How to mitigate glass cannon syndrome in Pathfinder?

Question

This is the more gamey part of my other question, What to do when your character is just too good?

In D&D 3.5e and Pathfinder, damage output grows very, very quickly. Much more quickly than hit points or healing grows. This leads to a drop in fun when any round can be a one-round kill for any PC or opponent. (See linked question for example anecdote and character build.) As splatbooks grow, so do options for more damage-stacking from variant class abilities, feats, magic, and loot. Once we hit even modest levels (8-12), many combats become about who's going to get in a full attack first, because anyone does enough damage with a full attack routine to drop... Well, themselves for sure, so any NPC of equivalent level. At level 10 many characters can do 100 points of damage a round without trying very hard.

Are there any technique changes or robust house rules one can use to keep this huge DPR escalation from happening? I know about E6, which definitely accomplishes this by simply stopping level inflation, but it makes it hard to use a lot of published adventures and supplements, which my gaming group, as a bunch of adults with jobs and kids and stuff, find quite valuable. I've thought about a simple damage cap of some kind (e.g. no more than +level to damage regardless of source) but that has a lot of system side effects.

I'm not interested in "changing" systems (we play a variety of RPGs, but Paizo's brilliant adventure work brings us often to Pathfinder) but ideas from them are welcome.
It seems to me that since this is a problem largely introduced/escalated in 3e/3.5e, it's not endemic to all D&D, and D&D versions aren't all that different, so it seems like there might be some kind of more gentle fix... (I have Basic/1e/2e experience too).

Answer 1

Max Hit Points for everyone

Okay, bear with me for a minute, because this is a little strange. My group did this once for a Pathfinder campaign. It had some interesting consequences, which caused us to not do it again, but it might be of some use.

Leveling with Max hit points

So, simple. Every level you gain the maximum allowable for your class plus your CON modifier. You have a lot more hit points to play with, and are less likely to get devoured in one fell swoop.

Enemies with Max hit points

Instead of taking the stats listed in the Bestiary, take the number and type of hit die and figure out their absolute maximum. This makes a whole lot of things very beefy and harder to kill.

Consequences

Fights tended to last a decent amount longer, especially with multiple hostiles, since most full attack actions wouldn't do enough damage to one-shot an enemy, but you could still take something out in one round if enough players allocated enough resources to it. Same could be said of the PCs. Even if something got off a really good hit against you, it was rarely enough to cause you to drop in the negatives, much less die.

Problems for Spellcasters

The amount of hitpoints we found outpaced the damage output of magic fairly early. That 6d6 fireball is usually barely enough to put a dink in an opponent with significant hit die. Insta-kill spells were still as effective as ever, though, since we didn't modify the saves of anything.

(This also tended to make blaster-casters memorize spells that were Maximized, so when we did see Fireball it was a flat 60 points of damage at later levels).

Conclusion

We actually found that this made battles last for too long. It didn't really change the dynamic for how the battle would go, aside from how many expendable resources the players would have to consume, but it certainly did away with insta-death problem we were having before.

Answer 2

Not the answer you want to hear, I’m sure, but...

I really do not think there is going to be any simple, straightforward way to accomplish what you want. Damage doesn’t just automatically grow, it’s a side-effect of various specific class features, feats, and items. Which means that any changes to the system are going to require going through each of those specifics to figure out what you need to do with it.

That’s going to mean... something close to rewriting the system. And it’s also going to make using premade modules very difficult, as characters will have to be modified and updated according to the houserules.

It may be easier to simply find a new system that doesn’t have the same problems. It’s going to be very difficult, though, I think to find a rules-heavy fantasy system à la Pathfinder, that has anywhere near the amount of support in terms of premade modules.

Which brings me back to E6. I really think it’s the best solution available here. Scaling characters down in level is annoying, but most likely less work than applying the sorts of systemic changes you’d need to accomplish what you want.

Things I think will not work

HP Inflation

A number of the systemic problems in 3.x come from the existing HP inflation relative to previous editions of D&D. For example, most of the Evocation school: those spells frequently deal damage similar to what they did in AD&D, but because everything has higher HP, they’ve lost a lot of effectiveness.

Ultimately, 3.x already has serious problems where debuffs are far more effective than simply damaging things. Increasing HP is only going to make that worse.

