Would it be game-breaking to allow casting spells at a higher level to increase the save DC for that spell instead of increasing damage?

Question

There are a number of spells that include a section on "At Higher Levels." The actual effect varies from spell to spell; longer duration, more targets, summon better creatures, etc. I am specifically asking about spells that satisfy the following conditions:

• The spell deals damage and when cast at a higher level than their base it causes additional damage.
• The spell deals more than damage. There is also a rider/sub-effect as part of the spell.
• The spell involves a saving throw. This saving throw may or may not be related to the damage the spell does. It is directly related to the rider effect.

For instance, Ray of Sickness:

A ray of sickening greenish energy lashes out toward a creature within range. Make a ranged spell attack against the target. On a hit, the target takes 2d8 poison damage and must make a Constitution saving throw. On a failed save, it is also poisoned until the end of your next turn.

At Higher Levels. When you cast this spell using a spell slot of 2nd level or higher, the damage increases by 1d8 for each slot level above 1st.

Another example would be Thunderwave:

A wave of thunderous force sweeps out from you. Each creature in a 15-foot cube originating from you must make a Constitution saving throw. On a failed save, a creature takes 2d8 thunder damage and is pushed 10 feet away from you. On a successful save, the creature takes half as much damage and isn't pushed.

In addition, unsecured objects that are completely within the area of effect are automatically pushed 10 feet away from you by the spell's effect, and the spell emits a thunderous boom audible out to 300 feet.

At Higher Levels. When you cast this spell using a spell slot of 2nd level or higher, the damage increases by 1d8 for each slot level above 1st.

No matter what level the spell is cast at, the Con save DC is a constant based on the caster's spellcasting ability modifier and proficiency bonus.

My question is, would it be overpowered to say that instead of causing more damage, the caster can opt to make the rider effect harder to resist?

For the sake of discussion, consider the rule to be for every slot level above base it increases the DC by 1. And it's completely either/or per casting. A caster could cast once to grant a better DC, and a second time to grant more damage. But they cannot cast a spell two levels higher and boost the DC by one and an additional damage dice.

So a 20th level wizard could cast a 1st-level spell using a 9th-level slot would have a final DC of (barring magic items): 8 (base) + 6 (proficiency bonus) + 5 (20 Int bonus) + 8 (higher spell slot) for a max of DC 27. That seems pretty high, but consider that a Androsphinx, a CR17 monstrosity, has the following bonuses to saving throws...

Dex +6, Con +11, Int +9, Wis +10

Most saves are still within the realm of possibility on a good roll.

So in the case of Ray of Sickness, instead of doing on average 45hp damage and a slim chance of poisoning, the spell would average 9hp damage, but a very good chance of poisoning. Thunderwave would generally be saved at higher levels so would average 5hp damage and that's it. But with the increased DC, it would do about 9hp damage and most likely shove.

Would giving this option to casters be a fair house rule?

---

Since it was brought up in the comments and the top answer; here is the why:

Because I am but a squishy wizard. And a number of times I have found myself cornered by a brute. Now I know that I can upcast my spells to do more damage, but I also know that it would never be enough to take the brute down. HOWEVER, I know that some of my spells do more than damage, but something that will help me escape. Ray of Sickness would give them disadvantage when I run and they get an Opportunity Attack. Shocking Grasp would also stop that Opportunity Attack. Or Thunderwave would push them 10 feet away to give me an even better chance to escape. So I would rather focus less on damage and more on secondary effects that would keep me alive.

Answer 1

By the Numbers

Let's compare the Ray of Sickness spell with the Weird spell.

Ray of Sickness says:

Make a ranged spell attack against the target. On a hit, the target takes 2d8 poison damage and must make a Constitution saving throw. On a failed save, it is also poisoned until the end of your next turn.

At Higher Levels. When you cast this spell using a spell slot of 2nd level or higher, the damage increases by 1d8 for each slot level above 1st.

So, a 20th level caster can spend a 9th level spell slot to deal 10d8 damage and possibly make someone Poisoned for a round (save DC 19). With the house-rule, the Wizard could instead deal 2d8 poison damage and possibly make someone Poisoned for a round with a +8 bonus to the save DC (total DC 27).

