I'm running 5e on a grid with minis. The one thing that I hate is when sorcerers or wizards cast a fireball that perfectly hits enemies in the AOE. The player counts the squares on the grid to determine exactly where the fireball can hit and he knows the perfect way for the fireball AOE to take effect. It will hit enemies but amazingly the blast stops just in front of an ally's face. A player who was attacked by 2 melee enemies in my game cast a fireball behind the enemies so that the enemies were hit in the explosion but he wasn't. In my opinion, this lacks verisimilitude.

I am looking for techniques or playtested house rules for adding in this kind of verisimilitude with area of effect spells.

I was considering making a house rule that wizards who cast area of effect spells like fireball need to make a perception check if they are centering their fireball onto the middle of an open space to accurately place a fireball. If they center the fireball on a creature, then they do not need to do this. But I'm not sure how this would affect the game, and I'm looking for an expert solution.

  • 6
    \$\begingroup\$ Mod warning: We expect answers to follow Good Subjective, Bad Subjective and not be pure opinion or speculation. Back It Up! \$\endgroup\$
    – mxyzplk
    Commented May 5, 2016 at 11:58
  • 9
    \$\begingroup\$ Is your concern only with large AoE spells or do you plan to implement something similar for other situations like martial combatants as well? Will arrows possibly drift to hit different squares than intended? What about magical arrows that might include some sort of AoE effect? Could a melee combatant accidentally hit the person next to his/her target? \$\endgroup\$
    – Ellesedil
    Commented May 6, 2016 at 21:55
  • 4
    \$\begingroup\$ My issue is primarily with aoe spells. I have no problem if an aoe spell is centered on a target but when it's centered onto the middle of blank square that just happens to be in exactly the spot to optimize it's blast radius and hit all enemies perfectly. If for example i had told the players you have 1 second to place the center of their aoe effect I know it wouldn't be perfect. \$\endgroup\$
    – user28536
    Commented May 9, 2016 at 5:39
  • 3
    \$\begingroup\$ Please explain how you consider all other spells & special 'magical' abilities (and just magic in general) to have verisimilitude where AoE spells do not. In other words, what makes it believable to you that a wizard could cast a spell like Fireball in the midst of combat, but not believable that they can cast it at a location of their choosing? Both being able to cast it and selecting a location are inherent in the spell description. \$\endgroup\$
    – TylerH
    Commented Dec 16, 2020 at 21:23

17 Answers 17


Boy, so many people lining up to tell you "don't do it that way it's badwrongfun!" I'll offer a differing perspective, which is yes, absolutely, use a house rule to this effect. It has the desired effect of adding verisimilitude without "nerfing" or "ruining" anything. I shall offer up real play experience and not pure opinion to demonstrate this.

I used this exact kind of house rule during all of my AD&D 2e days (a decade) and from time to time in 3e/Pathfinder days for the same reason; adding some verisimilitude to combat - far from negating the need for tactics, it instead makes you have slightly better tactics because you aren't relying on things being exact (like artillery and infantry in the real world). Planning for the possibility of friendly fire makes for incorporation of real world techniques which is always desirable to me (learning about real world weapons, tactics, history, science, etc. is one of the best benefits of RPGs that many people seem to want to stomp out nowadays). Also, adding a little bit of randomness to magic makes it not so overwhelmingly better than the martial options - some of the martial/magic power differential comes from "I have to roll all the time" vs "I just do it", so by making your mages roll to do things you equalize the playing field a bit.

House Rule Option 1: Grenade-Like Missiles

But to do this you need something that best as I can tell is missing in 5e, which is what previous editions call "grenade-like missile," "scatter diagram," or the "throw splash weapon." In 1e AD&D it was under Grenade-Like Missiles in the DMG (p.64) - you'd roll d8 for direction and d4 (short range) d6 (mid range) d8 (long range) for how far it landed from the target.

In its most recent 3.5e incarnation, it's written like this:

To attack with a splash weapon, make a ranged touch attack against the target. Thrown weapons require no weapon proficiency, so you don't take the -4 nonproficiency penalty. A hit deals direct hit damage to the target, and splash damage to all creatures within 5 feet of the target.

You can instead target a specific grid intersection. Treat this as a ranged attack against AC 5. However, if you target a grid intersection, creatures in all adjacent squares are dealt the splash damage, and the direct hit damage is not dealt to any creature. (You can't target a grid intersection occupied by a creature, such as a Large or larger creature; in this case, you're aiming at the creature.)

If you miss the target (whether aiming at a creature or a grid intersection), roll 1d8. This determines the misdirection of the throw, with 1 being straight back at you and 2 through 8 counting clockwise around the grid intersection or target creature. Then, count a number of squares in the indicated direction equal to the range increment of the throw. So, if you miss on a throw out to two range increments and roll a 1 to determine the misdirection of the throw, the splash weapon lands on the intersection that is 2 squares away from the target in the direction toward you. See the accompanying diagram.

After you determine where the weapon landed, it deals splash damage to all creatures in adjacent squares.

In other words, roll to hit the target place, and use d8 to scatter direction on a miss.

So, super helpful for those thrown mundane firebombs, but you can also use it for area effect spells that don't usually require a hit roll - you just require that touch attack on their target or target square. (You have to interpret what "range increment" means for the spell, I usually just did 1 square if it was dropping in short spell range, 2 squares medium, 3 squares long). You can use a real ranged hit roll or sub in something like an Arcana check at your discretion. (Sometimes in 3e I've used Spellcraft instead of a to-hit).

A fireball is most likely still going to hit that main guy you throw it at. It is less likely to perfectly get every one of a group of enemies, and it's also more dangerous to try to crisp everyone including the enemies in direct melee with your party members.

Then you get to apply other bonuses/penalties like you would for ranged attacks. I'm running a Pathfinder pirate game where PCs are often trying to heave fireballs and lightning boats at enemy ships while going full speed in a ship heaving up and down on the waves through a rainstorm. Those additional to-hit penalties make it really exciting, especially if some of their crew has already flown over onto the enemy ship! Similarly, if players play smart and set out ranging stakes and practice, then when the hordes of wild elves attack their encampment they could get advantage on the placement roll.