Damage caps

This is just awkward, I think. What constitutes a “single source”? For that matter, Sneak Attack is 1.25 damage per level: if you limit it to a max of +level, is a Rogue now expected to multiclass just because his Sneak Attack is too good?

Limitations on multiple attacks

Keeping people to one or two attacks would dramatically reduce the amount of damage dealt, but you’d be massively widening the gap between spellcasters and martial characters, plus you’d make attacking very “swingy” – one attack means it’s very easy to do nothing for the round, and that sucks.

Requested Legend details

I mentioned as a side-comment that I think Legend does a pretty good job avoiding rocket tag, but it’s not really relevant since Legend does not have any premade modules as of this writing, and conversion to Legend, while perhaps not as much work as you might expect, is certainly non-trivial. However, more information was requested, so here’s what I have to say on the matter.

Legend has a lot of very careful mathematics behind it, and the game has had some very thorough “destructive testing” done to it – conscious attempts to break things as a way to find the problems. It also has a very strong commitment to maximizing the significance of player choice. This starts with character building, where the total number of decisions is kept low while the importance of each decision is high so that rather than worrying over small details, you focus on big, important choices. This carries through to gameplay: it is considered a failure of the Legend system for a player to receive no opportunity to respond to a given threat, or for encounters to provide a player no way to contribute.

And most of all, one of the most sacred concepts in Legend is “A = A′,” the idea that a character of a given level should be capable of contributing on an equal footing as an ally, or providing a reasonably equal challenge as a rival, with any other character of the same level. Legend characters can vary a lot, but their level is supposed to tell you how dangerous they are.

I will also give some important details to be aware of coming from 3.x here on this subject: Legend espouses a very steep leveling curve. A character a level higher than you has a big edge. A character two levels higher than you will be tough for your entire party. A character three or four levels higher may, depending on the build, be all but impossible for even your group to fight. By the time you get to a five level difference, there’s almost no combination of lower-level characters who will be able to beat any character five levels above them.

Furthermore, if you are considering Legend, it also “auto-optimizes” quite a bit: Legend characters of a given level are capable of feats comparable to what a fairly well-optimized 3.x character can do, even if you make no particular effort to make them so. If that’s not the style of play you like, you’ll have to stick to lower levels.

Answer 3

Here are a few things to consider :

Tactics

If a full attack is "all" it takes to down someone, force Move actions. Could come from cover/concealment or combat maneuvers like Trip/Knockdown or Disarm. If you just play "rocket tag", you're implicitly accepting that if you miss, there's a good chance you're dead.

Note that this is exactly what happens IRL between powerful forces in a battle and in this respect, is not a "bug" of the system but a simple truth.

---

Intelligence

So you're targeted by a Master Ninja, who manages to sneak up on you and place a Death attack. Curse you, Hayabusa ! Wait, you made your save ! Time to grab your trusty bow and fill him with arrows...
Well, shit. The bastard has broken/stolen your bow as well. Not only does he make his escape, he's slowed you down by forcing you to have it repaired / replaced while you try to get it back.

This goes a step further in Tactics, and is to be expected of Ninja, for example. Study your target. Know their strengths and weaknesses as well as yours. "That Brilliant Bow of Badassitude could be a problem if I must escape. Can I get rid of it ? What about impairing the bowman ? Maybe I could blind him ?"

As an aside, the GM should make sure the players don't feel cheated though. Extreme competence is assumed as part of the characters (especially past level 10) and robbing them of that can be frustrating to no end for some.

---

Numbers

So they can take the Big-Ass-Monster down in a round ? What about 10 Medium-Sized-Nuisances ? One by itself could barely hope to hit them, but with flanking, teamwork feats and simply being all over the place, the little pests could prove to be annoying. Add in a Leader-type and you could even be worrying your characters.

---

Story happens

Take the characters down a notch or two through Story. Have them stripped of Rank and Privileges by the King or even make them straight Outlaws. Have them stranded on an island (and some of their equipment lost to the sea) after their boat got caught in a storm. Have their home base attacked (and their Mentor killed, leaving them unable to progress in their main class) while they were on mission.

