D&D 5E Removing Attack Rolls -- and maybe more? (Game Design / Theory Discussion)

[Blue]

One of the awkward bits of D&D is that we have tied that die roll to a narrative - If we make the "To Hit" roll, we HIT the target with the weapon. Meanwhile we have Hit Points that are specifically decoupled from narrative - they aren't "just meat".

If we decouple the d20 roll from that narrative, then we decouple "hit or miss" from "succeed or fail". If you succeed on the d20 roll, you do damage to the target - whether we narrate that as running them through with a sword, or them twisting their ankle trying to dodge, becomes a separate element - we can either use that as design space, or just leave it to convenience..

I'm completely with you. As mention, I am a proponent of miss damage - for the reasons you mention.

But what was proposed had no d20 roll at all, no success or failure - only draining of HPs, which is an effect.

But it is possible at range to not interact with the target at all, and I can't see a way to have the mechanics support that narrative. Every shot, even if you are not aware of it, will drain HPs.

 

[Umbran]

But it is possible at range to not interact with the target at all, and I can't see a way to have the mechanics support that narrative. Every shot, even if you are not aware of it, will drain HPs.

Once we have decoupled hit points from meat, I'm okay with that. Being unaware of the arrow is not really different from being unaware of the rogue's short sword as they sneak attack when hidden.

We are talking about heroic fantasy - imho, the results don't need to be "realistic". They just need some basic plausibility as a narrative. So, that unseen archer keeps you dancing away from arrows that just barely miss and tht you notice just before they hit - you get tired dancing away from them, using hit points.

 

[DND_Reborn]

Delightfully terrifying!

Yes, it is!

Kobolds, Giant Rats, and any other of the 1d4 damage low CR monsters suddenly become real KO machines for low level characters.

LOL, it isn't that great. Exploding d4 averages less than a regular d6...

How about for something that has multiple damage dice (e.g. Ogre 2d8+4)? I'm guessing any and all dice that are max damage "explode", yes (i.e. you don't need two "8s" to explode)?

Correct. Any die that rolls maximum explodes, so if you are rolling multiple dice any to all of them could explode. So, fireballs average 32 instead of 28, 3 magic missiles average 13 instead of 10.5, etc.

 

[Blue]

Once we have decoupled hit points from meat, I'm okay with that. Being unaware of the arrow is not really different from being unaware of the rogue's short sword as they sneak attack when hidden.

We are talking about heroic fantasy - imho, the results don't need to be "realistic". They just need some basic plausibility as a narrative. So, that unseen archer keeps you dancing away from arrows that just barely miss and tht you notice just before they hit - you get tired dancing away from them, using hit points.

Remember, this is every shot by every single creature, not just heroes. There are times when kobolds with slings at extreme range just aren't a threat to the PCs, and that's a narrative this can't support. No "cutsey" explanation like "the archer makes your dance" can cover every shot by every creature, it needs to be solid.

 

[Umbran]

Remember, this is every shot by every single creature, not just heroes.

I have not forgotten that.

There are times when kobolds with slings at extreme range just aren't a threat to the PCs, and that's a narrative this can't support.

What? By the book there is no extreme range. There's short range (30 feet for a sling), long range (30-120 feet for a sling) at which you are at disadvantage, and they are not effective beyond 120 feet.

So, since we are not using a d20 to hit here, we will need to figure out what "disadvantage" translates into, and we are good. We need to do that, and figure out what advantage means in this scheme, anyway.

 

[Lojaan]

I'm not sure what problem is being solved here?

In my experience, players like making attack rolls, and enjoy it when bad guys miss them.

Why remove something people enjoy? Or do your players not enjoy this?

 

[DND_Reborn]

I'm not sure what problem is being solved here?

This isn't a problem I am trying to solve, as I thought this should have made clear:

So, I guess that's it for now. This isn't really something I am considering adopting seriously, just the results from some observations about the purpose of bounded accuracy, how likely attacks succeed in 5E, and other factors.

Also, I am not saying this is a good idea or a bad idea, just an idea.

