What if Barkskin made the d20 have a minimum raw value to hit?

[Segev]

The obvious one is 16, meaning the attacker has a maximum of 25% chance to hit.

This would make the spell useless against anything that already needs a minimum of 16 or higher to hit the druid, but would be fairly nice against high to-hit creatures. I think that might actually make it worth its Concentration requirement. But would it be too good? Too much effort to track?

This would be instead of setting a minimum AC.

[Keltest]

I think "you can never have more than a 25% chance of hitting me" is kind of an insane defensive buff for a 2nd level spell, concentration or not, especially because you can cast it on other creatures as well.

[diplomancer]

I'd put it at something a lot lower, like 11. But I don't think it's a good idea, regardless of balance. It just feels weird that the Tarrasque and a Goblin have the same chance to hit the 15 or less AC Druid.

Such a spell could work perhaps, but with an entirely different "fluff" than "you have a particularly tough skin".

[Keltest]

You could do something like have the level 2 version set the min to 8, then add an extra point to the min for each spell level you upcast it. As a 9th level spell a minimum of 15 on the die to hit isnt too insane, especially for concentration.

[Sindeloke]

We've been doing "DR equal to 2x spell level", because if there are enough spells to fill out a golf bag you might as well make them an actual golf bag worth of spells; making something actually mechanically distinct means it's more likely to have a situation where it's actually the most useful and another where it's less, versus a multitude of spells that all just always add to the exact same stat and inevitably there's one correct solution.

Analyzing this change from that perspective, we go from a niche that's already occupied by shield/shield of faith/mage armor/robe of the magi/just actually wearing some damn armor, to a niche with nothing in it other than the self-only mirror image. That has potential. As you point out, it's the right choice against high +hit creatures, but less useful than a +AC spell against a low +hit creature, so it has its own use case. @Keltest might be right that 25% is too low; using mirror image as a guide, a 12 or higher might be better (not as good as the first image, better than the last image, and we can assume that 'cast on someone else' vs 'cast on self' part is balanced by the concentration).

@Diplomancer is also right, though, that the mechanic doesn't exactly match the fluff of "barkskin." You could get something a little more akin to it by saying that the target spontaneously grows a bunch of tough, leafy branches that stick out at weird angles and seem to try to grab enemy attacks, but I'd personally rename that to something more evocative of the idea, like... idk, "defensive growth" or "branching barrier" or something, and at that point we're pretty much just making a new spell.

[Goobahfish]

The mechanic is interesting, but it really doesn't fit the flavour of 'bark' skin.

It would make more sense with a aesthetic like clockwork soul or fortune vibe.

From a balance perspective, I agree 16 is high. Maybe 14? That is 1 in 3?

[Kane0]

*Checks notes*

Here's what I did to it:

Barkskin
2nd-level Transmutation
Casting Time: 1 action
Range: Touch
Components: S
Duration: 8 hours
You touch a willing creature. Until the spell ends the target's AC cannot be less than 17 regardless of what kind of armor it is wearing. When the targeted creature is struck by a critical hit or takes fire damage this minimum AC is reduced by 1 (maximum once per turn). The spell ends when this minimum AC reaches 10 or the target's natural AC.
At higher levels: The minimum AC increases by 1 per spell level above 2nd (reaching a maximum of 24 with a 9th level casting)

[MrStabby]

I think its a creative way to adjust the spell though I agree a flat requirement to hit is athematic and also maybe a hard to balance between too powerful and too weak.

If you want to make creatures harder to hit and/or better vs more elite enemies I might suggest something like not getting to add strength to to-hit rolls. It's a big bonus to AC, especially against higher level enemies but it does little at low levels and nothing vs archers or attack roll spells. I think it would be higher than a level 2 spell but it might have a role.

[Mastikator]

Then it would be a better mirror image. Barkskin is fine for a wildshaper, just allow it to upcast, say +1 AC per 2 spell levels. I think it's fine to have a protection spell that raises the floor but not the ceiling.