House Rule Option 2: Proximity Saves

One of my current Pathfinder GMs handles this same problem just by making all party members in close combat with anyone getting AoEd make a save for half/no damage, for the same reason. This works OK but I don't like it as much, mainly because it takes all the control away from the wizard. "Having to roll to hit" is not onerous and expected of other classes, but just saying that they can't really try to place it better and you just have to save if you're around is a bit odd and tends to just make AoEs one radius bigger. "How come an enemy in contact doesn't have to save too?" Although rogues and monks and such enjoy being able to use evasion with impunity in these situations.

Why This Isn't A Huge Mistake

The theoretical problems levied against house ruling this by other posters are invalid.

  1. "This takes control away from the player!" Only to the degree that control is taken away from any character (especially the fighty types) from having to roll a die to determine success. That's arbitrary, and other d20 derived games do have spellcasters roll for success and it doesn't "ruin" them.

  2. "This obviates the point of using a grid with its numerological perfection." No it doesn't, again, any more than fire bombs or ballista or anything else having to roll to place on a grid does. You don't have to go Theater of the Mind to get a little bit of imprecision. You could, but that's not a necessary solution.

  3. "This removes the need for tactics!" Really? Removing randomness removes tactics? Counterexample: the real world. Rather than removing tactics this instead ensures more realistic tactics.

  4. All kinds of tactical tips for your monsters to not bunch up and not get all fireballed. All fine, do those too, but does not bear on the validity of this house rule. Encounter design is an entirely different topic.

  5. "It's not RAW." No one cares; your game is your rules. See How do you help players not focus on the rules?

  6. "This must mean you are a meanie and are against your players." I assume you'll use the same rule when monsters fireball, so that's not really valid, and it assumes bad faith in that you don't really just want more realism like you say but you want to "stick it to mages." Plus, does it mean you hate fighters because you make them roll for stuff?

From my experience, all this house rule does (both variants) is make a) casters not drop AoEs sometimes in risky situations, and b) make martials have to be a little smarter because they can't rely on pinpoint spellcaster artillery precision. It changed the game a little, but not in any way any of the participants thought was "unbalanced." Instead, it ended up more realistic, which is one of my core gaming principles (YMMV). I haven't used these in 5e, but I am familiar enough with 5e to say I don't see any way in that this would have a different effect (as to be expected, as 5e is closest to a 2.5e in practice).

  • \$\begingroup\$ Comments are not for extended discussion; this conversation has been moved to chat. \$\endgroup\$
    – mxyzplk
    Commented May 5, 2016 at 11:55
  • 3
    \$\begingroup\$ I'm curious how you address class features like the Evocation Wizard's Sculpt Spells feature with this. Is the assumption just that, they spent the class feature to avoid having to be overly precise so this rule mostly doesn't apply for them? \$\endgroup\$ Commented May 8, 2020 at 12:24
  • 4
    \$\begingroup\$ I would add to this that if you intend to go this route, be damn sure you don't violate the same rule. If my DM says I can't pinpoint an AOE spell accurately because I essentially have to "eyeball" it, I would be pretty upset if my group got hit by a fireball perfectly with no enemy collateral. Make sure the rules you add are enforced on both sides! \$\endgroup\$
    – Carson
    Commented Dec 17, 2020 at 20:03
  • 8
    \$\begingroup\$ This is really good! Well, until you get to the "counterpoints". That's less good, and also not really backed up by anything except snark directed at other users. \$\endgroup\$ Commented Mar 8, 2021 at 19:37
  • 4
    \$\begingroup\$ @RevanantBacon I agree. I also don't find the argument that martial classes have to roll as convincing as the rest of the answer (which is overall great). There are already two rolls to determine the level of success of an AOE spell (saving throws and damage rolls), so this is a third roll to determine the degree of success. A more apt comparison would be requiring a martial character to roll to hit, roll for damage, and then having an enemy make a saving throw to possibly reduce damage. \$\endgroup\$ Commented Oct 15, 2021 at 16:56

Well, I don't think I need to tell you that it's within RAW, per the spell description of Fireball [emphasis mine]:

A bright streak flashes from your pointing finger to a point you choose within range then blossoms with a low roar into an explosion of flame. Each creature in a 20-foot radius must make a Dexterity saving throw. A target takes 8d6 fire damage on a failed save, or half as much damage on a successful one. The fire spreads around corners. It ignites flammable objects in the area that aren’t being worn or carried. At Higher Levels. When you cast this spell using a spell slot of 4th level of higher, the damage increases by 1d6 for each slot level above 3rd. [Pg 241 PHB]

I'd suggest upping your tactics game as opposed to taking control away from a player. The player is getting satisfaction out of playing tactically - instead of taking that satisfaction away from them, have your enemies play smarter.

  • Have a larger creature harass the wizard, who will be too concerned about their own safety to try to support the other players with a fireball.
  • Have enemies fan out once they notice a wizard to avoid the AoE, or have them try to draw the players into a trap, a narrow hallway. Bring some ranged enemies to give yourself some options to challenge the players. Bring enemies that climb, leap, fly, or otherwise challenge formations of the players.
  • Use terrain to block line-of-sight for the wizard.
  • Use enemies with counterspell, or other creative ways to shut down the wizard.
  • Use enemies with some kind of Dex save advantage, like the Shield Master feat that lets them negate damage on a successful save. Hell, you could throw a fire elemental at them.

Generally, if a player is coasting through challenges, find ways to make it harder and more interesting, without making the player feel like their abilities are useless.