Bounded Accuracy replaced escalating attack modifiers and AC (the "arms race / treadmill effect") with escalating damage and hit points.

In my experience, players like making attack rolls, and enjoy it when bad guys miss them.

As so mine (sort of), but this discussion is about are they necessary?

Why remove something people enjoy? Or do your players not enjoy this?

Frankly, my players enjoy rolling damage more because now the excitement/ tension is will you do a lot of damage or a little? By moving the critical to the damage roll, this is still present to adding more excitement (rolling maximum) and because damage dice are smaller than the d20 happens more often. They also cringe more often when I roll maximum against them.

5E has roughly a 60-70% hit rate, often slightly higher due to buffs IME. It became so common that it started to lose the excitement since PCs hit way more than they missed. I had one former AD&D player who was playing 5E with me and surprised to learn he hit with a roll of 6 at low levels (he was in tier 1, but I don't recall his level at that time). I told him, "Yeah, you hit a lot in 5E, it was designed that way."

So, the question I am postulating really is why bother having a roll if you are most likely going to succeed? It becomes unnecessary and slows down the game, adding little enjoyment.

Also, it adds disappointment when you DO miss. Which is why some people like or want a system where you cause less damage (but still some) or some other effect can happen on a miss. This way you still feel like you are contributing.

The sort of changes (again, as a design experiment) would be more about an "attack effectiveness" roll. When you attack, you are nearly always effective in making your target respond (dodge, soak, injuring them, etc.). The "damage" roll, modifier by your offensive abilities and your target's defensive abilities, would then determine how "effective" your attack was. If the target has sufficient defenses to handle everything you dish out (i.e. damage), they can negate your effectiveness entirely.

Finally, I seriously doubt I would implement this for D&D, but after reading more about Bounded Accuracy I realized it became (IMO) the next logical step. There are systems out there that do this sort of thing, but more by making it a contested roll. I am further removing the hassle of the defensive roll by making it passive--as it is in D&D.

Hopefully that explains everything in a clearer light.

 

[Tales and Chronicles]

Some OSR systems such as Knaves and Maze Rats (IIRC) do not use ''to-hit'' rolls and they work quite well. Combat is more dangerous because HP WILL be lost when you enter combat, no escaping it. The sheer load of characters hit points in 5e will be challenged.

 

[HammerMan]

Once we have decoupled hit points from meat, I'm okay with that. Being unaware of the arrow is not really different from being unaware of the rogue's short sword as they sneak attack when hidden.

We are talking about heroic fantasy - imho, the results don't need to be "realistic". They just need some basic plausibility as a narrative. So, that unseen archer keeps you dancing away from arrows that just barely miss and tht you notice just before they hit - you get tired dancing away from them, using hit points.

in more modern terms your HP is your plot armor.

batman has 500hp, so yeah, 1 hit from darksaid would kill him (he is only mortal) but the first 499hp are him dodging and avoiding the deadly blow

 

[Lojaan]

This isn't a problem I am trying to solve, as I thought this should have made clear:


Bounded Accuracy replaced escalating attack modifiers and AC (the "arms race / treadmill effect") with escalating damage and hit points.


As so mine (sort of), but this discussion is about are they necessary?


Frankly, my players enjoy rolling damage more because now the excitement/ tension is will you do a lot of damage or a little? By moving the critical to the damage roll, this is still present to adding more excitement (rolling maximum) and because damage dice are smaller than the d20 happens more often. They also cringe more often when I roll maximum against them.

5E has roughly a 60-70% hit rate, often slightly higher due to buffs IME. It became so common that it started to lose the excitement since PCs hit way more than they missed. I had one former AD&D player who was playing 5E with me and surprised to learn he hit with a roll of 6 at low levels (he was in tier 1, but I don't recall his level at that time). I told him, "Yeah, you hit a lot in 5E, it was designed that way."

So, the question I am postulating really is why bother having a roll if you are most likely going to succeed? It becomes unnecessary and slows down the game, adding little enjoyment.

