D&D Blog. Should Fighters get multiple attacks?

[am181d]

I'd like to see Fighters attack once per round (or twice while dual-wielding) and then have their damage dice scale as they level.

So a 20th level Fighter can do a ton of damage with his two-handed sword, but he's only rolling to hit once. (And since he's 20th level, he's probably got a few different options for adding a special effect to his attack, like a trip or a sunder or a slam, etc.)

 

[Mercule]

The warlock in 3.5, was also a simple magic user.

Good catch. I also don't mind the idea of the "simple" mage being an extension class, with the PHB being simple = fighter, complex = wizard.

simple fighter - 3.5 fighter - lets call him fighter
complicated fighter - 4e fighter - lets call him ritual warrior or tactician or something

I guess I've always viewed things like the paladin, cavalier, even ranger as "complex" fighters. I don't know that we need to invent a new catch-all class for complex fighter types, so much as recognize what niches the other fighter-like classes fill:
- Fighter = basic combat ability, much as one would expect from a non-officer soldier or mercenary.
- Paladin = a warrior, first and foremost, but one with a few extra tricks from piety.
- Cavalier = Fits the mold of the courtly warrior. Focus on expertise with knightly weapons and some social abilities, but weak without the tools of a noble. Good with the rituals of honorable combat.
- Swashbuckler = Generally viewed as a light-weight warrior, but really more of a tactical, improvising warrior focused on being very good with light, adaptable weapons and in using terrain to his advantage.
- Kensai = master of one weapon or style to the neglect of others. Could be rolled into the core fighter more easily than the others (maybe).
- Barbarian = natural fighter that lives by "the best defense is a strong offense". High output, high hit points, lousy armor. Not specialized in any weapons.
- Ranger = The wilderness, archery equivalent to the swashbuckler. Potentially more geared toward strategic dominance than tactical, as well. Often defenders of the civilized world from certain, specific threats.

YMMV on the list or definitions, but the point remains that combat is so varied that making the fighter more complex and adding options above being good at sticking the pointy end into the other fellow usually brings along some baggage about what sort of options you want, which can lead to another class. Maybe a kit-like mechanic applied to a base fighter is the best way to solve that. Maybe that's only relevant in certain cases (e.g. cavalier and swashbuckler, but not ranger, paladin, and barbarian). The basic concept of a fighter really is pretty darn basic, though.

 

[Fanaelialae]

(Although I'm not sure design is any harder than for increasing damage. Having both is the problem.)

The reason, IMO, that design for multiple attacks is harder than increased damage, is that multi-attacking introduces "jumps" in damage.

For example, if Joe the Fighter gets +1 damage per level, his damage output scales smoothly (presumably keeping pace with the scaling hp of the creatures he's facing).

If Bob the Fighter gets +1 attack at 5th level, his damage out is only where it should be (presumably) at 1st and 5th level. In between those levels, he falls behind and may struggle.

Now, you could give Bob a damage bonus as well, but that's where the math breaks. At 5th level, his output will suddenly double, meaning that those bonuses suddenly double as well and Bob is way ahead of the curve. Now the game is set to easy.

All of this is before factoring in the increased impact of crits and accuracy that multi-attacking grants. A damage modifier is easy to analyze mathematically because it is force additive, but multiple attacks are a PITA because they are a force multiplier.

I'm not outright opposed to multiple attacks, but I think they should probably be limited to two attacks against a single target at most, and the second hit should just be rolled weapon damage (it shouldn't add any damage modifiers). I think that would make it much easier to include multiple attacks without breaking the math of the game.

 

[Hassassin]

The reason, IMO, that design for multiple attacks is harder than increased damage, is that multi-attacking introduces "jumps" in damage.

Not necessarily - or at least they don't need to be large. Take the Trailblazer approach where at level 6 you get two attacks at -2. That increases your typical damage by 50-70% depending on what you need to hit. Make it -4 and you are down to a 10-40% increase.

Of course, 3/2 attacks from AD&D were another approach to reducing jumps.

All of this is before factoring in the increased impact of crits and accuracy that multi-attacking grants. A damage modifier is easy to analyze mathematically because it is force additive, but multiple attacks are a PITA because they are a force multiplier.

(Assuming a confirm roll, crits get taken into account automatically.)