  • 26
    \$\begingroup\$ I'd only add try not letting your spell-slingers be able to aim at a point 20 ft beyond the villains because of walls, trees, or other obstacles. If they can't see the target, they can't hit it with a spell. \$\endgroup\$ Commented May 1, 2016 at 5:59
  • 12
    \$\begingroup\$ Also, there shouldn't be a clear divide where enemies are on that side and allies on this side. If the enemies use tactics to mix into the party during combat (in Pathfinder, these would be things like Acrobatics and 5-ft steps), the entire party will either have to work harder to maintain their battle lines so that the wizard can AoE the whole enemy group, or they'll have to start taking fireballs that either only hit one or two bad guys or hit allies as well as enemies. \$\endgroup\$ Commented May 1, 2016 at 15:02
  • 5
    \$\begingroup\$ The issue isn't strategy. I'm aware of what I can do to make it challenging (thanks for more ideas btw!) It really is just this 1 situation. \$\endgroup\$
    – user28536
    Commented May 2, 2016 at 1:47
  • 1
    \$\begingroup\$ Good answer, and useful rules citing! I'm not sure it 100% addresses the original point, though. OP seemed ok with having characters able to place fireball exactly where they wanted it, but had a problem with their implied ability to know which space in a featureless field was exactly 30 feet away from enemies/allies. Unless I misunderstood them. \$\endgroup\$ Commented Oct 15, 2021 at 17:04

This meta-game accuracy is a purposeful feature of using the optional grid rules — that kind of tactical detail is the whole point of using a grid.

An obvious alternative that eliminates miniatures-based player precision is to not use the optional grid rules. There's some discussion of imprecise AoE handling on DMG page 249 (in short “make a call, consult this table if you like”), and our question D&D 5e and "Theatre of the Mind" in combat goes into more detail about how to handle gridless combat in general.

If gridded combat is strongly your preference though, embrace it and its accuracy. It's a feature of grid-based combat that your players are — quite naturally — using as intended. Instead of saying it's unrealistic, ask yourself “Huh, how do spellcasters place their AoEs so accurately?” and put some imagination into how that could be believable. My own reaction is to compare them to a real profession that requires intense precision at a distance, like high-level professional athletes (baseball pitchers are incredible in the precision they achieve, for just one). When I consider what highly-trained reflexes can accomplish in our own world, precise placement of fireballs stops threatening my sense of verisimilitude.

  • 4
    \$\begingroup\$ Yes! Love this answer. My initial instinct was not to nerf the character's placement, but spend more time on them describing how their cool character is performing such an amazing feat with such accuracy. \$\endgroup\$
    – webnesto
    Commented Apr 2, 2021 at 21:40

In the earliest rules, Chainmail with its Fantasy Supplement (Gygax & Perren, 1971), the fireball referenced the catapult rules for its mechanic. That includes the following optional rule (2E, p. 10):

Fire Optional: Roll two different colored dice. One color is for an over-shoot and the other is for an under-shoot. To decide which number to use you take the highest of the two. Miss is in inches, shown by dice spots. If they tie then the rock land [sic] at the specified range. This method is simple but effective.

Now, there's actually a second typo here; where it says "highest" it really should have said "lowest". A mathematically equivalent way of putting that would be: roll 2d6; for every point below 7, the shot lands 1" short, and for every point above 7 the shot lands 1" long. This gives a quasi-bell-shaped distribution of possible impacts around the target space. Compare to the OD&D rule for bombing scatter (Vol-3, p. 27-28), where the math is correctly expressed. The rule was maintained in some form up through Holmes Basic D&D (1979), where it appears as the suggested mechanic for Giants' rock-throwing ability.

Alternatively, there is an historical document believed to be a pre-publication draft of Original D&D titled "Beyond This Point Be Dragons" which features this rule for the spell:

A Fireball is thrown by the Magic-User, and the accuracy will vary with the distance of the intended target. Targets within 50 feet can be hit with 99% accuracy, at 55 feet 95% accuracy, and the accuracy decreases at 5% per every 10 feet.

So: I sympathize with the OP's intuition about the fireball (catapult) ability, and the writers also felt likewise, even as early as 1971, where the rules above were attached to the spell in question. I do use the Chainmail rule in my D&D games today for both the fireball and lightning bolt spells (which itself referenced the the old cannon rules in its initial appearance). It's a simple resolution, it gives a little bit of added drama/tension to the spell being cast, and my players have seemed pretty happy with it.

More on the fireball legacy: Spells Through the Ages - Fireball.

  • 5
    \$\begingroup\$ This is a fine way to deal with shooting into melee without going into drama, arguing about playstyle, or arguing about any number of details. It adds a nice touch of uncertainty into "shooting into melee" which is sort of what dice rolling was all about in the first place. \$\endgroup\$ Commented May 1, 2016 at 21:31
  • \$\begingroup\$ This is a good idea, though I think there's a more modern example in 5E to borrow from: the normal/max range of ranged weapons; simply shorten the range of Fireball and make that new distance the 'normal' range, and then add the original distance back as a "long" range, so you're in effect aiming the fireball at disadvantage. Because it's an AoE spell, it'll cast just fine, but if you aim beyond the 'normal' range, roll 1d4 (for example) for aiming accuracy and on a miss, it lands centered 5 feet off the intended center in a cardinal direction based on the 1d4 roll result. \$\endgroup\$
    – TylerH
    Commented Mar 8, 2021 at 18:53
  • \$\begingroup\$ But expect to receive arguments about it being unfair, and questions about whether all spells (or at least, all AoE spells) now having this requirement for consistency. \$\endgroup\$
    – TylerH
    Commented Mar 8, 2021 at 18:53

Prepare for a long answer...

This is, ultimately, a sign of a flaw in your enemy tactics, not in the player's use of Fireball. Wizards are scholars with years of study and practice under their belt. Sorcerers literally have magic in their blood, a part of their very being that they have grown up with and come to know as well as any part of their body. So not only are they doing what they should do as players, and playing well, but what they are doing makes sense for their characters. Neither a Wizard nor a Sorcerer gets to the point where they are throwing Fireball spells around without an understanding of what they are capable of. The only times they should have to roll for their aim are when they wouldn't be able to tell where they are aiming.