---

Change focus for a while

Once all is said and done, if the combat has become so easy it's boring, it may be time to try a little courtly intrigue for a change. Or why not a mystery ? Possibly meshing with "Story happens" above, there are plenty of opportunities to take the players / characters out of their comfort zone while staying true to the setting. They'll be happy to resume bashing heads once they've stumbled for hours finding a tangible threat to pounce on. Or hey, maybe they'll actually enjoy trading piques with the Jester and decide they want a piece of land and a throne of their own ?

Answer 4

The problem that keeps coming up with these answers is that broad ranging rules changes like the ones listed here is that they negatively affect systems that are just fine, and don't need adjustment. This makes me think that the solution should be something that only affects the 'rocket tag' parts of the game: high damage, save-or-die spells, and other extremely potent effects. My idea is to introduce a mechanic that is specifically targeted at these powerful abilities.

Awesome Points!

Players and important NPCs get 1 (or more) Awesome Point per day. At any time (even off-turn) a player or NPC can spend an Awesome Point to ignore all effects of a specific creature for 1 round. For example, the would let a fighter ignore a full round of attacks, or remain unaffected by a disintegrate.

More powerful enemies may need more Awesome Points in order to make fights last longer.

You may also want to add a restriction that players may only use their Awesome Points when their lives are directly threatened. That is, unless you're okay with players occasionally using them to prevent less-than-deadly effects.

Note: I call them Awesome Points because there are a thousand and one different 'special point' kind of abilities floating around, and I wanted to use a distinct name. Feel free to change it, of course.

Answer 5

I like using Action Points (aka Hero Points, Destiny, etc.). I usually steal the Destiny Point rules straight out of Star Wars Saga with little to no modification, but since you're running Pathfinder, I'd recommend Hero Points, which allows characters to Cheat Death.

It doesn't exactly solve inflating damage, but it does allow players to avoid one-hit deaths themselves. As a GM, you can turn the tables and have a pool of Villain Points to spread out among your bad guys... which can be fun as long as you don't abuse it.

Answer 6

The easy way is to look at how those modules NPC are built and follow the module's expectation.

I'm DMing City of the Spider Queen (a D&D 3e module) and all I get is 16th level wizards with a huge amount of unsynergic feats. This is enough to cripple their effectiveness to the point where anyone I've asked how to play them in an effective way told me to change their feats.

I know from your other question that the current problem is due to high damage output on a bow where haste and rapid shot apply.

Just don't. Spread your ability points where you don't need them (except for Constitution, you don't want a glass not-cannon). Unfortunately the game provides no in-rules solution to the problem. Be unoptimized. Really unoptimized. At the point of not preparing haste. At the point of ignoring that feats exist.

I know it's not a great solution but I guess it works.

Answer 7

One approach you could take (which is not without its problems) is to have NPCs play by different rules to PCs. This is a common 'solution' employed in computer games of all ilks. If the PCs are turning out to be glass cannons then up the HPs of NPCs and down their damage.

The main issues here are that the whole thing breaks down if you have player vs player combat or your glass-cannon problems are limited to a subset of the party.

Unfortunately - it's the best I think you can do short of making wide-scale changes to the system.

Answer 8

This points to one of the largest deficiencies in the AC system of armor protection in my opinion, i.e. it uses on the deflection capabilities of armor, and not the mitigation capabilities. There is the head nodding and hand waving that the AC includes both aspects, and a miss isn't always a miss, but it's really what I said in practice- hand waving. Missing a hit roll by one isn't described by us as players as you contact but do no damage if the guy is in full plate. It's described as missing, and for good reason.

The Dungeons & Dragons Wiki describes a bit of what I've used in the past, i.e. Armor as Damage Reduction.

I've played with that before, and indeed, it does help. I've also played with the numbers, as the help seemed to fall off as the characters increased in level, which is how I realized that without scaling, this isn't enough. So I've tried a couple of additions.

1. Alter armor training, so that fighters actually get a multiple to the mitigation above. This does not apply if flat-footed.
2. As feats add to damage, add feats to add to damage mitigation. That way it becomes a choice; if a fighter is a tank, then feats that add to mitigation would be the choice of the day, rather than damage.
3. As above, but with additional ability choices.
4. Have magic items that actually increase the damage mitigation of the armor beyond the innate plusses.

This isn't an easy fix, but it did work for me, and with the addition of other tactical options, made combat more of a fluid experience, rather than just an exercise in character building and ability activation.

Answer 9

Defence in D&D and Pathfinder is more than just a high AC.

