.

 

[Mokona]

The average roll needed to hit should be higher at first level and slowly decline as you level. So you get "better" at hitting average opponents as you level. Exact numbers don't matter at this stage but say you need a 13 to hit at first level and you need a 8 to hit at 20th level. At first you really want combat advantage because it gets you closer to 50/50 and over time you can afford to take penalties for environmental effects (or extreme action) and still hit 50/50.

The gap between skilled users and non-skilled needs to get narrower. In 3e that gap started at a +8 spread and got eternally wider. In 4e the spread also started out at around the same level (+9) but only increased slightly after that. Both of those ignore Skill Focus feats which widen the gap in favor of specialists. I would hold it down to a skilled-unskilled gap of only +2 and then let specialists really shine by handing out +5 for Skill Focus.

Hitting is fun. All player's should hit roughly the same amount of times if there is any binary effects (riders like status effects, penalties, marks, zone control, &c). If hits only convey damage then it's ok to design a system where those who do more damage hit less often and vice verse (but I think such a simplified system with 5e is unlikely).

 

[KidSnide]

I think if you do research about what players find fun in actual play, you get results that are very different from 50/50 chance of success.

To take a trivial (but often misunderstood) example, PCs need to have a dramatically better than 50% chance of succeeding at an average encounter. Otherwise, you get unacceptably short campaigns as 99.9% of them will end in the first 10 encounters.

If PCs hit only 50% of the time, players will feel like they were completely ineffective in combats shockingly and unacceptably often. Even low-accuracy characters need to hit more than 50% of the time, or the experience will be unbearable. The consequence is that high accuracy characters will hit almost all the time. That's OK. That's the benefit of high accuracy.

Post-math-fix 4e got this mostly right, at least in the heroic tier. I have high hopes that 5e will adopt at least some of that wisdom.

-KS

 

[Gansk]

Trailblazer has come to the conclusion that the PC's should succeed around 70% of the time for the game experience to seem neither too easy nor too hard.

3e is measured against this goal for the following when PC level = monster CR

PC attack vs. monster AC
PC AC vs. monster attack
PC save vs. monster DC
PC DC vs. monster save

Buy the pdf if you want to know how the math turns out.

 

[kitsune9]

It's a simple question with a pretty intricate answer, so it'll be interesting to see the responses!

The most basic of chances that allows for character variation would give you a three-point system (25%, 50%, and 75%)

I like the first one with a standard deviation of 5% to each set and max at 3 deviations. However, I acknowledge that to build a system with this many permutations may too daunting to do so locking the numbers into a 3 point to 5 point system is probably a more worthwhile endeavor.

 

[kitsune9]

Trailblazer has come to the conclusion that the PC's should succeed around 70% of the time for the game experience to seem neither too easy nor too hard.

It was an excellent read. It's really good for players who want to create a more balance system (or create imbalance!) and use their mathematical modeling.

 

[Yora]

As a very lose base, I'd say 50% for the average person to overcome the average obstacle and 75% for a specialist to overcome an average obstacle.

The big problem with D&D is that character level can make a very significant difference so 50% at 1st level very soon becomes 95% at 10th level for the same task.
Thankfully this is something that the designers want to adress.

 

[nightwalker450]

For combat, I'd rather they raise accuracy, and lower damage (or raise HP). If combat has a chance to take a while, it'd be better if the players usually hit on their turn, instead of waiting each round, just to miss.

 

[howandwhy99]

as a baseline measure that ignores all situational modifiers and presupposes an "average" adventurer, what should the chance of success with an action?

how often should the average fighter hit with an attack?

What about the average rogue?

How often should a saving throw succeed?

How often should characters succeed on skill checks that they are good at?

What about skill checks that they're not good at?

should characters be differentiated more by their chance of success or what they can do upon succeeding?

should a fighter and rogue have similar hit chances but the fighter can do some sort kind of special maneuver that adds rider effects?

Or should the game be simplified so that the fighter hits more often?