Creating arbitrary restrictions on the players interferes with their fun. It is important to remember that players having fun is a primary goal of the game, and a big part of the DM's job is ensuring everyone has a good time. If you feel like it's "BS", evaluate why you feel cheated. Is it just that you feel the players aren't getting their money's worth, metaphorically speaking? Or, do is it because on some level you want to "beat" the party? If the latter, I would say you are taking the DM's role as antagonist too seriously. Your job as DM is not to win against the players. Share in their triumphs and cringe with them at their losses, even if you don't let them see, because your job is ultimately, whether they win or lose, just to give them a good story to tell.

Which, to answer your question, brings us back to the former case: Are you making things too easy?

Be aware of the intelligence of your NPCs. They don't need to just run up and attack. This isn't a videogame with limited AI. Think about what they know, and how they'd use it. Can they tell who's a caster? Do they know what casters can do? Imagine the look on the players' faces when, at the first Magic Missile, one of their adversaries points and bellows "MAGE!", and others push past the front lines to descend upon them, or scatter and take cover, switching to ranged weapons.

Even more, though, think about encounter design. Battle begins long before the first blade is drawn. Think about how smart the enemies they could come into contact with before they ever see them, and what they might do to prepare. Will they all be grouped together at the start? If so, ask why. If not, what are they doing? who's away from the group. Maybe it's an ambush and ranged units have already taken sniping positions. Maybe they're hunting the party and have scouts on point, with a second wave of fresh fighters bringing up the rear. Perhaps they've got a caster or casters of their own. If so, you'll need to think about their tactics. Do they focus on counterspelling? Supporting the vanguard? Picking off stragglers? Do they know offensive, support, or "control" magic?

More than that, don't leave loot until after battle. While random loot is fine, it should be distributed logically. If any of the enemies can use a magic item, they probably will, barring a good reason not to. In addition, a few hand-picked treasures can go a long way. Maybe the bandit leader has taken a suit of armour of fire resistance as a trophy. Perhaps a crafty, roguish assassin is using a Cape of the Mountebank or Dust of Disappearance to get the drop on their targets, and the party caster in the back makes a nice unguarded target. Or, even without magic items, the introduction of some dragonborn or gensai (or something like a goliath with a truly massive bow), can shake up a combat and force different tactics.

tl;dr Versoin! - Rather than nerfing the players' abilities, put them in situations where "just throw a fireball at it" isn't the most effective tactic. You'll feel good because they aren't making short work of your encounters all the time. They'll feel good because it will give them exciting, interesting, and memorable encounters to talk about. Everybody wins.

Disclaimer: Be open with your players when you do this. Tell them that they've shown they can handle your encounters, and you want to give them more of a challenge. Then, every once in a while, throw them a bone and toss in an encounter that they can just throw a fireball at (or the equivalent). That way, they won't feel like you're just trying to screw over their favourite tactics, even if that's exactly what you're doing. Also, don't be afraid to pull your punches and let an enemy or two prove woefully inept if it turns out you may have overcompensated a bit, or have some reinforcements arrive if it seems too easy.

  • 6
    \$\begingroup\$ Fighters are athletes with years of study and practice under their belt. Barbarians literally have rage in their blood, a part of their very being that they have grown up with and come to know as well as any part of their body. So not only are they doing what they should do as players, but what they are doing makes sense for their characters. Neither a Fighter nor a Barbarian gets to the point where they are Dueling Dragons without an understanding of what they are capable of. The only times they should have to roll for their hits are when they wouldn't be able to tell where theur target is. \$\endgroup\$
    – Yakk
    Commented May 5, 2016 at 14:02
  • 13
    \$\begingroup\$ @Yakk - There's a big difference between aiming at a fixed point in space and knowing how big the boom is, and trying to hit the small vulnerable spot on a moving target. Besides which, BAB scales with level, while AC does not, so Barbarians and Fighters do in fact get to the point where what once was difficult becomes a trivial task. In other words, you're comparing apples and cabbages. \$\endgroup\$
    – user28753
    Commented May 5, 2016 at 15:33

A fireball is in D&D what artillery is on a battlefield. Just like intelligent soldiers know how to deal with artillery, intelligent monsters know how to deal with magic users.

Continuing the military analogy there are basically five techniques:

  • Dispersal, so that AoE will strike a limited number of combatants
  • Cover, using natural or manufactured obstacles to break line of effect
  • Camouflage, hide so you can't be targeted
  • Closure, get in among the mage's allies so he has to either not use his AoE or bring down friendly fire. 5e opportunity attack rules make it easy to slide around enemies.
  • Counter-battery fire, the Mage is the most dangerous enemy? Fine, take him out first.

When my players try to do something tricky using a spell, sometimes I ask them to make Spellcraft checks. "You're trying to throw a fireball to hit this one guy but not the guy standing next to him. I need a Spellcraft check, and if you fail you hit them both. Does that work for you?" My players seem to take that pretty well. (This does tend to result in them putting lots of points in Spellcraft but that's usually a good idea anyway.)

I'm playing in Pathfinder, but you could do the same thing with an Arcana check or something similar. I guess Perception, as you describe, works too.

Another thing you might try would be to say: "If you're on the very edge of a blast spell such as fireball, you get advantage on your saving throw." This represents the bonus because the fireball doesn't completely fill your square, and the bonus because it's running out of fire so far from its center.

But more generally, if spellcasters' spells are bothering you, the answer is to have more combats per day. Spellcasters are very strong when they can rest after each combat, and they're much weaker when they have to conserve spells.

  • 3
    \$\begingroup\$ The main problem with using Arcana is that while Intelligence is how Wizards understand magic, it's not how Sorcerers do...which actually brings up an entirely different problem in using skill checks. Skill selection in 5e is already hugely limited compared to any of the 3.x based rulesets. Forcing casters to invest one of their choices for better control, particularly when there's no easy "all casters need this anyway" skill any more, is thus much more limiting than in previous editions. \$\endgroup\$
    – user28753
    Commented May 1, 2016 at 16:07
  • 2
    \$\begingroup\$ The easy answer to that would be to have the skill check be based on whatever stat the spellcaster is using to cast the spell. \$\endgroup\$ Commented May 26, 2017 at 16:50

As a DM, it appears from this (non) problem that you are missing the point from two directions.