Also, it adds disappointment when you DO miss. Which is why some people like or want a system where you cause less damage (but still some) or some other effect can happen on a miss. This way you still feel like you are contributing.

The sort of changes (again, as a design experiment) would be more about an "attack effectiveness" roll. When you attack, you are nearly always effective in making your target respond (dodge, soak, injuring them, etc.). The "damage" roll, modifier by your offensive abilities and your target's defensive abilities, would then determine how "effective" your attack was. If the target has sufficient defenses to handle everything you dish out (i.e. damage), they can negate your effectiveness entirely.

Finally, I seriously doubt I would implement this for D&D, but after reading more about Bounded Accuracy I realized it became (IMO) the next logical step. There are systems out there that do this sort of thing, but more by making it a contested roll. I am further removing the hassle of the defensive roll by making it passive--as it is in D&D.

Hopefully that explains everything in a clearer light.

Ahhhhhh I see what you mean now. Thank you for explaining it so well. You're saying if the 'to hit' roll always hits, do we even need it? And if good design is removing all rolls that either have no chance of success or no chance of failure then there is definitely an argument to either removing it, or somehow combining the 'to hit' and damage rolls.

I do feel like removing the attack roll generates a big chunk of cognitive dissonance tho. It's harder to get your head around. How about removing the damage roll? Make it so that each weapon/spell has a set damage value that is modified by the attack roll, like;

Miss by more than 5 = Miss (no damage)
Miss by less than 5 = glancing hit (half damage value +ability mod damage)
Hit = normal damage (damage value + ability mod)
Hit by more than 10 = critical damage ( 2 x damage value + ability mod)

So, for example, a longsword could have a damage value of 3/6/12 (glance/hit/crit)

You could also have some fun playing around with different weapons like, giving a scythe a 0 for glancing damage, but 3 x for crit etc..

This is kind of a fun idea but I know my players enjoy rolling dice* so I think even though this could be faster, it could also be less fun for them.

*Except for d4s. No one enjoys rolling d4s.

 

[DND_Reborn]

Ahhhhhh I see what you mean now. Thank you for explaining it so well. You're saying if the 'to hit' roll always hits, do we even need it? And if good design is removing all rolls that either have no chance of success or no chance of failure then there is definitely an argument to either removing it, or somehow combining the 'to hit' and damage rolls.

Sure, no problem. I am glad it is clearer now.

I do feel like removing the attack roll generates a big chunk of cognitive dissonance tho. It's harder to get your head around. How about removing the damage roll? Make it so that each weapon/spell has a set damage value that is modified by the attack roll, like;

Miss by more than 5 = Miss (no damage)
Miss by less than 5 = glancing hit (half damage value +ability mod damage)
Hit = normal damage (damage value + ability mod)
Hit by more than 10 = critical damage ( 2 x damage value + ability mod)

So, for example, a longsword could have a damage value of 3/6/12 (glance/hit/crit)

You could also have some fun playing around with different weapons like, giving a scythe a 0 for glancing damage, but 3 x for crit etc..

I reality you are suggesting the same thing: combining the attack roll with the damage roll.

With your suggestion, "misses" would be extremely rare IME, maybe 10% of the time on average, and damage would become more static since the values are set, as shown in your example of the longsword.

If you can wrap your head around the idea of the attack yields the damage, you are already there, just looking at it through the glass so it is mirrored in a manner of speaking.

I am saying if your damage is minimal (or negated entirely by the target's defense), your attack was not very effective; as where if you roll critical damage and the defenses can't over come it, your attack was very effective. Since rolling damage yields more results, I think it is a better way to go (no shock there, huh? ).

BUT, if you wanted to explore your version, I would make it this:

Attack modifier = half listed weapon damage die + double ability modifier + proficiency

Damage = d20 roll + Attack modifier - target's AC

Let's say you are 5th level. have STR 16, and using a longsword. Your attack modifier would be 4 (half d8) + 6 (double STR modifier) + 3 (proficiency) = +13.