Combat was 50% base chance, 2 of 6 for most other rolls.
"Average" Fighter "to hit" odds differ by level at the baseline. +1 (5%) from above 0-level character.
"Average" Thief "to hit" odds differed too.
Saving Throws were about 25% early on, 75% later, but these had wide variance and lots of modifiers late in the game (higher levels).
Skills were kind of a screw up. But 2 of 6 on a d6 starting. "Good at" could just be called more.
Not good at skill checks could just be the default 1/3rd odds.
What is possible after succeeding is singular originally. Differentiation was by ability odds. Lots of stuff was largely auto-success or fail though.
Rider effects would have been optional auto-successes by ability. For instance, monk stunning was optional, but required higher odds to perform.
Fighters hitting more often kind of makes sense. They are more capable at fighting. All the other elements are extra stuff.

 

[Sunseeker]

Before feats that improve hitting are taken, this is my opinion:
Fighter: 75%
Hitting is what the fighter DOES. If the fighter isn't hitting, then they are doing nothing, they have little else they can do, so they need to be able to do their main thing very, very reliably.
Paladin: 65%
Hitting is what a paladin is good at, but they're capable of doing other things when they can't hit. A paladin who can't hit can give buffs and sometimes heals to their allies.
Rogue: 60%
Like Paladin's, rogues have a lot of things they can do in a fight besides hitting, traps, devices, they've got that laundry-list of skills for a reason.
Cleric: 50%(variable), given there are so many ways to build a cleric, I think 50% is the only real way to represent them. Melee-hit shouldn't be their primary feature unless they're a specific build focusing on it.
Sorcerer: 40% Like most casters, sorcerers don't hit things, but given that their training is in the School of Hard Knocks as opposed to The Arcane University, they should be hitting more than your average Wizard. Spell "hit" depends really on the reflex saves of their opponents.
Wizard: 30%, Wizards pretty much don't hit, they can, but even if they do they're not really going to accomplish much. Like Sorcerers and other casters, spells are defended against, not attacked with.

The chart is basically reversed when looking at reflex saves. Wizards at the top, Paladin's at the bottom.

And that's how I break down the base classes, beyond that, specific builds will go a long way to improve the actual hit, feats and so on.

 

[JRRNeiklot]

An average first level fighter should hit an average unarmored person about 55-60% of the time.

 

[Crazy Jerome]

In D&D, given the (linear) nature of the d20 and the generally expected ranges, I don't think my preference is possible--at least not without a lot of jumping through hoops that would probably be unplayable.

I like 65% as the default base for average character tries average activity. Then increasing specialization should get them as much as +15% on that--with each +5% being harder to get than the previous one. Likewise, being somewhat capable but lagging behind could drop as much as -15%. Then situational modifiers could bump another +15% up (being very stingy with the +10% and +5% options, and little or no stacking), and as much as -25% down. Not coincidently, a base 25% would be the starting floor for any ability that was possible but just barely for that character.

Or possibly more useful, I wouldn't mind a system where 1 was auto fail, 20 was auto succeed, 2 was "auto fail unless there is a darn good reason why not," and 16-19 was increasingly auto succeed unless there is a darn good reason why not." Once you get to either boundary, you are fighting for "reason why not" rather than another +1.

 

[Kannik]

After playing games such as those using the Silhouette system my inclinations shifted to this rather verbose feel:

A Competently Trained individual should succeed on a Moderateskill task/challenge about 75% of the time.

By Competently Trained I mean trained to a level to which you would expect someone would have if that was their profession and they had been doing it for a year.

By Moderate I mean something that would be slightly out of the norm for that profession.

So, for example, a relatively new auto mechanic who has to rebuild a transmission should be able to do it with a 75% chance of success (and failure here of course means it takes them longer than anticipated). NOTE: I am not an auto mechanic and maybe rebuilding transmissions is really easy.

Nicely, in combat against a similarly skilled opponent, to keep the players from being frustrated I have found that a 70%ish hit ratio is a good balance, which lines up with the competent ratios above. Note that this is in a system where one hit != one kill -- if you're playing in that game a 70% hit ratio would end up with many dead PCs! Better alter that to something else...

peace,

Kannik

 

[TerraDave]

One principle I would like to see, regardless of the specific numbers: as you level, the % chance of doing something goes up.

There are some caveats:

Attacking: this might hit a plateau at some point, or maybe you sacrifice a certain chance to hit for something else.

Skills: Should not plateau, but maybe you tradeoff or also get additional abilities (faster, more, better...)

Saves: (or defenses against deadly and debilitating things): This is potentially tricky, but I know how I want it to work: higher level characters and monsters should save most of the time.

Not good: untrained, non-core stuff might not get better.