[Amnestic]

Then it would be a better mirror image. Barkskin is fine for a wildshaper, just allow it to upcast, say +1 AC per 2 spell levels.

Honestly you could make it +1AC per spell level for upcast purposes. 4th level slot for 18 AC? 6th level slot for 20? 9th level slot for 23?

Those are all probably fine, assuming you're actually taxing the spell slots in a day.

[Segev]

As a concentration spell, it typically fails within the first hit or two. Which means as an AC booster, it goes down sometimes before it makes any difference at all.

I have never cast it as a player, and when I have used it as a DM, it didn't affect the fight much even when I remlved the Concentration requirement by making it a potion.

[Mastikator]

Honestly you could make it +1AC per spell level for upcast purposes. 4th level slot for 18 AC? 6th level slot for 20? 9th level slot for 23?

Those are all probably fine, assuming you're actually taxing the spell slots in a day.

Yeah it's probably fine, being concentration 1h and all. 23 AC is good but not amazing at level 17, but I don't think full casters in general should have such easy access to high AC values, expending a high level spell slot for it is IMO hard access.

[Keravath]

The version limiting the die roll required to hit would be too powerful.

The current version gives a minimum AC16. Under what circumstances would the revised version requiring a die roll of 16+ be better? The answer would be always except in the case of a creature with +0 to hit in which case they would be the same.

In addition, the 16+ to hit version only gets better as the +to hit increases. Consider in tier 3 and a creature with +10 to hit. Requiring a 16+ to hit is the same as granting an AC of 26. Against a dragon or something with +14 to hit, the 16+ requirement is like giving an AC of 30.

So, no :) ... the version requiring a fixed die roll is broken in terms of the 5e system of bounded accuracy. (Interesting idea though).

Some of the other suggestions here though, scaling it with spell level, removing concentration and extending duration seem pretty good though if you are going to scale Barkskin with level you'd probably also want to scale Mage Armor. A 16AC minimum is equivalent to Mage Armor with a 16 dexterity (pretty typical for a wizard). On the other hand, Barkskin will stack with wild shape and one of the significant limitations of the wild shape forms is the AC. An AC of 16 is a substantial improvement already over the typical 11 AC for a bear. Scaling that up is likely to help all the moon druid wildshapes and moon druids often have the spell slots for upcasting and the biggest need for improved AC.

[DruidAlanon]

Honestly you could make it +1AC per spell level for upcast purposes. 4th level slot for 18 AC? 6th level slot for 20? 9th level slot for 23?

Those are all probably fine, assuming you're actually taxing the spell slots in a day.

Meanwhile, with medium armor & sheild + magic initiate feat you could cast shield, a 1st lvl spell, and have 24 AC. Spending a 9th level spell for 23 AC seems unecessary.

Barkskin would have been just fine if it was 16AC, 8hours, no con, just like mage armor. That would fix the spell but would simultaneously break Moon druids even more.


On the actual proposal of the thread, a 25% chance would kinda work fine as a reaction? i.e. on a sucessful attack, you roll a 1d4 and you have 25% to hit. With a flavour of something like a fungal body or so. That would in similar lines with silvery barbs. As a 2nd spell I think that wouldn't have been broken, I think.

For example, a standard to-hit probability is 60% - 65%.

Sivery barbs:
with 60% probability to hit, the combined probability is 21.6%
with 65% probability to hit, the combined probability is 27.4%
On average (60-65%) it is 24.5%.

Therefore, forcing a re-roll and a 25% would put such a spell right next to Silvery Barbs. Then add the advantage on top of it and you'd have the exact equivalent in the Druid list.

[Theodoxus]

Not saying I'd do this to barkskin, but I could see design space where you're protected by magic similar to, but stronger than, Mirror Image (so definitely higher level) where AC and To Hit don't matter. You just need to roll higher than X on a d20 to hit. Definitely sounds like a mixture of Abjuration and Divination magic.