A damage modifier isn't that easy either. You don't know what the 3e rogue's +1d6 every two levels actually means before you know his weapon and strength. Ditto for a +1/level increase.

If you flatten the math, even a +1 to attack / three levels is "jumpy".

 

[thewok]

This, or a near facsimile of this. I thought that Saga did multiple attacks really well. You could choose to make multiple attacks if you wanted, but every attack would take a penalty (for example, make two attacks at -5 or three at -10).

One of the most important elements of this was that all attacks were made with the exact same bonus, allowing you to roll all your attack dice at once (and avoiding the mathematical horror of 3e's iteratives). It also made multiple attacks a strategic choice, which gave fighting types something more to think about.

Now I'm not sure if the Saga penalty was right (-5 is pretty steep), and making it a feat tax was a bummer, but like so much in Saga I think the idea was good, it just needed tweaking.

The great part about Saga was that it took feats or talents to be able to do these things. But, you could also take feats and talents to make that one attack you get per round even better. And (on a larger scale), talent and feat selection could make two people of the same class that were almost totally different from each other.

My friend made a dual-wielding scoundrel. He liked to put out as many blaster bolts as possible in a round. So, he started taking things like Double Attack, Triple Attack, and the talents that allowed him to reduce the penalties on those.

My scoundrel was built for accuracy. I chose talents and feats to help with that one shot. If I aimed before firing, I could knock an enemy down three or four spots on the condition track, depending on how well I rolled damage.

I think that each class should be fairly basic on its own. One attack per round, and so on. Then, as the character gets more levels, he should be able to take feats that allow him to specialize in how he wants to play. If the Fighter wants to make more attacks per round, then he can spend feats to do it. If he wants one massive attack per round, he can spend feats for that, too. If he wants a more balanced Fighter, able to do more damage to one target or spread it around, then he can take both kinds of feats. He will have breadth of ability, but not mastery.

And I think that should also be a viable choice, but it shouldn't be like the 3E bard, which was by class design. It should be by choice of the player, and it should be a perfectly acceptable way to play (and excel at) the game.

 

[Mercule]

I had this thought in another thread, but it's probably more relevant, here. What's anyone think?

Could this be the answer to multiple attacks and/or weapon specialization? At 7th level, the fighter rolls 2d20 for all attacks and takes the better. At 14th level, he rolls 3d20 and keeps the best.

I could probably add some structure around that to make it feel more like multiple attacks or super-specialization, too. Maybe grant the bonus rolls for free and then have a feat tree or two to represent different focuses.

Power Attack
Prerequisite: two or more attack rolls.
Effect: Starting with the highest die rolled, compare the result to your target's AC. If only the highest result would hit, do normal damage. If the second highest roll would also hit, add a competence bonus equal to your strength bonus (1.5 str bonus for two-handed weapons) to the damage dealt.

Improved Power Attack
Prerequisite: Power Attack, three or more attack rolls.
Effect: As per Power Attack, plus if you have a third attack roll and it also would hit, the bonus increases to double your strength bonus (triple for two-handed weapons).

Whirlwind Attack
Prerequisite: two or more attack roles.
Effect: If your highest attack roll would hit your target, pick a second target you threaten. If your second highest attack roll would hit that target, that target takes half the damage you role against the primary target.

Improved Whirlwind Attack
Prerequisite: Whirlwind Attack, three or more attack rolls.
Effect: As per Whirlwind Attack, plus if the second attack hits, and you have a third attack roll, you may choose a tertiary opponent you threaten to compare this roll against. If it succeeds, this target also takes half the damage rolled.

Weapon Specialization
Prerequisites: proficiency in weapon, two or more attack roles.
Effect: If only your highest attack would hit your target, apply damage as normal. If your second highest attack would also hit, apply half your BAB (formula assumes 3e BAB) to the damage dealt.

Weapon Mastery
Prerequisites: Weapon Specialization in weapon, three or more attack roles.
Effect: As per Weapon Specialization, but if your second and third attacks also hit, you apply your full BAB to the damage dealt.

I have a couple ideas on Weapon Finesse and Two Weapon Fighting, but you get the point.