  1. Taking the adversarial position, you versus the players. That isn't what this game is. Your efforts are somewhat like a good teacher or a good coach: give your players challenges of increasing difficulty so that they grow. The players aren't playing against you, they are trying to overcome the challenges that you create for them. There is a non-trivial distinction in that point of view.

With or without a grid is irrelevant in this case. What is relevant is getting too emotionally invested in your monsters. (Been there, done that). You are neutral, as the DM, as regards to either side in a fight. The original role the DM played in this game (when it was invented) was "referee." That role is still a part of what a DM does (not all of it). Part of your challenge is to balance the roles of referee with that of tactical planner/executor for the monster/NPC side.

  1. Beyond the points of player agency in other answers that I won't repeat, one of the hardest things to do is to represent how good a character is, particularly one who uses a given spell frequently. To use a sports analog:

Just as a golfer gets better at hitting lob shots with practice, just as a pitcher improves accuracy by finding his "eye" for a spot on the plate, just as a quarterback improves accuracy with timing and practice, and just as a basketball player improves eye-to-hand-to-target accuracy with his 3-point shots or free throws, your wizard gets better at placing a fireball "just so" by working on it. We have a lot of abstractions in this game, and a wizard is going to try to place that spell as accurately as possible as often as possible.

Why rain on that parade?

Your player is not playing against you, he's trying to defeat the monsters. Make the monsters play harder or smarter: that raises the challenge for the players and their characters.

Regarding your comments about cheese and shooting into melee or casting spells/AoE into melee: this is 5e, there isn't a rule about shooting into melee with missile fire any more than there is a rule about how close you can get to a fireball without being burned. That's either a judgment call, or a case of measurement. Make a ruling/judgment call and press on. That's a DM role: making rulings.

  • A proposed ruling if you feel that you need something that fits into this edition of the game: when a PC ally is too "close to the edge" of that AoE spell, apply the mechanical feature used the Monk's and Rogue's Evasion (7th level) class ability which is that the PC takes half damage on a missed Dex save, or no damage on the Dex save. (PHB p. 79 and 96 respectively). Since the PC is "somewhere in that 5' square" you can explain this (with an eye to realism) as the result of the combination of six seconds of fighting and movement combined with the timing of when the spell goes off. Since the game isn't any more granular in detail than the six second round, this both allows for a chance to avoid damage, and to have some consequences for 'shooting into melee' ... try it the next time you play.
  • Experience from an earlier edition: what we did in earlier editions isn't necessarily germane to 5e since adding more rules seems to contradict the idea behind this edition (to take nothing away from Daniel Collins' excellent response). We used a variety of rulings / conventions about shooting into melee in earlier editions -- each table had its own variation on that. If a DM thought someone was cutting it too close, I saw any number of rulings along the lines of "oops, you toasted the thief for X damage" or have the thief roll a dodge chance when the fire ball went off a bit too close. The proposed ruling above tries to keep such a house rule within the general theme of this edition.

Think through your homebrew or modification of fireball, and consider a playtest rather than just springing it on your players: if you decide to make the wizard roll for a "to hit" (that spot) with the fireball once, then why aren't you making the wizard roll to hit (the spot) each time a AoE spells is being used?

If you want to have a rule for what you see as cheese, homebrew / modify fireball and be done with it. After you do that, based on your introduction, you can be sure that your player(s) will complain about that too. Then you, the DM, will do what in response to that? The relationship between you and your players can make or break the game.


There are basically three options here.

  1. RAW: Don't ignore the part of the spell description that indicates that the environment is affected by the spell as well. I.e.

    It ignites flammable objects in the area that aren’t being worn or carried.

    Meaning, sure the spell may not affect the player directly, but perhaps it burned the rug the ally is standing on, and now they take 1 point of fire damage from the burning environment, or the fire spread in other ways. However, this won't help you with other Area of Effect spells, which do not describe such environmental hazards.

  2. RA(DM)F (Rules as the DM's fun): You want spells to be less predictable, and you don't think that characters should have laser point accuracy. At our table, we added a bit of a random modifier to ranged area attacks for things like grenades. First, you roll a 1d4 indicating how far off target you are, (A 4 results in being directly on target making the d8 superfluous) and then you roll a 1d8 to decide in which direction you missed your intended target. The d4 result of 1-3 has for us either been 1-3 feet or 5-15 feet depending on the item and what made the most sense. Another option which you can try, but I have not done myself so I don't know how well it works, is to declare that when it says you can "choose a point", it really means you can choose a clearly identifiable item acting as a point. So you can target a rock that is being used for cover or an item of note in the room, and that becomes your target, but you can't actually target an arbitrary abstract mathematical point. Personally, I think such a rule would cause too much book keeping, but it might work for your group.

  3. Something Else: Perhaps this is an indication you prefer close quarter combat, rather than open areas, are you have the enemies which aren't adjacent to the players grouped together too much. Your problem with the accuracy of the spells may actually be an indication that you dislike another aspect of the way the game is running, and this is only a symptom.


House Rule Suggestion: Firing into Melee

For the last few months my group has been using a house rule for 5e we've called, "Firing into Melee". It adds two rules

  • Two creatures are in melee if either has attempted to attack the other within the last round and neither creature has moved since then.
  • An area effect (be it spell, attack, trap, or something else) that affects a creature also affects any creatures in melee with that creature.

The effect of this has been that our blasters can still toss out fireballs as before but may need to skip some enemies to avoid hitting our front liners. We've found it helped with our immersion as, for example, the fighter who is dueling it out with some hobgoblin can't be picked off with (what is functionally) a well aimed hand grenade without harming the hobgoblin.