Mid to high level PCs, NPCs and monsters have access to other options e.g. damage reduction, protection/immunity from energy, healing/teleport spells linked to contingency, summoned cannon fodder/illusions/mirror images/concealment etc etc

Not only are they available, they are relatively cheap. In the original question, where are the assassin's arcane/divine caster hirelings? They don't even have to be there to have buffed him in advance.

An additional +1 to AC gives a 5% better chance of avoiding damage from an attack, gaining concealment turns a hit into a miss 50% of the time! Consider the advantage the assassin has if he starts the encounter with greater invisibility and has spring attack. The bowman has to guess correctly where to shoot, hit and overcome a 50% miss chance every round.

Answer 10

Maybe it's something that changed in Pathfinder, but my 3.5e experience doesn't jive with the idea that damage is the problem at high level. Sure damage can get dangerous, but it's got a lot of defenses to mitigate it if you choose to use them. Save or die OTOH is just plain lethal as soon as you get a bad roll, which will inevitably happen.

That said, if you're focused on damage at higher level... are you doing any scouting? If you can plan ahead a bit and know what you're going to fight, spells like Resist Energy are incredibly effective. Complete Arcane has a Mass version that's only a third level spell, too. Protection From Energy is a similar option, which one is better really depends on how many times you get hit.

For physical damage, there's obvious options like Stoneskin for DR 10, or the cheaper Heart of Earth for the same thing, some temporary hit points, and critical immunity with its sister spells. High DR can really ruin the day of anything that needs to hit a lot of times to dish out that 100 damage. (Sadly most of the DR abilities classes have don't scale high enough to really matter.)

There's other staples like Blink & Mirror Image. Magic Vestment and Barkskin can boost your AC a lot fairly easily. Heroes Feast can give the entire party some extra temporary HP and a will save boost for most of the adventuring day. Defensive spells are really powerful, if you're using them.

If for some reason you can't use them (maybe you get ambushed a lot and can't keep spells up constantly), then it gets trickier. If it's a problem you really want to solve by the most direct means necessary, the DM could simply create a custom magic item that grants permanent DR and energy resistance. That's going to curb all incoming damage really effectively and doesn't require any new rules at all, but it's not going to help you against Save or Die. (Hero Points are a good way to make those less frustrating for players.)

Answer 11

The problem here is that DnD really relies on the DM for balance, not the RAW. Despite all the improvements pathfinder has made, the RAW still have all kinds of balance problems, and its very easy for an encounter or an entire campaign to go off the rails. The DM has to be free to dynamically scale the encounter as it is happening to keep the dramatic tension. You dropped that assassin in one round, but what about his friend? Sticking too closely to the published encounters is the main problem here. The DM's job is not to stick to the RAW or the text in the AP, the DM's job is to give everyone a fun game. No one likes encounters that are too easy or too hard. Try to prepare variations in advance.

That said, archery damage output of the kind in your example is beyond ridiculous for 12th level. You shot 6 D8+29 arrows in one round, which is 201 average damage if all 6 hit. A D8 is worth 4.5 points on average, so that is 45D8 points of potential damage...compare that to evocation spells or hit dice...its no mystery what the problem is, archery + classes with static damage boosts (paladin, samurai, etc.) is just out of line. Paizo probably made a mistake with these massive damage boosting abilities. Compare 3.5 smite evil to pathfinder smite evil, for example.

Like you said, characters in pathfinder are glass cannons. But that is how DnD characters have always been...at low levels. An enraged barbarian can one-hit any 1st or 2nd level character. The high damage output in pathfinder seems to be an attempt to keep this playstyle at high levels.

The solution is dead easy, simply double the hit points of every PC and NPC in your campaign. Double any healing effects as well.

However, I'm not sure the glass cannon syndrome is as bad as you think. It actually leads to more tactical gameplay. A deadlier combat arena leads to more realistic PC behavior. When the PCs are safe from harm, they waltz into every room with no thought to tactics, traps, or preparation.

The bottom line is the DM just has to do a better job of scaling the encounters to the party, its ridiculous to expect the AP to know what your party is capable of. Don't stick so closely to the book. Right now we are playing through a campaign written for 4 players, but we have 8 PCs. You just have to make changes. If you really can't make changes, then maybe try skipping the party ahead to a higher CR section.