Is this Visions of Danger spell balanced for level 7?

Question

So I am looking to homebrew a 7th level sorcerer spell because I am not entirely happy that the options match my characters theme.

I have seen Visions of Danger in Pathfinder 2e and have made an attempted conversion into 5e rules. I am looking for advice on if it is balanced at my first pass.

Visions of Danger: 7th level illusion

Casting time: 1 action

Range: 120ft

Components: I don't care

Duration: 1 minute (required concentration)

Effects: You create a 20ft radius sphere filled with dangerous looking illusions of your choice at any point you choose within range. The illusion spreads around corners but not through total cover.

When a creature enters the area for the first time, or starts its turn there, the creature must make an Intelligence saving throw. The creature takes 8d8 phychic damage on a failed save and has its speed reduced to 0 as it believes it is grappled by the illusion. On a successful save the creature takes half damage and disbelieves the illusion so will take no additional ill effects on any future turns.

As an action, a creature 'grappled' within the area can attempt another save to disbelieve the illusion. A successful save will end the speed reduction and allow movement as normal as the spell effectively ends for them.

My line of thinking:

There are very few high level spells with ongoing damage effects, so this was a tricky one. I started with Cloudkill (5th level) so that is the lowest level this could be. Scaled up to 7th level Cloudkill would do 7d8 damage, but scaling is never good so I added an extra 1d8.

I then added the extra effect of the speed reduction because I feel that especially at high levels a spell like this needs some way of keeping people in the area.

The ability to disbelieve the illusion is unusual in 5e, as although disbelieving illusions is a thing in the rules I am not aware of it with a damage spell. This probably means there is a lot of additional tracking to do, but my DM will be ok with this and I will help keep track of stuff so I don't take that kind of thing as being a negative as I might if I was creating this to actually publish or if I were answering someone else asking this question.

I have also looked at Incendiary Cloud (8th level) which does 10d8 damage vs 8d8, and heavy obscurement vs grapple. Both Incendiary Cloud and Cloudkill move, but they are area denial spells for their entire duration, unlike this spell.

I therefore think the riders balance out, and the lower damage on this spell means 7th level is about fair.

My only concern is that Incendiary Cloud is (certainly in optimisation circles) a terrible spell, but it is official so I can't exactly clone it, call it lower level and claim it to be balanced.

A bit more reasoning:

Some spells to consider in terms of balance are Phantasmal Force (2nd level), which is where I took the 'use your action to disbelieve' from. The way I play that spell is that I always make the character make the check, and I narrate it as them dealing with the illusion. So if they fail then they swung at the imaginary creature and missed, but if they succeed they realise they hit and it is an illusion. I know the spell doesn't say that exactly, but there are so many arguments over how that spell works I think this is the only fair use.

Also (Evards) Black Tentacles which is similar in that it does damage in an area over a period of time and restricts the ability to escape. That is a 4th level spell doing 3d6 damage. You can't upcast it, but if it followed the usual upcast rules it would do 6d6 at level 7. That is less that this spell, but it includes difficult terrain that I don't have, restrained which is far worse than grappled, and is a check (not a save) to escape. So not only is it harder to save from after the initial save, but the difficult terrain means that most things are going to be stuck in the area for more than one round even if they do save. With this spell one save ends it for them entirely, no being pushed back in, no failing to get out of the area etc.

From a powergaming perspective a spell which remains even after a save is a million miles better, because you can all but ensure the target stays in for the duration even if they make their save all 10 times. This spell has effectively zero synergy with anything other than spells like Bane which make saving throws harder.

Also Maddening Darkness (8th level) was brought up in an answer, but that has more range (not much of a difference), massively larger area, 10 times the duration and the targets can't see (you can't see them either, but you cast the spell presumably when that is a bonus to you). It will take most creatures 2 turns to escape without using an action, but I find it heavily metagaming to assume that a creature knows to make for the nearest (or indeed any) edge in order to escape if they can't see the exit. The spell also lingers, meaning that any good synergy can keep an enemy in the area for a very long time. It does the same damage, but if you cast it in the circumstances it is meant to be used for you will hit way more targets than this spell has any chance of doing.