Some oddities: Larger creatures

This rule does create some odd situations for larger creatures. If a giant is punching at a halfling you make think it would be possible to catch the giant in a fireball but avoid the much smaller halfling. This rule makes no allowances for that kind of thing though the DM can of course rule them on the fly. In our experience we've generally handled these kind of situations by granting a much smaller creature advantage on their save as they duck/dodge/what-have-you behind their larger opponent.

  • \$\begingroup\$ After reading this again, I find that I really like your house rule. I'll propose it to our group for our next play session. \$\endgroup\$ Commented Jun 13, 2018 at 16:30

Don't use a grid system

If you are tired of pin-point accuracy, I would suggest you stop using a pin-point accurate grid system. In my experience, it's fairly easy to switch to a non-grid system while still using minis. Actually in one of my campaigns, we frequently switch between them, based on the level of sloth we're feeling. Ultimately, it's up to you as DM to facilitate this change.

So, why remove the grid system? Well, for one, it allows creatures and players to position themselves in between the usual 5 foot increments. This gives you more flexibility to place monsters, obstacles, and even things that can protect against fireball, like large barriers (the fire spreads around corners, but you can still say a barrier helped them dodge it). In addition to this, it makes it difficult for your players to calculate the perfect area of effect. They're still going to ask you "does this hit that guy?" and you need to determine if that's the kind of information you're going to give out. I'd suggest you don't tell them anything beyond what their character ought to know. Let them pick a point, and determine the results. No take-backs.

And finally, a caveat. This sort of system of positioning can be really subjective. You need to try your hardest to be fair and use the positioning as an element of the challenge. Give them some leeway for ranges of attacks (don't say, "nope, that guy is 5.01 feet away, you cannot attack him"). If they're doing too well, fudge the numbers a bit to your advantage. At the end of the day, where you think things are is where they are.

If you handle it correctly, it should stop the ease-of-use of fireball, but without taking away player agency.


Here's one option:

The wizard can point to any square they like, and the fireball is centered there, just like the spell implies, but there's nothing to say they have infinite time to figure out optimal placement -- in reality people are moving, distances are hard to judge, and so on -- if they have to stare at the map and start counting squares, then the character is taking that time (to figure out the equivalent -- this person is 23 feet away in that direction and that one is 17 feet that way, and so they should place the center exactly here) before the spell casting begins.

If it only takes them a couple of seconds, maybe their spell is just the last one to go this round. But if it takes them very long, I'd say that they spent their time figuring placement very accurately, and their spell will be delayed until next round.

This approach can apply more broadly to players who spend time trying to over-optimize everything in a way that make it seem like chess rather than a pitched battle -- not just area of effect spells.

  • \$\begingroup\$ Have you done this, and if so what are the pros and cons? \$\endgroup\$
    – mxyzplk
    Commented May 5, 2016 at 13:50
  • 2
    \$\begingroup\$ @mxyzplk A GM I have played with has used a very similar sort of approach against all kinds of micro-strategizing and indecision during combat. It works well; it can be annoying if you're the one trying to over-strategize but it seems to improve the experience, in a similar way that mechanics in computer games designed to get you not to grind can seem annoying if you try to grind but ultimately make for a more enjoyable game. \$\endgroup\$
    – Glen_b
    Commented May 5, 2016 at 15:59
  • 1
    \$\begingroup\$ Does this unduly penalize players that are just plain slower than others? That would be my biggest worry, we have a couple players who are just not as quick on the trigger as others. \$\endgroup\$
    – mxyzplk
    Commented May 5, 2016 at 16:01
  • 4
    \$\begingroup\$ I haven't downvoted this answer but also haven't upvoted it; I feel like it needs a little more detail and showing pros/cons from experience to not just be "another random suggested approach." \$\endgroup\$
    – mxyzplk
    Commented May 5, 2016 at 16:03

I agree with the concern, largely because it bothers me that combat can be treated as a place where calm, methodical calculations have any place. I mean, sure, it's part of the genre where you're optimizing things numerically all of the time (who's going to keep their +1 sword after finding a +2, just because the +1 is prettier or was given to them by their grandfather?) But there's a huge gap between optimizing skill and equipments choices vs relying heavily on ridiculous levels of precision in the heat of combat.

But even more than verisimilitude, I care about game balance, and in my mind the real problem is that Fireball is clearly too powerful when usable in this way. (Even when the monsters are using it -- it just tilts the style of combat a little too much for my taste, regardless of which side it favors.) So I'm all for house rules that fix both problems (balance and verisimilitude) at the same time. There are all kinds of possibilites (note that it's been a long time since I've played, so I'm going to ignore the exact rules of the various versions):

  • Space expansion. A fireball is a sphere, and if the ceiling is lower then it will expand out more. (Different versions disagree about this, but whatever.) And even if the wizard is an expert at knowing the exact height of a ceiling that is both fully visible and flat, real ceilings are unlikely to be either of those. Caves are lumpy and dark. Buildings have slanted ceilings and beams and air vents (holes). I wouldn't try to do the math for any of this, just accept that the exact radius varies a bit.

  • Targeting. It may be in the rules, but even with a perfect grid to guide them, it seems ridiculous to me that a wizard would be able to target and hit an exact spot in space (particularly when the ground is lumpy! Height matters.) Fireball is not meant to have the unerring accuracy of Magic Missile; it's supposed to be a somewhat messy, splashy type of spell. How far off can be dependent on whatever you want it to depend on.

  • Targeting again. So maybe your wizard is really good at placing the center of the burst. How long does it take? You could allow them tighter placement in exchange for more time spent aiming.

  • Positioning. If anyone is using dexterity to avoid getting hit, they're not stationary! (I suppose you could allow them to be -- "you're guaranteed to avoid the blast as long as you're willing to act as if you have a dexterity of 3".)