You roll a total 18 (roll 5 + 13) vs. AC 15. 18 - 15 = 3 damage.
You roll a total 26 (roll 13 + 13) vs. AC 15. 26 - 15 = 11 damage.
You roll a total of 15 (roll 2 + 13) vs. AC 15, no damage.

This might sound confusing at first, but for us old dogs used to working with THAC0, it is no big deal.

This is kind of a fun idea but I know my players enjoy rolling dice* so I think even though this could be faster, it could also be less fun for them.

*Except for d4s. No one enjoys rolling d4s.

Yeah, that is an issue with your idea (and my suggested version). Another reason just rolling the damage is useful, since the d20 would still be used for other things, and all the weapons keep their different damage dice (even the dreaded d4... ).

 

[Ulorian - Agent of Chaos]

Perhaps:

Light armor provides resistance vs. B/P/S of 10 damage or less.
Medium armor provides resistance vs. B/P/S of 25 damage or less.
Heavy armor provides resistance vs. B/P/S.

(Picking numbers out of the air.)

DR as damage resistance, not a flat reduction... I like that a lot! How about instead of 'X damage or less' just say 'up to a maximum of X damage reduced' (and add that clause to heavy too).

 

[LordEntrails]

These ideas always ... amuse? me. At my table, the idea of reducing randomness is... damn near a non-starter.

Why?
Simple because half of my table can run the calculations in their head on round one of exactly how many rounds, on average, it will take if things are static. to defeat or lose to the other side. They know more than just DPR, but have an innate feel for how many rounds an encounter will take, how much of their resources will be used up, and what risk level they are at. 40 years of gaming with people with advanced math & engineering degrees will do that.

Its why we use re-roll initiative each round. Why the DM almost never shows all the adversaries on round 1, why the players almost never optimize their characters, why we all do wild and weird stuff. So the outcome is always in question, so the game remains fun.

But that's our table. Do what you need to keep or grow the fun at yours.

 

[DND_Reborn]

At my table, the idea of reducing randomness is... damn near a non-starter.

Sadly, that is the issue... it really isn't all that random when you are hitting nearly all the time anyway...

Otherwise, as a DM I do pretty much everything you mention: re-roll initiative, cascading encounters, must roll hp at each level, etc. I love randomness, but again, attacking just doesn't fit as much as I'd like...

So, at any rate, this isn't sometime I am doing, just a design exercise, etc.

 

[LordEntrails]

Sadly, that is the issue... it really isn't all that random when you are hitting nearly all the time anyway...

Otherwise, as a DM I do pretty much everything you mention: re-roll initiative, cascading encounters, must roll hp at each level, etc. I love randomness, but again, attacking just doesn't fit as much as I'd like...

So, at any rate, this isn't sometime I am doing, just a design exercise, etc.

Earlier you (?) state the hit rate is typically %60-70. (Though sometimes higher.) That, to me, is not "nearly all the time". Removing attacks and just having damage only decreases randomness, by whatever that miss chance is or so. And from what I understand, your method doesn't introduce any increased randomness in having damage only.

In my experience, having two random events in a chain, increases the over entropy feel of the system. So even if you add randomness to the damage roll, you still only have a single instance of randomness, not two instances.

 

[DND_Reborn]

Earlier you (?) state the hit rate is typically %60-70. (Though sometimes higher.) That, to me, is not "nearly all the time".

Yes, it was me. At often (due to buffs, advantage, etc.) it is effectively much higher-- high enough to me that it is "nearly all the time." YMMV, of couse.

Removing attacks and just having damage only decreases randomness, by whatever that miss chance is or so. And from what I understand, your method doesn't introduce any increased randomness in having damage only.

First, it isn't really removing attacks so much as incorporating attacks into damage as other game systems do.

Instead of thinking of it as a "damage" roll, think of it as "attack effectiveness" roll--damage is only part of that. In 5E terms, since damage is already "there" I am just spit-balling with that mechanic.

But, no, it doesn't increase randomness.

