UA Starter Spells, specifically, Chaos Bolt

[jnayls]

I feel like the Bolt should have a greater chance to jump with higher spell slots. But not to the point of being broken or abusable. I would hope they had made it scale better, but...


Chaos Bolt
1st-level evocation
Casting Time: 1 action
Range: 120 feet
Components: V, S
Duration: Instantaneous
You hurl an undulating, warbling mass of chaotic energy at one creature in range. Make a ranged spell attack against the target. On a hit, the target takes 2d8 damage. Choose one of the d8s. The number it rolled determines the type of damage, as shown below.
d8 Damage Type
1 Acid
2 Cold
3 Fire
4 Force
5 Lightning
6 Poison
7 Psychic
8 Thunder
If you roll the same number on any two d8s, the chaotic energy leaps from the target to a different creature of your choice within 30 feet of it. Make a new attack roll against the new target, and make a new damage roll, which could cause the chaotic energy to leap again.
A creature can be targeted only once by this mass of chaotic energy.
At Higher Levels. When you cast this spell using a spell slot of 3rd level or higher, each target takes extra damage of the type rolled. The extra damage equals 1d8 at 3rd, 2d8 at 5th, and 3d8 at 7th.

My attempt to fix a spell that's not broken, but also doesn't shine quite as bright as it could. Emphasis mine, on the key changes from original.

[Eragon123]

What if you cast it as a fifth level spell and get two doubles?

As written only one bounce, is this the intent? Why not scale it up to ninth level? It scales fairly poorly anyway.

[pwykersotz]

That can get out of hand pretty quick. I'm not great at probability math anymore, but the odds of any two numbers being doubles is pretty insane with 5 dice.

I did a quick check using the WOTC dice roller, and I rolled 5d8 27 times before I failed to get at least one double.

[Naanomi]

I would only have the dice that roll doubles actually 'bounce', so the attack would get weaker (and less likely to bounce) as it went

[jnayls]

Chaos Bolt
1st-level evocation
Casting Time: 1 action
Range: 120 feet
Components: V, S
Duration: Instantaneous
You hurl an undulating, warbling mass of chaotic energy at one creature in range. Make a ranged spell attack against the target. On a hit, the target takes 2d8 damage. Choose one of the d8s. The number it rolled determines the type of damage, as shown below.
d8 Damage Type
1 Acid
2 Cold
3 Fire
4 Force
5 Lightning
6 Poison
7 Psychic
8 Thunder
If you roll the same number on any two d8s, the chaotic energy leaps from the target to a different creature of your choice within 30 feet of it, making only one leap per damage roll. Make a new attack roll against the new target, and make a new damage roll, which could cause the chaotic energy to leap again. Each time the energy leaps, remove one damage dice to a minimum of 2d8.
A creature can be targeted only once by this mass of chaotic energy.
At Higher Levels. When you cast this spell using a spell slot of 3rd level or higher, each target takes extra damage of the type rolled. The extra damage equals 1d8 at 3rd (total of 3d8), 2d8 at 5th (total of 4d8), and 3d8 at 7th (total of 5d8), and 4d8 at 9th (total of 6d8).


Eragon123
What if you cast it as a fifth level spell and get two doubles?

As written only one bounce, is this the intent? Why not scale it up to ninth level? It scales fairly poorly anyway.

Valid points, corrected. I do feel like the bounces allow it to scale differently.


pwykersotz
That can get out of hand pretty quick. I'm not great at probability math anymore, but the odds of any two numbers being doubles is pretty insane with 5 dice.

I did a quick check using the WOTC dice roller, and I rolled 5d8 27 times before I failed to get at least one double.

Also valid. Didn't consider the doubling situation, also not the greatest at probabilities, and I think that corrects the ability somewhat.