I'd also see the stylistic modes (power, whirlwind, finesse) being mutually exclusive, but specialization (and, maybe TWF) being additive. You could also add other feats allowing combinations, like:

Berserker
Prerequisites: Power Attack, Whirlwind Attack
Effect: The effects of Power Attack and Whirlwind attack are cumulative. Apply Power Attack damage to the first target before calculating damage to the secondary target.

 

[FitzTheRuke]

The warlock in 3.5, was also a simple magic user.

I think it would be neat if in the core rules they could have a:

simple fighter - 3.5 fighter - lets call him fighter
complicated fighter - 4e fighter - lets call him ritual warrior or tactician or something

simple mage - 3.5 warlock or something - warlock
complicated mage - 3e wizard - lets call him wizard.

This would give more choice to everyone.

Sounds good to me... and very likely what they are going for (probably not exactly, but I bet they are thinking along those lines.)

 

[Fanaelialae]

Not necessarily - or at least they don't need to be large. Take the Trailblazer approach where at level 6 you get two attacks at -2. That increases your typical damage by 50-70% depending on what you need to hit. Make it -4 and you are down to a 10-40% increase.

Of course, 3/2 attacks from AD&D were another approach to reducing jumps.

No, by doing that you only reduce jumps in average output. You still see huge gains (doubling) in minimum and maximum output per round, which also should be accounted for.

With an additive increase, it's very obvious by how much you're increasing damage because you increase minimum, maximum, and average all at once. Without the kludge of attack penalties, I might add.

(Assuming a confirm roll, crits get taken into account automatically.)

A damage modifier isn't that easy either. You don't know what the 3e rogue's +1d6 every two levels actually means before you know his weapon and strength. Ditto for a +1/level increase.

A flat damage modifier is very easy to calculate. As for the rogue's sneak attack bonus, it's still just a very straightforward range. The same goes for weapon and strength.

However, if a rogue can make 1 attack at +5, 2 attacks at +3, or 3 attacks at +1, it's significantly more complex. Those 3 attacks styles produce very different results depending on the the AC they target. Those three attack sequences will produce a very wide range of numbers, and therefore be much harder to balance for.

If you flatten the math, even a +1 to attack / three levels is "jumpy".

I'm pretty sure they're only talking about flattening the attack bonus. +1 damage every level is a fairly moderate and reasonable rate to scale at.

Multiple attacks make everything jumpier. If I give +1/3 levels to a PC who can only make 1 attack, that's still just +1 damage every 3 levels. Perhaps +2 on a crit. If I give it to someone with 3 attacks, that +1 might be worth +0 (if your attack penalty causes you to miss with all three attacks), all the way up to +6 (critting with all 3 attacks).

In essence, what I'm saying is that it's not enough to just look at the average damage. You also have to look at the minimum and maximum damage. Otherwise you end up with orcs who can one shot a healthy fighter with a better than average attack sequence, because the math is only balanced against the assumed "average".

Simply put, it can easily lead to broken math. That's not the type of game that I, personally, want to play.

 

[WarlockLord]

The question I have is:

Who cares?

Whether or not the fighter gets multiple attacks does not fix the core of the fighter's problem: that it's an inherently low level concept which cannot compete with the divine avatars and archmages roaming the land at high levels. Fix THAT, then we can worry about the minutiae.

 

[Hassassin]

No, by doing that you only reduce jumps in average output. You still see huge gains (doubling) in minimum and maximum output per round, which also should be accounted for.

True, there's a jump in the variance. However, increasing damage increases variance even more.

[sblock=Example math]Suppose attacks are balanced to hit with a natural 8 or higher. Assume e.g. damage 1d8 without bonus (any will do).

To get equivalent average damage with two attacks, you need to apply a -6 penalty. Variance increases about 20% (standard deviation 10%).

To get equivatent average damage with double damage, you still need to apply a -6 penalty. However, variance increases 80% (stddev 35%).[/sblock]
Multiple attacks lead (unsurprisingly) to more consistent combat performance than damage increases.

Simply put, it can easily lead to broken math. That's not the type of game that I, personally, want to play.

My point was that they aren't significantly more difficult to analyze. You can balance the effects on average damage and variance pretty much as easily as with attack bonus or damage increases. I prefer multi-attack because it leads to more consistent performance, which should be the fighter's "thing" IMO.