In my experience, having two random events in a chain, increases the over entropy feel of the system. So even if you add randomness to the damage roll, you still only have a single instance of randomness, not two instances.

You might feel that way, but in the end it really doesn't work. If anything, it does create a more non-linear feel, but your results still range from miss (0 damage) to critical (double damage).

My idea has the same results: 0 damage (your "miss") to critical damage.

 

[p_johnston]

So I do like the concept of not having an attack roll just damage but I'm not sure 5e is the system for it.

So the big problem is that 5e is really not built to deal with Damage Reduction (DR) very well. (Note: I've tried introducing DR to my games before so I do have some actual play experience). To boil it down 5e treats the amount of damage a Character (PC or NPC) can do as a lump sum when calculating it's effect in battle whether or not that damage comes from a single attack or multiple. When you have DR this assumption breaks down pretty heavily, especially when the DR number starts to get high.

Let's look at some examples from both PC's and NPC's (Note I will be using average damage for pretty much all calculations)
PC: Let's look at the rogue vs the monk at say 5th level. The rogue is a consummate striker. All their damage comes from a single well placed strike. at level five they are probably doing 1d8+3d6+4 (19) damage. A monk at the same level through multi attack is doing 3d6+12 damage (24) without flurry of blows, each hit dealing 1d6+4 (8) damage. Now just for simplicity let's use page 274 of the DMG and say the average AC of a cr 5 monster is 15. That translates to 5 DR. Now the rogue get's to do 14 damage vs the monks 9. Ok now let's up it to level 11. The rogue is dealing 1d8+6d6+5 damage (31). The monk is dealing 3d8+15 (30) damage without flurry at 10 damage a pop. Now the average CR for a level 11 monster is 17, which would translate to a DR of 7. The rogue now deals 24 damage to the monks 9.

NPC: Let's look at a level one heavy armor fighter with a shield. His AC is 19 which translates to a whopping 9 DR. This means at level one pretty much any monster below CR 2 will end up dealing about 1 damage an attack. Now let's bring them up to level 5 and pit them against a couple of monsters that should be roughly the same challenge. Let's say AC 21/DR 11 now due to plate. They are fighting a Bullette and then a Troll. The Bullette deals 30 damage an attack or 19 through DR. A pretty solid hit. The troll deals 11 with one attack and 7 with it's other to for 3 damage a round.

To make DR work in 5e you need to not only rebalance every class to bring them back in line but you also end up needing to redo the entire way you calculate CR and encounter balance, and change a bunch of spells to make them fit. Exploding damage die do help this somewhat but, I would guess, not nearly enough.

TLDR: Replacing AC with DR breaks the fundamental combat math of 5e in such a way that will require you to change almost every aspect of the game to make it work.

 

[DND_Reborn]

So I do like the concept of not having an attack roll just damage but I'm not sure 5e is the system for it.

It isn't ideal, I am sure, but I think some of your numbers are off a bit. We've actually used AC as DR for a while now, and I really haven't had any issues with it. But I'll review your examples:

PC: Let's look at the rogue vs the monk at say 5th level. The rogue is a consummate striker. All their damage comes from a single well placed strike. at level five they are probably doing 1d8+3d6+4 (19) damage. A monk at the same level through multi attack is doing 3d6+12 damage (24) without flurry of blows, each hit dealing 1d6+4 (8) damage. Now just for simplicity let's use page 274 of the DMG and say the average AC of a cr 5 monster is 15. That translates to 5 DR. Now the rogue get's to do 14 damage vs the monks 9. Ok now let's up it to level 11. The rogue is dealing 1d8+6d6+5 damage (31). The monk is dealing 3d8+15 (30) damage without flurry at 10 damage a pop. Now the average CR for a level 11 monster is 17, which would translate to a DR of 7. The rogue now deals 24 damage to the monks 9.

The "average" AC is as follows (from my monster database of over 700 creatures):
tier 1: AC 13
tier 2: AC 15
tier 3: AC 17
tier 4: AC 19

So, I agree at level 5 using an estimated DR 5 (AC 15-10) is correct and DR 7 for level 11.