Naanomi
I would only have the dice that roll doubles actually 'bounce', so the attack would get weaker (and less likely to bounce) as it went

Sort of... And I stole your idea somewhat to address the other concerns, so Thanks 😁

[Arcangel4774]

That can get out of hand pretty quick. I'm not great at probability math anymore, but the odds of any two numbers being doubles is pretty insane with 5 dice.

I did a quick check using the WOTC dice roller, and I rolled 5d8 27 times before I failed to get at least one double.

Chaos bolt can t bounce to the same person twice, it's really not too much a worry that it's gets insane, unless you have TONS of mooks to blow up.

EDIT* By this I mean there's no reason to take away damage as it's not hitting a guy more than once, and frankly it's damage is not overly competitive with other aoe spells. A fireball or cone of cold at the same level will probably do more damage to the same amount of guys, unless there spread out in such away that the only way to hit them all is to daisy chain it.

[pwykersotz]

Chaos bolt can t bounce to the same person twice, it's really not too much a worry that it's gets insane, unless you have TONS of mooks to blow up.

EDIT* By this I mean there's no reason to take away damage as it's not hitting a guy more than once, and frankly it's damage is not overly competitive with other aoe spells. A fireball or cone of cold at the same level will probably do more damage to the same amount of guys, unless there spread out in such away that the only way to hit them all is to daisy chain it.

I agree with you analysis, but disagree that it's not something worth correcting. The fact is that a high chain is nearly guaranteed, and spells shouldn't be designed in such a way as to deal damage so far off curve, even if the uses are niche. There are lots of ways to limit it and still keep it feeling interesting and make it useful, but not limiting it seems like a poor answer.

That's why a lot of us run ideas through the forum, after all. To identify those edge cases that could prove problematic.

[Kryx]

RAW Chaos bolt is a pretty terrible 1st level spell. It has a 1/64 chance to bounce (1/8 * 1/8) and does pitiful damage and scales horribly.

I've changed it to 3d8, scaling by 1d8 at each higher level, and bouncing on a 8 from the original 3d8. That puts it in line with other spells by my spell balance numbers (https://docs.google.com/spreadsheets/d/1N4QC6EmXE0avgk8jK1aubJcaFoZDYw8b_DuPHh8aBTc/edit#gid=639488216). This way it has a bout a 33% chance to bounce each time. If it bounces it'll do more damage than a 1st level spell, if it doesn't it'll do slightly less damage than a 1st level spell.

[Eragon123]

Acutally there are 8 pairs and 64 combinations. so it's a 1/8 chance of bouncing.

[jnayls]

Took me a while to calculate the probabilities, but this is what I've come up with (note that I do believe this is a gross simplification of way this probability is truly calculated but... I could find no better.)

6d8 (~92.3%)
5d8 (~79.5%)
4d8 (~59.0%)
3d8 (~34.3%)
2d8 (~12.5%)

Based on the formula as follows:

n = number of sides on dice
y = number of dice rolled
n * n-1 * n-2 ... / ny
8 * 7 * 6 * 5 * 4 * 3 / 86
8 * 7 * 6 * 5 * 4 / 85
8 * 7 * 6 * 5 / 84
8 * 7 * 6 / 83
8 * 7 / 82

And based on this and quick look through Kryx's spell doc, I've bumped the based to 3d8 with a minimum of the same. So that at 9th it is 7d8 (~98.1% chance of leaping on first strike). Seemingly I would probably never cast Higher than 5th myself, but it does scale well against hordes and the like at higher level slots.

Note: Any Statisticians out there feel free to educate me on this as once you get to 9d8 this method of calculating falls apart showing that there is a chance (however slim) of rolling non-repetitive, even though its (irl) impossible.

[Kryx]

Acutally there are 8 pairs and 64 combinations. so it's a 1/8 chance of bouncing.
I misremembered RAW due to this thread's discussion and my changes. You're correct. I've updated my doc - it's still terrible by RAW.

[pwykersotz]

Acutally there are 8 pairs and 64 combinations. so it's a 1/8 chance of bouncing.