Though I suppose if you start with a 1st level spell, and starting at nat 11 to hit, and increase +1 per spell level to max out at needing a nat 19 with a 9th level spell, that might work. 50/50 to hit as a first level spell, anytime you can force disad on rolls would make it a decent spell... Probably better than mage armor (though I'm crappy at mathing out probabilities...) Just not sure if it should be 8 hours, non-conc, or 1 hour (w/ or w/o conc) or 10 minutes... I could also see it being akin to shield, 1st level, reaction, all attacks that round need to roll a nat 16 or something...

[Amnestic]

Could make it damage reduction.

Reduces weapon damage by [x] (minimum of 1) until [y] damage reduced, at which point the spell ends. Upcasting increases X or Y or both.

I dunno what a decent balance point is. Reduce it by 3 each until 24 points reduced? Effectively +24 hit points if it all gets used up, but since it requires concentration and shorter duration it needs a small boost over Aid's 15, perhaps.

The "minimum 1" is so concentration saves still gets pipped on those taking damage.

Kinda like how stoneskin in 3e worked. If HAM can exist as a weapon damage reduction number, barkskin can too (and you could include a clause about them not stacking if you cared to).

[cZak]

Could make it damage reduction.

Reduces weapon damage by [x] (minimum of 1) until [y] damage reduced, at which point the spell ends. Upcasting increases X or Y or both.

I dunno what a decent balance point is. Reduce it by 3 each until 24 points reduced? Effectively +24 hit points if it all gets used up, but since it requires concentration and shorter duration it needs a small boost over Aid's 15, perhaps.

The "minimum 1" is so concentration saves still gets pipped on those taking damage.

Kinda like how stoneskin in 3e worked. If HAM can exist as a weapon damage reduction number, barkskin can too (and you could include a clause about them not stacking if you cared to).

This always seemed more thematic to me for the name
Maybe Resistance to Bludgeoning & DR vs piercing

[KorvinStarmast]

Segev: your proposal is in the 'too fiddly' category. It is that kind of die and number manipulation that 5e explicitly tried to get away from. (Yes, there is that residual thing with a Sprite's arrow putting you to sleep on a 5 or les that they did not clean up). Not a good fix for Barkskin.

This is a better one.

*Checks notes*

Here's what I did to it:

Barkskin
2nd-level Transmutation
Casting Time: 1 action
Range: Touch
Components: S
Duration: 8 hours
You touch a willing creature. Until the spell ends the target's AC cannot be less than 17 regardless of what kind of armor it is wearing. When the targeted creature is struck by a critical hit or takes fire damage this minimum AC is reduced by 1 (maximum once per turn). The spell ends when this minimum AC reaches 10 or the target's natural AC.
At higher levels: The minimum AC increases by 1 per spell level above 2nd (reaching a maximum of 24 with a 9th level casting) I'd go with "not less than 16 to start with and a max of 23" but I find this to be a better attempt to replicate the idea behind barkskin.

[Oramac]

We've been doing "DR equal to 2x spell level"...snip

Personally, I think this is the way to go. Just a flat across-the-board damage reduction based on the spell level. Depending on the wording, it could also apply before the concentration check too, which helps offset the need for concentration.

[Psyren]

I prefer the temp HP version from the playtest, provided that druid AC (specifically moon druid) is improved. Barkskin should be useful in melee; the version proposed here is too strong while the existing version is too weak.

[rel]

The proposed change doesn't seem in line with the 5e design ethos for whatever that's worth.

The spell will actually get stronger as you level up and face enemies that tend to hit you more reliably but for less of your total HP. Again, not sure how relevant that is.

[DruidAlanon]

The proposed change doesn't seem in line with the 5e design ethos for whatever that's worth.

The spell will actually get stronger as you level up and face enemies that tend to hit you more reliably but for less of your total HP. Again, not sure how relevant that is.

you can get the same number (25%) under plausible assumptions with a reroll, just like silvery barbs. I'd change it to a reroll triggered as a reaction.