Weird says:

Each creature in a 30-foot-radius sphere centered on a point of your choice within range must make a Wisdom saving throw. On a failed save, a creature becomes frightened for the duration. The illusion calls on the creature's deepest fears, manifesting its worst nightmares as an implacable threat. At the end of each of the frightened creature's turns, it must succeed on a Wisdom saving throw or take 4d10 psychic damage. On a successful save, the spell ends for that creature.

Its duration is "Concentration, up to 1 minute" (10 rounds).

That 20th-level wizard can spend a 9th level spell slot to deal between 0 and 40d10 damage and leave the target(s) Frightened for up to 10 rounds (save DC 19).

Conveniently, both the Poisoned and Frightened condition grant disadvantage on attack rolls and ability checks. While Frightened also precludes the target from voluntarily moving closer to the source of the condition, it also requires that the source be within line of sight to continue working. For my purposes, those effects are close enough to continue going.

Now, to the opponent: the aforementioned Androsphinx, with saves of:

Dex +6, Con +11, Int +9, Wis +10

and, let's ignore any other abilities they might have, choosing instead to take its saves as roughly-normal-ish for a CR 17 creature.

If it's the only baddie, our wizard now has 3 choices for their 9th level spell slot:

• 2d8 (9) damage + (+11 save vs. DC 27 means they need to roll a 16 =) 75% chance of Poisoning for a round, 25% chance of saving.
• 10d8 (45) damage + (+11 save vs. DC 19 means they need to roll an 8 =) 35% chance of Poisoning for a round, 65% chance of saving.
• (+10 save vs. DC 19 means they need to roll a 9 =) 40% chance of Frightening for the first round and 40% chance of dealing 4d10 (22) damage on that first round, with the chances continuing for the next 5 rounds; using 5E's assumption that effects end after one round, that means we can pretend the creature is Frightened for a round and takes damage once. (60% chance of saving.)

All of that taken together means that the wizard can trade either 13 or 36 points of damage for a ~50% reduction in the chance that the creature's attacks aren't made with disadvantage in the first round after casting the spell compared to casting a 9th level spell.

By the numbers, that looks like it's roughly a fair trade.

At the Table

However, a numerical analysis cannot account for everything. There are three other factors that affect the answer.

Spell Selection

Allowing casters to increase the DC of statuses may affect spell selection, allowing casters to slot in more utility-ish spells at higher level reasoning that they can get by with lower-level spells that might not quite pack the same direct-damage punch but will still apply statuses that will help the party handle the monster and still have a Wish or Time Stop prepped. This feeds directly into...

Versatility

Allowing casters to upcast spells in 2 ways functionally increases the number of spells they have at their disposal. Versatility itself is powerful (in fact, it's the primary driver of the Tiers of Classes). While 5E may be better than its predecessors in minimizing the differences between class tiers, casters are usually on top largely due to their versatility. Additionally, versatility leads into...

Table Time

Having more options means it takes longer for a player to take their turn. Allowing spells to be upcast to do more damage or to increase the DC of non-damaging effects increases the options the player has. Therefore, on average, turns will take longer and combat will be slower.

But, Why?

Setting aside, momentarily, all questions of balance and fairness, house rules ought have a reason behind them. Ideally, that reason would speak either to a weakness in the extant rules or to some unique bit of world building in the particular world in which the campaign takes place. House rules require everyone involved to remember something that isn't in the books (and, in my experience, often isn't written down anywhere at all), adding additional cognitive overhead; that overhead should pay for something concrete.

So, is there a specific pain point that this house rule is seeking to fix or a specific bit of world-building this is seeking to reinforce? Or, does this just seem like a nifty idea?

Conclusion

In conclusion, this option probably won't break the game. It'll tend to make casters a little more powerful and to slow combat a bit, but it probably won't make a huge impact, most of the time. And, it's that last bit that makes this GM extremely hesitant to endorse it: I just don't think it'll add anything to the game beyond a couple moments of "hmm; should I bump up damage or the chance that the status effect takes effect?".

In many ways, it reminds me of the 3.5 version of Power Attack:

On your action, before making attack rolls for a round, you may choose to subtract a number from all melee attack rolls and add the same number to all melee damage rolls. This number may not exceed your base attack bonus. ... If you attack with a two-handed weapon, ... instead add twice the number subtracted from your attack rolls.

As both a player and GM, I've spent dozens of hours watching people (myself included) debating whether to take a -2 or -3, especially when wielding a 2-handed weapon. It was painful, and I think and fear that's what this house rule is strongly leaning towards.