I'm behind on my statistical math, but I know for certain that this is wrong.

Exhibit A: The Birthday Problem (https://en.wikipedia.org/wiki/Birthday_problem)

[Kryx]

I'm behind on my statistical math, but I know for certain that this is wrong.
A pair on 2d6 is 1/6. See https://math.stackexchange.com/questions/1181129/why-are-the-probability-of-rolling-the-same-number-twice-and-the-probability-of

The same would be true for 2d8: 8* 1/64 = 8/64 = 1/8. 12.5%

[pwykersotz]

A pair on 2d6 is 1/6. See https://math.stackexchange.com/questions/1181129/why-are-the-probability-of-rolling-the-same-number-twice-and-the-probability-of

The same would be true for 2d8: 8* 1/64 = 8/64 = 1/8. 12.5%

This is frustrating, because I used to know how to explain this. The statistics are unintuitive.

I'll let Wolfram Alpha talk for me. (http://www.wolframalpha.com/input/?i=5d8)

http://www4f.wolframalpha.com/Calculate/MSP/MSP6302273fc3i23h74h4000001cghdhdehf0edhfe?MSPStor eType=image/gif&s=43

As you can see, the odds of rolling all different faces on your 5d8 are about 20%. The odds of rolling at least one double is around 80%.

Edit: In other words, jnayls' math is right on the money.

[Kryx]

Yes, but you said 2d8 having one pair being 12.5% was wrong, but WolframAlpha shows that as well: http://www.wolframalpha.com/input/?i=2d8

The houserule that jnayls is suggesting is madness. The spell becomes absolutely ridiculously powerful the higher level you cast it.

[pwykersotz]

Yes, but you said 2d8 having one pair being 12.5% was wrong, but WolframAlpha shows that as well: http://www.wolframalpha.com/input/?i=2d8

The houserule that jnayls is suggesting is madness. The spell becomes absolutely ridiculously powerful the higher level you cast it.

I may have misunderstood if you think I said that. I have been talking about 5d8 the entire time. I thought that the comment I responded to was talking about a flat statistical certainty across all dice rolls. That is probably my bad.

And yes, the original houserule is excessive, my first post was to that effect. I think the dropoff is one way to reel it in. Probably not the way I would go, but I don't know what way I'd go, so I'm no help on that angle.

[Eragon123]

I'm behind on my statistical math, but I know for certain that this is wrong.

Exhibit A: The Birthday Problem (https://en.wikipedia.org/wiki/Birthday_problem)

I am speaking only for the two dice case. Which I am right. Yes more dice more problems.

[jnayls]

Addressing no one in particular, just voicing my opinions to be discussed. Feel free to redress me, or poke holes. This is why I posted in the first place: Feedback and discussion.

I am under the impression that burning a higher spell slot nets you a true gain when using a spell. Sure it doesn't scale terribly well, that's for sure. But the chances go up significantly the many ways (some I have not posted) that I have rewritten it, and that is by design. Its not fireball, sure, but the chance for crits and leaps can make it useful, and may even outperform in certain edge situations.

Fireball's 8d6 with a dex save aoe 20' radius vs. 3rd Level Chaos Bolt's 4d8 attack with possible crits, and a better chance of leaping is hardly complicated and not a terrible trade off.

Maybe I'm just of the camp that Fireball is not a spell that every blaster should feel required to learn. I would like my Wild Mage Sorc to be able to cast Chaos Bolt throughout his adventure without having to "trade up" as it fits better than other spells on the list. Purely for flavor at least.

The way the spell is written (and I am glad they wrote one as a base) I don't feel like they made it great or even good. I could just burn chromatic orb and when damage is rolled just pick my damage type with a d6 and two fewer types. And I've got the Chaos Bolt with better damage and yeah sure it doesn't leap, but there was only a 12.5% chance of that happening anyways.

I could always bump it up to a 2nd level spell if it is balance thing.

[Jerrykhor]

Addressing no one in particular, just voicing my opinions to be discussed. Feel free to redress me, or poke holes. This is why I posted in the first place: Feedback and discussion.

I am under the impression that burning a higher spell slot nets you a true gain when using a spell. Sure it doesn't scale terribly well, that's for sure. But the chances go up significantly the many ways (some I have not posted) that I have rewritten it, and that is by design. Its not fireball, sure, but the chance for crits and leaps can make it useful, and may even outperform in certain edge situations.

Fireball's 8d6 with a dex save aoe 20' radius vs. 3rd Level Chaos Bolt's 4d8 attack with possible crits, and a better chance of leaping is hardly complicated and not a terrible trade off.

Maybe I'm just of the camp that Fireball is not a spell that every blaster should feel required to learn. I would like my Wild Mage Sorc to be able to cast Chaos Bolt throughout his adventure without having to "trade up" as it fits better than other spells on the list. Purely for flavor at least.

The way the spell is written (and I am glad they wrote one as a base) I don't feel like they made it great or even good. I could just burn chromatic orb and when damage is rolled just pick my damage type with a d6 and two fewer types. And I've got the Chaos Bolt with better damage and yeah sure it doesn't leap, but there was only a 12.5% chance of that happening anyways.

I could always bump it up to a 2nd level spell if it is balance thing.
Poking holes is my specialty, I'm glad you asked :smallbiggrin:

I'm not sure how you can think that a 3rd level Chaos Bolt is 'not a terrible trade off' against Fireball. Because it is exactly just that.

Both spells have the same weakness, i.e the fewer the enemies, the lesser damage they deal. Chaos Bolt at worst, misses completely and does nothing. Fireball however, even if every target saved, they still take half damage.

Chaos Bolt also has a high chance of simply hitting one target and nothing else, while Fireball hits every target in the 20ft radius.

As written, Chaos Bolt is a completely garbage spell. Sure, you're not obliged to take Fireball, but in terms of blasting, it is one of the best, no doubt.

[Kryx]

Hole poking intended to be helpful/insightful:


I am under the impression that burning a higher spell slot nets you a true gain when using a spell. Sure it doesn't scale terribly well, that's for sure. But the chances go up significantly the many ways (some I have not posted) that I have rewritten it, and that is by design.
A 1st level spell cast at 3rd level is not intended to be as good as a 3rd level spell. If it is your intention to change this system then you have a large task before you.

Additionally current higher level spell usage scales linearly. You've made a spell that scales not only linearly, but also exponentially. Assuming your earlier numbers a 4th level spell has a 79.5% chance to bounce. The math would be multiples higher than any other 1st level spell scaled to 4th level.


Fireball's 8d6 with a dex save aoe 20' radius
Fireball is probably the worst spell to base balance off of. By my calculations it should be 6d6, not 8d6. The developers have stated that it is purposefully an outlier, along with lightning bolt. It should not be used as a balancing point unless you decide that all spells should be at that power level (I chose to make those outliers match other spells and more closely match the DMG recommendations)

Additionally comparing an AoE spell vs a primarily single target spell isn't such a good comparison. Something like Blight would be a better choice.


Maybe I'm just of the camp that Fireball is not a spell that every blaster should feel required to learn.
If Fireball is overepresented in your games then fix fireball as it's an outlier. Otherwise you'll need to fix every spell. Though to be fair I still see a fair amount of fireball and lightning bolt in my games even in their balanced forms - because they're still amazing spells.

===================

Not that you must use this version, but perhaps it was unclear what I use so I'll restate my houseruled version which by my calculations is balanced as a 1st level spell, scales based on the current scaling system, and works great in my game so far:

You hurl an undulating, warbling mass of chaotic energy at one creature in range. Make a ranged spell attack against the target. On a hit, the target takes 3d8 damage. Choose one of the d8s. The number it rolled determines the type of damage, as shown below.

If you roll an 8, the chaotic energy leaps from the target to a different creature of your choice within 30 feet of it. Make a new attack roll against the new target, and make a new damage roll, which could cause the chaotic energy to leap again. A creature can be targeted only once by this mass of chaotic energy.

At Higher Levels. When you cast this spell using a spell slot of 2nd level or higher, each target takes extra damage of the type rolled. The extra damage equals 1d8 for each slot level above 1st.

[jnayls]

Kryx
Hole poking intended to be helpful/insightful

Thanks.

I forgot about the blurb in the DMG (pages 283-284) about how to scale spells with damage. And I haven't been able to notice that they do like to break their own guidelines a lot it seems. (i.e. Fireball, Lightning Bolt, Cone of Cold, Destructive Wave, and even Cure Wounds is way off the scale that they set.) They even suggest that a missed shot should give you a 25% increase in damage compared to "half on a save" that is common.


You hurl an undulating, warbling mass of chaotic energy at one creature in range. Make a ranged spell attack against the target. On a hit, the target takes 3d8 damage. Choose one of the d8s. The number it rolled determines the type of damage, as shown below.

If you roll an 8, the chaotic energy leaps from the target to a different creature of your choice within 30 feet of it. Make a new attack roll against the new target, and make a new damage roll, which could cause the chaotic energy to leap again. A creature can be targeted only once by this mass of chaotic energy.

At Higher Levels. When you cast this spell using a spell slot of 2nd level or higher, each target takes extra damage of the type rolled. The extra damage equals 1d8 for each slot level above 1st.

The added spell slots do or do not add to chances of leaping? Cause I actually think I like the this works. Either way. But the scaling would change the same way as I've described if the former is used (not necessarily a bad thing). It's not clear either way, and i think spelling it out would be necessary for any gameplay.

[Kryx]

I forgot about the blurb in the DMG (pages 283-284) about how to scale spells with damage. And I haven't been able to notice that they do like to break their own guidelines a lot it seems. (i.e. Fireball, Lightning Bolt, Cone of Cold, Destructive Wave, and even Cure Wounds is way off the scale that they set.) They even suggest that a missed shot should give you a 25% increase in damage compared to "half on a save" that is common.
WotC does seem to sometimes ignore guidelines for spells and other areas. That's why I created my spreadsheet - to try to balance each spell so that all spells can be more competitive choices.

Cure Woulds by their metric should be 2d10 (11). It's currently 1d8 + spellcasting (4.5+3-5 = 7.5-9.5). But it has no chance to miss so I think that's why it is lower. I up it to 2d6 + spellcasting (7+3-5 = 10-15) because I perceive healing to mostly not be worth it in combat and this is intended to nudge that a bit by making healing slightly more powerful than damage per level. It also makes it more reliable.



The added spell slots do or do not add to chances of leaping? Cause I actually think I like the this works. Either way. But the scaling would change the same way as I've described if the former is used (not necessarily a bad thing). It's not clear either way, and i think spelling it out would be necessary for any gameplay.
In my version, like RAW, Chaos Bolt does not increase the chance of bouncing at higher levels - I need to clarify that in the wording. It is intended that the bounce chance always is a 33% chance which is what "at least one 8 from 3d8" is. So 33% of the time it does more damage than a 1st level spell and 66% of the time it does slightly less.

If it scaled based on level it would scale exponentially as well as linearly like I mention above.

[OJPhoenix]

Took me a while to calculate the probabilities, but this is what I've come up with (note that I do believe this is a gross simplification of way this probability is truly calculated but... I could find no better.)