Then the question is whether this should be a concentration spell and if so, for how many mins (I'd go for 1min) and at what level (>= 3 I pressume). Also, I wouldn't probably allow this on Legendary actions, to partly account for the overscalling issue that you mention.

[KorvinStarmast]

Then the question is whether this should be a concentration spell Since Mage Armor isn't ... barkskin should not be.

[Psyren]

If it's not concentration then you can stack Barkskin and Stoneskin. The intent seems to me to be that you run with one or the other, ideally one at low levels and the other at higher levels when you're wealthier.

[Mastikator]

If you remove concentration then it becomes considerably stronger, in which case adding regenerating temp hp would make it overpowered, IMO without concentration a min AC of 16 (or even 17) is fine, but no more.

[Tanarii]

For self buffing, Barkskin is already +3 to +5 AC for most animal forms.

And of course it's not a self-only spell. It's similarly rocking on the AC 12 arcane squishy.

Barkskin has a solid niche and my experience was it got used with some frequency in Tier 1, starting to drop off in late Tier 2. Which felt perfect to me.

[stoutstien]

For self buffing, Barkskin is already +3 to +5 AC for most animal forms.

And of course it's not a self-only spell. It's similarly rocking on the AC 12 arcane squishy.

Barkskin has a solid niche and my experience was it got used with some frequency in Tier 1, starting to drop off in late Tier 2. Which felt perfect to me.

Eh 16 isn't high enough to bank on it lasting that long. I'd take warding wind 9 out of 10 times as a moon druid if I was looking for a spell to have up while shifted.

[diplomancer]

Honestly? I'd just put it at level 1. It becomes a decent option to concentrate on at lower levels, and there's no need for spells to be relevant at all levels (in fact, few lower level concentration spells are relevant at all levels). Maybe allow it to upcast for +1Ac/level.

[Willie the Duck]

The obvious one is 16, meaning the attacker has a maximum of 25% chance to hit.

This would make the spell useless against anything that already needs a minimum of 16 or higher to hit the druid, but would be fairly nice against high to-hit creatures. I think that might actually make it worth its Concentration requirement. But would it be too good? Too much effort to track?

This would be instead of setting a minimum AC.

Given that this:

does not stop an existing miss from being a miss, and
stops a certain sub-set of hits from being hits

I don't see this as being significantly different from bringing back a '% miss chance,' akin to 3e blur. This means you don't have to roll two dice, and there are situational differences (ex. a miss chance allows a forced miss on a 20, negating a crit, while this method does not), but overall they seem much the same. I'm not sure if they got rid of those for a specific reason (probably the design ethos point rel made), but in general I haven't missed the added fiddliness (referencing KorvinStarmast's point) that presented.

Overall, D&D has two major metrics of straightforward combat durability -- AC and HP. Whenever someone proposes adding another, my general question is 'what benefit does the game acquire by including this instead of another spell that deals with AC or HP (or no new spell at all)?' 5e added Advantage/Disadvantage, which is useful in reducing the number of +/-s built into the system, serving the whole bounded accuracy endeavor* (and it seems to do a fairly good job of it**). It also addedwildly enhanced the amount of temporary HP-granting abilities, useful compared to healing and just adding HP (I think only Aid does this, right?) in that you can do it preemptively, and that it doesn't stack (so having multiple sources of it is diminishing returns). I'm not sure what this proposed added to-hit mechanic does that actively enhances the game, other than just adding one more protection you can throw on a character who is already buffed up in all the other avenues (i.e. they already have access to Shield, someone has Shield of Faith on them, they have Aid and temp hp and the through one way or another the opponent will be at disadvantage to hit, etc.).
*most (and the places where this isn't true being sore points for many) abilities which instead provide actual +s or -s to hit are much more limited -- one-offs like Shield or concentration spells, etc.
**whether bounded accuracy was a worthwhile goal to begin with is a question I get why we are still debating...