I think it can seem more powerful than Maddening Darkness, but that is because the situations where Maddening Darkness are truly useful are few and far between, and that isn't a balance issue.

Lastly: Intelligence saves are rarer for monsters to have, but I don't believe that the type of saving throw is accounted for in standard balance equations. If you want to include it in your analysis please do, and I would happily change it to a Wisdom save, but it is worth nothing that I am specifically looking to pigeon hole my character into Intelligence based psychic damage spells, so in a way it is a weakness of mine that I would be building into with this spell. The DM knows this and I know something like a mindflayer dragon is likely just around the corner for example and we have both discussed that I have no issues with their being some fights that are just going to be tough for me mechanically. That doesn't make much difference for standard balance, but if you end up thinking this is level 7.5 then it might tip the balance that for me 7th is appropriate.

Describing the flavour of this spell to explain why I picked these mechanics:

Mostly I will create an army of heavily tentacles eldritch horrors grabbing and attacking, so when a character uses their action to disbelieve the illusion that would be narrated as battling the horror in an attempt to escape. A failed save would mean they missed, or failed to cause damage. A successful save would mean their blow went through and they realised it was an illusion. So it isn't that they just think really hard about it or have to suspect it is an illusion, it just represents them dealing with it. If they do something else with their action, then that is ok but they would have no reason to suspect it is an illusion and doing something else wouldn't give them any clues because they aren't interacting with it physically.

Answer 1

This should probably be a level 8 spell

There are multiple factors influencing how good a pure damage spell is to deliver impact: area, range, the raw damage output per round, how fast that damage is delivered (earlier is better), what save is used to avoid the damage and what damage type it is, because opponents may have resistances. Then there are secondary effects like grappling and impairing vision that can have a big influcene.

Saving Throws: For monsters up to CR20 from the Monster Manual, the average chance to fail a Constitution save is about 51%, for Wisdom it's 59%, Dexterity 62% and for Intelligence it is a whopping 71%. Monsters do not get smarter as much as they get tougher with increasing CRs.

Damage Types: Monsters often have resistances, and some damage types are more resisted than others. For monsters up to CR20 from the Monster Manual, poison is the most resisted damage, with 37% of monsters being resistant, while psychic damage is rarely resisted, with only 4% of monsters being resistant.

The chart below shows the expected cumulative damage dealt for each round, accounting for the influence of damage types, saving throw types, and the fact that you end the effect with a successful saving throw for Visions of Danger:

As one can see, during the first two rounds, the impact is pretty much the same as for the level 8 Maddening Darkness. It then starts to decline, but even by round 4 is still above a Cloudkill upcast to level 7. The average fight only lasts 3-4 rounds, so the earlier rounds are more important for determining how to evaluate this, and this means the damage behavior is a better match for a level 8 spell, even if you account for the fact that upcast damage tends to be a bit lower than that of level-native spells. In the first round, it even slightly beats the level 8 comparator (due to saving throw).

Speed 0 vs Obscurement

The secondary effects are important too. This is still an area effect spell that can be expected to hit at least 4 opponents (DMG, p. 249) and often many more, so you are likely to cast it against mobs. Even on higher levels, not all opponents can teleport, and in particular creatures that appear in numbers may not have any way to escape the effect other than making the save.

Stopping a creature from moving can be very useful in and of itself, too: if that creature lacks ranged attacks, not only has it no way to escape, it also has no way to close the distance to you and your Sharpshooter pal, who can kill it off risk-free from a distance.

Compare this with Maddening Darknesses obscurement: that effect is symmetrical, at least unless you employ tricks like a warlock with Devil's Sight. It will stop you from targeting the creatures in the effect as much as it will stop them from targeting you.

You can argue that narratively, blocking sight should work nearly as well as being immobilizing to keep creatures from finding exits from moving out of the spell's area (and I am with you on this how hard it is in real life, even if you know the surroundings), but keep in mind that in most cases the PCs are intruding on creatures on their home turf, in surroundings that they know well. At least my DM would let them know where the exit is and move there without much difficulty at all, so this will be heavily dependend on how your DM runs it.

Also observe how much lower Evard's is in damage output, compared to cloudkill, while it is only one level lower at 4th instead of 5th. Its effectively less than half the damage output. While restrained is stronger than speed 0, a large part of that must be due to keeping the creatures in its damaging area instead of just obscuring vision. If you applied the same discounting logic to the damage of Maddening darkenss, then for Visions of Danger to be a level 7 spell, you would nearly need to half the current damage at least during the early rounds.

I think overall, especially in combination with sustained damage, reducing speed to 0 is stronger than blocking vision, and further pushes this in the direction of level 8.

Range, Area and Duration

Both these factors are minor influences on power for the spells in question and should not change the leveling.

The range of 150 feet for Maddening Darkness is a bit longer, but both spells are not short in range, and both of them are out of counterspell range, so I think that difference is not material.

The area is larger for Maddening Darkness, but in my experience, areas larger than 20 to 30 feet diameter in radius are very situational and can be as much a liability as they can be a benefit. If you are fighting in open areas like a desert, then the large area is great. If you are fighting in the confined space of a dungeon (as often happens), then the large area can be a real downside, because you may not be able to place the center far away enough to not hit your own team. This means it makes the large radius spell more situational. A 20-footer, you can pretty much always employ usefully.

The duration is also longer for Maddening Darkness, but as discussed above, most fights are over long before the minute has passed, so at best this can serve to block a choke point for longer while you are trying to get away. Again, can be nice, but is quite situational, and not an important power consideration.

Conclusion

While this should be level 8 as written, if you used a Wisdom save instead of Intelligence, this would hew closer to a level 7 damage profile, and could work as a level 7 spell. Alternatively, if the Intelligence save is important to you, you could use 7d8, which would have a similar effect on overall damage output in aligning this better to level 7. I think based on the Tentacles/Cloudkill comparison, it would still be very solid in that form as a level 7 spell.

Here are the charts for using a Wisdom save instead, or for using 7d8:

Answer 2

Let's compare to Maddening Darkness, an 8th level spell that I consider a good 8th level spell.

MD:

• 60 ft radius, significantly larger
• 8d8 psychic versus Wisdom save
• Magical Darkness

VoD:

• 20 ft radius
• 8d8 damage vs Wisdom save for half
• Speed = 0 for a failed save
• Note there is no save possible at the end of every turn in this very damaging area.

I suggest that your spell is strictly better than Maddening Darkness, mostly due to the movement speed effect. Creatures will effectively be trapped there unless carried out by a different creature who passed the save (or some other magical movement effect). Maddening Darkness also has trouble being cast since you need a lot of space to get such a large radius in play, and only Warlocks will reliably have a way to keep enemies in it.

Answer 3

Not strictly a numerical balance thing but "as an action" type recurring saves don't look like a very good fit for illusion type effects.

After all, if the target fails the initial saving throw, they've got practically zero reason to suspect this is an illusion and will thus never use their action for a follow-up save.

I'd recommend therefore instead something like this:

At the end of each of its turns, the target can make another Intelligence saving throw.

Although it must also be noted that Phantasmal Force, another illusion/phantasm spell that can damage, features a similar active disbelief mechanic:

The target can use its action to examine the phantasm with an Intelligence (Investigation) check against your spell save DC. If the check succeeds, the target realizes that the phantasm is an illusion, and the spell ends.

However this is counterbalanced by the fact that as well as the damage being fairly low for a 2nd level spell, it's a strictly single-target spell while yours can potentially affect dozens of targets.