Would increasing the persistent damage of Witch Bolt be unbalanced?

Question

Some People (myself included), feel that Witch Bolt is sub par.

I was thinking of changing it so that increasing the spell level also increases the persistent damage, as well as the initial damage.

This would change the last text to be:

At Higher Levels: The initial and persistent damages both increase by 1d12 for each slot level above 1st.

Would this cause any balance issues?

Answer 1

It is not overpowered, but still a bad idea

As I have shown in another answer, it is not unbalanced.

However, it would just add to the current problems.
The spell is a classical trap, looks good on paper, very weak in practice, but on higher levels at least it looked as weak as it was.
With your change it keeps being subpar in disguise.

Answer 2

Yes, this is overpowered

Witchbolt does not require an attack roll for the ongoing damage. Increasing by 1d12 for each level would allow a 5th level caster to deal 30d12 damage, 27 of which is without an attack roll or saving throw, with a single 3rd level slot. This is very strong. Remember that Warlocks will always be upcasting it so allowing increased ongoing damage significantly increases damage output.

As a compromise I would suggest increasing ongoing damage for every second or third slot above first. This is still a decent damage bump but not quite so game breaking.

Math breakdown

Assume character of 5th level, CHA 20(+5) and Prof +3

• Eldritch Blast w/ Agonizing Blast + Hex:

  • 2 Blasts, +8 to Hit, 1d10+5+1d6 per hit.
  • Average Damage vs AC: \begin{array}{|c|c|c|c|c|} \hline AC& 12 & 15 & 18 & 21 \\ \hline DPR & 22.525 & 18.55 & 14.575 & 10.6 \\ \hline DPM & 225.25 & 185.5 & 145.75 & 106 \\ \hline \end{array}

• Witch Bolt (RAW)

  • 1 Bolt per round, +8 on first hit then automatic, 3d12 then 1d12

  • Average Damage vs AC: \begin{array}{|c|c|c|c|c|} \hline AC& 12 & 15 & 18 & 21 \\ \hline First Round & 16.575 & 13.65 & 10.725 & 7.8 \\ \hline DPM & 66.3 & 54.6 & 42.9 & 31.2 \\ \hline DPM* & 78 & 78 & 78 & 78 \\ \hline \end{array}

• Witch Bolt (Scaling every level)

  • 1 Bolt per round, +8 on first hit then automatic, 3d12 then 3d12

  • Average Damage vs AC: \begin{array}{|c|c|c|c|c|} \hline AC& 12 & 15 & 18 & 21 \\ \hline First Round & 16.575 & 13.65 & 10.725 & 7.8 \\ \hline DPM & 165.75 & 136.5 & 107.25 & 78 \\ \hline DPM* & 195 & 195 & 195 & 195\\ \hline \end{array}

• Witch Bolt (Scaling every second level)

  • 1 Bolt per round, +8 on first hit then automatic, 3d12 then 2d12

  • Average Damage vs AC: \begin{array}{|c|c|c|c|c|} \hline AC& 12 & 15 & 18 & 21 \\ \hline First Round & 16.575 & 13.65 & 10.725 & 7.8 \\ \hline DPM & 116.025 & 95.55 & 75.075 & 54.6 \\ \hline DPM* & 136.5 & 136.5 & 136.5 & 136.5 \\ \hline \end{array}

Another comparison at 9th Level. Now casting in a 5th level spell slot

• Witch Bolt (Scaling every level)

  • 1 Bolt per round, +8 on first hit then automatic, 5d12 then 5d12

  • Average Damage vs AC: \begin{array}{|c|c|c|c|c|} \hline AC& 12 & 15 & 18 & 21 \\ \hline First Round & 29.25 & 24.375 & 19.5 & 14.625 \\ \hline DPM & 292.5 & 243.75 & 195 & 146.25 \\ \hline DPM* & 325 & 325 & 325 & 325 \\ \hline \end{array}

• Witch Bolt (Scaling every second level)

  • 1 Bolt per round, +8 on first hit then automatic, 5d12 then 3d12

  • Average Damage vs AC: \begin{array}{|c|c|c|c|c|} \hline AC& 12 & 15 & 18 & 21 \\ \hline First Round & 29.25 & 24.375 & 19.5 & 14.625 \\ \hline DPM & 187.2 & 156 & 124.8 & 93.6 \\ \hline DPM* & 208 & 208 & 208 & 208 \\ \hline \end{array}

* Assuming Witch Bolt hits for DPM calculations. Allowing for miss chance significantly skews data. It is very difficult to accurately show the average damage for witch bolt as is in an 'all or nothing' type spell. Showing actually average damage would not be representative as it either does full damage or none.

It is not representative to take the actually average values from the middle rows as this assumes you do nothing with your turn after you miss with Witch Bolt. Instead you would either cast is again or revert to the Eldritch Blast DPR for the rest of the minute.

Summary

As can be seen from the tables above, this change doesn't significantly alter damage output against low or medium AC targets. The balance problem is the massive bump to damage against high AC targets. In the last example the problem only gets worse as the spell slots increase.

In general if a homebrew changes a relatively poor spell into a must have damage spell it is likely overpowered. I would suggest play testing the scaling with every second level option. While a decent bump over RAW it is really just bringing a weak spell in line with stronger alternatives, which was the intent of this homebrew.

Answer 3

Yes, overpowered

The ongoing damage of Witch Bolt does not require an attack roll. Increasing its damage would essentially allow the caster to "always hit" and make it a superior single-target action to all other spells, particularly if the caster upcast it at a higher level.

Of course, a target could move out of the spell's range. But they might not because:

• They are engaged in melee and would provoke attacks of opportunity
• They are prone
• They have their movement reduced by some effect
• The caster has the Spell Sniper feat Confirmed by Mike Mearls

You point out that Witch Bolt is underpowered. This is true in cases where chance to hit is not a factor, but your solution would make it overpowered against harder to hit targets (high AC, invisible, hidden, etc.):

Below is a spreadsheet that calculates average damage per action of various spells (and a GWM barbarian for comparison) after the first round because the math of simulating the first round in addition to the following rounds gets a bit more complicated.

Of note:

• It is comparable to Magic Missile, another "auto hit" level 1 spell but with the potential to do much more damage.
• Eldritch Blast with Hex does better most of the time if you have Agonizing Blast and hex but not all characters will have access to either/both, and boosting Witch Bolt in this way renders build choices that grant access to these as irrelevant or sub optimal.