  • Damage. Ok, I know the rules say there's a sharp boundary between getting fried enough to kill your average human 3 times over, and feeling a warm breeze on your face as you watch your enemy get engulfed in flames. (It would make more sense for your enemy's body to shield you from the worst of the blast.) How to take that into account without unbalancing the spell too much is a little dicey -- but still, there's a reason why foot soldiers don't chuck grenades behind the backs of the opponents they're facing.

  • Non-damage effects. Fireballs are bright. Opponents are facing away from the bright shiny thing. Allies are facing directly towards it. (In a cave, for full reality, your allies would be blinded for several seconds, and it could take minutes for their eyes to adjust. Unless the wizard called out for them to close their eyes, which might not work out so well for them. I do not suggest taking it to that extreme, though!)

  • Commotion. I doubt fireballs would be very quiet, and they're certainly bright. If you need to notify everyone of your presence for a mile around, it seems like setting off a fireball would be a very effective tool. If you don't want everyone coming over to see what's going on, though...

Anyway, my point is that there are lots of rationalizations available, but you need to justify whatever you do based on game balance as well as verisimilitude. And sure, you can adjust monsters' tactics, but that is unsatisfying to me personally because you'd be adjusting them for something patently implausible, which means their resulting tactics start to diverge from "reality" too much. They'll end up spreading themselves so thin to avoid ridiculously well-targeted fireballs that they'll end up letting opponents through to stab them in the back.

  • 4
    \$\begingroup\$ I used to play back in the day on the 2nd e rules. Fireball was easy to restrict because it was horribly, horribly destructive. cracked armour, brittle swords, burned potions, the whole area reeks of burnt hair.... and the screams of your enemies burning alive... Any player that dropped a fireball rules-lawyer-style on a grid would have had to deal with the aftermath of the other characters... you drop a fireball inches from the face of my Int 6 fighter and unless you're in a horrible, horrible situation, you're going to wish you killed him in the blast radius. \$\endgroup\$
    – mgjk
    Commented May 2, 2016 at 20:44

Personally, if I were doing this, I'd buff the spell (getting nerfed feels bad), but make it more uncontrollable - which is sort of what you want.

Change it from a 20' radius to a 15' + (d4 x 5') radius. You're guaranteed to get all the bad guys you would have gotten with the vanilla version, but maybe you might get more. Some allies might get scorched. Collateral damage :P

If you want to give back some control, a successful Arcana or Perception check could give you "advantage" on the radius roll (2d4, keep 1) while still maintaining a little bit of chaos. And obviously, if they really want control over spells like fireball, they could be an evocation wizard, and just exclude their friends.


I do this

I use a houserule that functions much in the way you are asking for.

When the wizard is casting on 'known terrain', in an area they are at least passingly familiar with (or if novel, they have at least 'paced it off' before the battle) I allow them to place the center of their AoE where they would like.

If they do not have at least passing familiarity with the dimensions and obstructions of the space into which they are launching their AoE, I require an Arcana check, much as in Dan B's answer. As the comments on that answer suggest, an Arcana check favors a wizard, but that is typically the class of my AoE casters. If I had a sorcerer or druid, I would just call it "a check using your spell attack modifier".

I set the DC for the check as standard for any check. An 'Easy' check might be centering the AoE on a square the caster saw at the beginning of the fight, but which is now obscured by darkness or a fog cloud. 'Moderate' might be something like landing the AoE perfectly on the field to hit all opponents and precisely miss friendlies, when all of them are in motion. 'Hard' might be something like sending a fireball 'steaking from your finger' through an arrow slit and landing it in what you assume is the center of a tower, even though you haven't actually been inside the tower.

If the character makes the check, they place the center of the AoE exactly where they want it. If they fail the check by one class (for example, it was Hard and they failed a Hard check but would have made a Moderate one), the center is one intersection off (5') from where it was desired, in a random direction. Failing by two classes is two intersections (10') off, etc.

In practice, this emphasizes the 'home field' advantage for defenders and encourages the PCs to scout the areas they think combat might take place. It incentivizes them to make 'easier' placements rather than harder ones, and if they choose to attempt more difficult placements, to feel great when they work out and to accept that sometimes they roll poorly and something goes astray – either they don't hit as many foes as they want or there is some friendly fire. As their spells get more powerful (and the consequence of a failure becomes greater) their spell attack bonus grows, and the chance of failure decreases. There is seldom a failure of epic consequence, but when it happens it is because they have chosen to risk it. It allows a certain amount of role-played tension between characters that have a higher tolerance for risk and those that are more risk-averse.

In general the responses of my players to this house rule have been favorable.


I'm running 5e on a grid with minis. The one thing that I hate is when a sorcerer or wizards casts a fireball that perfectly hits enemies in the aoe. The player counts the squares on the grid to determine exactly where the fireball can hit and he knows that the perfect way for the fireball aoe to take effect. It will hit enemies but amazingly the blast stops just in front of an allies face. A player who was attacked by 2 melee enemies in my game casted a fireball behind the enemies so that the enemies were hit in the explosion but he wasn't. In my opinion, this loses verisimilitude.

There are few meta-issues here that I think are at the core of the issue...


"Verisimilitude" - The game is about fighting fire breathing dragons; thwarting the machinations of evil gods; defeating vampires, liches, and other undead; riding a Unicorn or a Griffon; and, yes, casting and manipulating magic! All of those things require a fair bit of imagination to conjure, let alone play in any meaningful and balanced way. The game allows players to throw fiery magic balls of Troll-death and its fun!

Sorry, but verisimilitude is out the door, go play Twilight 2000 if you want realistic combat.

Keep it going...

That said, I do agree there is a problem with players planning the perfect circle AoE to only hit the enemies. The issue though isn't the spells, it is that play virtually stops while the player (usually players) figures out the perfect spot to land the spell.

I see Wizards as having a clear idea of what their magic can and cannot do. They will know their limits as clearly as an Cleric healing the wounded, or a Thief picking a lock, or a Fighter sizing up an enemy. But unlike the Cleric, Thief, or Fighter - the Wizard can pin-point their magic... and that's fine. Although unlike the Cleric, Thief, Fighter, or any other class, a wizard without spells is about as good as a pitchfork wielding commoner. Let them have their moment.