Rogue 5: 19 - 5 = 14
Monk 5: 23 - 5 = 18. I assume you are applying the DR to each monk attack to arrive at 9?

Rogue 11: 31 - 7 = 24
Monk 11: 29 - 7 = 22. I assume your 9 is the same thing?

Anyway, in the OP I specify that damage should be based on level, and attacks is one way to go but really you shouldn't apply DR to each attack with that model, but treat the attack action as grouped damage.

Then, if you look at the damages above, you'll see they are both adequate and close to each other.

NPC: Let's look at a level one heavy armor fighter with a shield. His AC is 19 which translates to a whopping 9 DR. This means at level one pretty much any monster below CR 2 will end up dealing about 1 damage an attack. Now let's bring them up to level 5 and pit them against a couple of monsters that should be roughly the same challenge. Let's say AC 21/DR 11 now due to plate. They are fighting a Bullette and then a Troll. The Bullette deals 30 damage an attack or 19 through DR. A pretty solid hit. The troll deals 11 with one attack and 7 with it's other to for 3 damage a round.

This would be a feature. Armor was very effective, so DR being higher due to it makes sense. But many creatures deal more damage than you seem to think. For example, an orc averages 9, but can deal up to 15! And with exploding damage dice, damage will be a bit more.

Again, if you group the troll's damage together, it is 29 damage, which would be 18 even against a PC with AC 21 when apply the DR once against the sum of damage for the "attack".

So, I think that solves any issue you have? If not, we can discuss it further.

 

[p_johnston]

So, I think that solves any issue you have? If not, we can discuss it further.

So if your just pooling all attacks together before all attack damage together before applying DR it does help with balance. With that I think that it would make trying the system out playable, however I would still be hesitant to use it.

There are still a few issues
1) Pooling all attacks as one before applying DR introduces a few problems of it's own. The one that comes to mind immediately is that it means you have effectively eliminated multi attack from the game. Now everyone just gets to roll more dice when they would get more attacks. There are two problems I can see with this
1a) You can't split your attack anymore. Before If someone with three attacks started to attack an enemy and brought them down with the first one they could move on and finish off the remaining two attacks on another enemy. Now your just going to be pooling every attack together to make sure you bypass the DR.
1b) It will (I suspect) make a lot of classes and playstyles feel a lot more samey. Playing a great weapon barbarian, a duel wielding fighter and a rogue currently feel a lot different mostly based on the number of attacks you end up getting. Now all three just get a handful of dice to throw each time.
2) You still have problems with DR when it comes to attacks where you don't get to do multi attack. The biggest example is opportunity attacks would kinda suck in this system for anyone except rogues. But the cleave feature of great weapon master is another example of something that gets really nerfed.
3) Exploding damage die are a cool idea but on average won't add all that much damage (average of about 1 extra point of damage)
4) This ends up cutting out advantage which will end up effecting a lot of abilities and spells.
4b) no advantage combined with no splitting attacks will likely end up making combat less dynamic then it already is. Why bother tripping, using the help action, or anything else if you can't get advantage? 5e already heavily incentivizes martials to just run up and mindlessly attack a single target. This just makes that worse.
5) There are a lot of minor rules, abilities, and spells that become much better, much worse, or just don't work if your never rolling to attack.
For example
Stunning strike becomes a whole lot better when you no longer have to roll to hit. Same with paladin smites, arcane archer arrows, all blade cantrips, etc.
How does bardic inspiration work now without attack rolls?
To flesh out point 4 a bit, how does advantage work? Am I just as effective hitting people while blind and prone as I am regular?
Connected does blindness/deafness now just not do anything?
How about the half-orc/barbarian extra dice on a crit feature?
How does magic missile work now that everything is doing what use to be it's niche?
And these are just the abilties that I can come up with off the top of my head. I'm sure there are many more that are similarly broken.

To reiterate I do think that the idea of a system where rolling to hit isn't a thing is interesting. I do not think that trying to hack 5e into being that system is a good way to go.