6d8 (~92.3%)
5d8 (~79.5%)
4d8 (~59.0%)
3d8 (~34.3%)
2d8 (~12.5%)

Based on the formula as follows:

n = number of sides on dice
y = number of dice rolled
n * n-1 * n-2 ... / ny
8 * 7 * 6 * 5 * 4 * 3 / 86
8 * 7 * 6 * 5 * 4 / 85
8 * 7 * 6 * 5 / 84
8 * 7 * 6 / 83
8 * 7 / 82

And based on this and quick look through Kryx's spell doc, I've bumped the based to 3d8 with a minimum of the same. So that at 9th it is 7d8 (~98.1% chance of leaping on first strike). Seemingly I would probably never cast Higher than 5th myself, but it does scale well against hordes and the like at higher level slots.

Note: Any Statisticians out there feel free to educate me on this as once you get to 9d8 this method of calculating falls apart showing that there is a chance (however slim) of rolling non-repetitive, even though its (irl) impossible.

You're math makes perfect sense to me, so I'm guessing you forgot that the descending multiplication for the numerator is actually about how many possible different outcomes exist on each dice, and seeing as how the ninth dice has 0 possible different outcomes, you actually end up with this:

8 * 7 * 6 * 5 * 4* 3 * 2 * 1 * 0 / 89 = 0

leaving that 0 out gets you 0.0030040741 which i'm assuming is that slim chance of failure you were talking about, but I could be wrong

I must admit, it took me a moment to realise the formula you wrote out was the chance of failure, which needs to be subtracted from 1 to get the chance of success :smalltongue:

Also, I really like this edited version of Chaos Bolt, it makes perfect sense that you'd get more chaos when you load it up with more fuel :smallbiggrin:

[xroads]

Chaos Bolt
1st-level evocation
Casting Time: 1 action
Range: 120 feet
Components: V, S
Duration: Instantaneous
You hurl an undulating, warbling mass of chaotic energy at one creature in range. Make a ranged spell attack against the target. On a hit, the target takes 2d8 damage. Choose one of the d8s. The number it rolled determines the type of damage, as shown below.
d8 Damage Type
1 Acid
2 Cold
3 Fire
4 Force
5 Lightning
6 Poison
7 Psychic
8 Thunder
If you roll the same number on any two d8s, the chaotic energy leaps from the target to a different creature of your choice within 30 feet of it. Make a new attack roll against the new target, and make a new damage roll, which could cause the chaotic energy to leap again.
A creature can be targeted only once by this mass of chaotic energy.
At Higher Levels. When you cast this spell using a spell slot of 3rd level or higher, each target takes extra damage of the type rolled. The extra damage equals 1d8 at 3rd, 2d8 at 5th, and 3d8 at 7th.


I like this! Probably not as op as other evocation spells at the same level. But certainly seem like it could be a fun spell to cast, particularly at higher levels. I'd seriously consider it as a caster.

[Finger6842]

2d8 at first level is really strong, the 'any damage type' is very strong and keeps the spell relevant in later levels, and 60' per jump is too strong, the initial 120' range is already incredible, shortbow is only 80' and longbow is only 150'

and then the Sorcerer twins the bolt, action surge, twin again. Whole armies of orcs and kobolds will die, potentially hundreds of feet apart as long as there is one every 60'. Fireball is a single element and small radius so NPCs that are immune, resistant or just spread out will definitely render it ineffective.

I really think this is a spell designed to be powerful at low levels and kept to smack the masses or hurt the magic resistant stuff. It can be swapped out eventually.

[JackPhoenix]

2d8 at first level is really strong

It's certainly not: it's on average 9 hp. Burning Hands has 3d6 (average 10.5) in AoE, Magic Missile 3*1d4+1 (3*3.5 is also 10.5) with autohit and best damage type, Chromatic Orb is 3d8 (13.5) and you actually get to choose the energy type, Thunderwave gets 2d8 (9) in AoE and push on top...

But hey, it's better than Witch Bolt with 1d12 (6.5)...


the 'any damage type' is very strong and keeps the spell relevant in later levels

It's not "any damage type", it's "random damage type", you don't have much choice over what damage you actually get. You may end up rolling 2* fire against fire immune creature and waste spell slot. Chromatic Orb has less options, but the choice is entirely yours.


and 60' per jump is too strong, the initial 120' range is already incredible, shortbow is only 80' and longbow is only 150'

Shortbow is up to 320' and longbow up to 600' (though usually at disadvantage), doesn't cost you a spell slot, and can do more damage. 120' is good, but Fire Bolt or Eldritch Blast can do the same, as cantrips.


and then the Sorcerer twins the bolt, action surge, twin again. Whole armies of orcs and kobolds will die, potentially hundreds of feet apart as long as there is one every 60'.

He doesn't, because the spell can target more than one creature, and is thus ineligible for twin spell. And because it's leveled spell and not cantrip, he can't quicken it for two casts either, so he can get two castings with Action Surge, but that's it. And note that the original spell scales with d6, not d8, so the chance to bounce doesn't increase with higher slot.

Even if ALL that worked...you'd have to hit with every bounce AND roll doubles on it. I'm not gonna do the math, so I'll use Jnayl's calculations: to get more than 50% chance for bounce, you'd have to use level 5 spell slot (with the OP's modified spell). Let's see what other options are there for dealing with armies of kobolds and orcs at that spell level, shall we? Cloudkill has better damage (though with worst damage type in the game, neither kobolds or orcs are immune to poison), decent AoE, lasts a while and moves. It makes better dent in an army than Chaos Bolt, and potentionally forces them to disperse and get to worse tactical situation. We can Animate Object or Conjure Elemental to get something to beat the army for us... again, both are better options than Chaos Bolt (so is Animate Dead from 5th level slot, for that matter). We can cast Maelstrom for even greater AoE than Cloudkill and better damage type, that doesn't move but doesn't block line of sight either and has the same terrain control potential.

[Kryx]

2d8 at first level is really strong
No it isn't. It's one of the lowest damage spell in the game.

Chaos Bolt: 2d8 (9) with a 12.5% chance to bounce.

Some other similar 1st level spells:
Catapult: 3d8 (13.5)
Guiding bolt: 4d6 (14) and advantage on the next attack
Ice Knife: 1d10 (5.5) + 2d6 (7) = 12.5
Inflict wounds: 3d10 (16.5)
Magic Missile: 3d4 + 3 (10.5) with no miss

2d8 for a first level spell is a bad spell. Chaos Bolt is a bad spell by RAW.

As I mentioned above "I've changed it to 3d8, scaling by 1d8 at each higher level, and bouncing on a 8 from the original 3d8". That makes it a totally worthwhile spell.

[Finger6842]

No it isn't. It's one of the lowest damage spell in the game.

Chaos Bolt: 2d8 (9) with a 12.5% chance to bounce.

Some other similar 1st level spells:
Catapult: 3d8 (13.5)
Guiding bolt: 4d6 (14) and advantage on the next attack
Ice Knife: 1d10 (5.5) + 2d6 (7) = 12.5
Inflict wounds: 3d10 (16.5)
Magic Missile: 3d4 + 3 (10.5) with no miss

2d8 for a first level spell is a bad spell. Chaos Bolt is a bad spell by RAW.

As I mentioned above "I've changed it to 3d8, scaling by 1d8 at each higher level, and bouncing on a 8 from the original 3d8". That makes it a totally worthwhile spell.

Well, it is UA so testing it is the point. I thought lesser resisted damage types did less damage because they were more likely to succeed all the time. Clearly I'm wrong about that.

Let's scale it to 40d8 at 9th level, that way it'll be worth while as compared to meteor swarm too, totally awesome. Perhaps we can level the playing field completely, every spell is now aoe 3d8 adding 2d8 for every spell level, and does random damage types. It'll be easy to choose then. Oh, mêlée classes need a way to cast it since they have no other way to do that kind of damage.

Yea, I'm baiting you, I apologize. Seriously.