However, I think if you gave the player about as much time to pin-point their attack as you give the Fighter to roll their attack dice, I have a feeling you'll see a lot more PCs getting roasted, or at the least more enemies surviving. Just because the Wizard can pin-point an attack does not mean they get more time to think about the best place to pin-point. With that in mind the PC might fudge a bit further away just to keep PCs out of the way - but that means engaged enemies might be missed too!

Yes, they can pin-point. But if you keep the play going... they won't have time to.

Crafting Rules

Creating... crafting... house rules are something we all do. And a lot of those house rules can make a bad situation worse. I've learned both professionally (designing computer games) and in my own table-top games that a level of care must be taken when Crafting Rules, especially those designed to 'fix' other rules.

I've seen several suggestion here, as well as elsewhere, that are to impose new requirements or restrictions: randomize the pin-point, adding a skill check or attack roll, applying modifiers based on stats, etc.

  • Random Pin-Point: In these rules, the desire is to add a random factor into Magic that the other classes have to endure. The main problem in doing so it can nerf the Wizrd - remember, the Wizard has a finite number of attacks while the other classes do not endure this restriction! The power of the Wizard class is the inherent ability to pin-point an attack, taking this away is actually a severe limitation.
  • Skill Checks or Attack Rolls: These can be hit or miss, usually miss, as a Player generally invests in Int or similar skills and forgoes combat prowess so as to maximize their magic (& pin-pointedness...) Applying these rules at character creation can be fair in so long as the skills/attacks are something the Player could choose normally and don't detract from the Wizard's primary focus -- throwing Magic. Applying these rules after character creation only slams the Player for pre-game and in-game choices already made, causing a new problem... an unhappy Player!
  • Stat modifiers: Unless the stat in question is Intelligence, don't bother - you end up with the same results as Skill/Attack above... an unhappy Player.

After learning the hard way, I steer clear of house rules that modify X rule with Y rule since doing so invariably causes new problems (usually with unhappy players because of the new restrictions to any desired type of play).

I hate is when a sorcerer or wizards casts a fireball that perfectly hits enemies

Sorry, but that's what magic users do, and you as GM should let them. If you hate it so much to create limitations, the problem isn't the rules, or the players knowledge of said rules...

Don't create rules that restrict, limit, or hamper a Players' desire to play the way they want to play!

Crafting Encounters: Dming Advice

How you craft your encounters is by far the best solution to Players use of AoE.
Yes, it isn't a rule per se and yes, it doesn't address players pin-pointing their attack (see above). But it is the best thing you as GM can do to reasonably change how Players use pin-point AoE In a way, this is the best rule of all -- you get to use your imagination in crafting interesting encounters. A great deal of the the issue of players using AoE spells can be mitigated when crafting the encounters in the first place.

Anyways, what's good for the goose is good for the gander! Why aren't there any enemy Wizards lobbing balls of fiery doom exactly where only the PC's get hit?

I strongly urge you to NOT apply any rules to address the issue. Just keep things going and keep the combat interesting and yeah, occasionally the Players will fire off a Fireball that only roasts the enemies. That is because doing so is a lot of fun - and your primary job as the GM is to ensure that that is what the players are having... FUN!

  • 6
    \$\begingroup\$ Simulation is one of the "8 kinds of fun" people get out of games (8kindsoffun.com, or specifically explained WRT RPGs angrydm.com/2014/01/gaming-for-fun-part-1-eight-kinds-of-fun). You are, in really long fashion, just saying that's not the kind of fun you like and so the OP should not do it and should do a different kind of fun. That's unhelpful. \$\endgroup\$
    – mxyzplk
    Commented May 5, 2016 at 16:06
  • \$\begingroup\$ @mxyzplk 8kindsoffun.com lists Sensation, Fellowship, Fantasy, Discovery, Narrative, Expression, Challenge, and Submission. What exactly is the "Simulation" here? or was the list changed from 2016? \$\endgroup\$
    – enkryptor
    Commented Oct 15, 2021 at 16:10

My problem with this whole argument is that you are assuming they (wizard) would be having to make calculations in order to know exactly where to place the fireball as to not hit themselves/allies when they are danger close.

Does a quarterback have to do calculations when avoiding the defense to know exactly how to throw the ball get to the receiver?

Does a pitcher have to do calculations in order to know exactly how to throw the ball to land in the strike zone?

Does a soccer player have to do calculations to have to know how to kick the ball so it will fly right into the corner of the net?

While a lvl 5 wizard shouldn't be compared to a professional athlete I would compare them to amateur ball player and they are still very accurate even at that level, and when its something that you work this hard at is something that you should be able to, with very little thought, be able to do (know the limits of).

The best answer I would recommend for someone who absolutely has to impose some kind of restriction on this spell would be to have them make a skill check accuracy roll and add proficiency bonus + int modifier of 8. Meet/beat perfect hit, below they are 1 square off in, roll 1d8, for a 1 is 1 square behind, for a 5 is 1 square in front of targeted square. For a critical failure it lands 3 squares in rolled direction meaning you possibly hit large number of allies or missing enemies completely wasting spell slot.
But this said, I would then want the same added to targeting an enemy.
I am not a fan of your premise for this complaint on the whole and honestly it feels like you just don't like the fact that this can be used in the listed scenario which just makes me glad I'm not in your campaign if this is how you are trying to run things.

**Note: I understand this is a 5 year old post, stumbled upon it looking for something else and had to put my thoughts down on such a bad idea.
No I did not read any of the other answers, ain't nobody got time for that.

  • \$\begingroup\$ This answer is 80% disagreeing with the question, and 20% a suggestion for what to do. Have you implemented your skill check at the table? How did it go? Did your players like it? \$\endgroup\$ Commented Oct 15, 2021 at 22:22

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .