Lurker Underneath's page

I think basically the game needs to be streamlined and simplified in a lot of areas, some things need to be sorted out, but the game also needs to be more open in other ways.
- There should be more options that aren't locked away in feat trees. For example, combat options like a lesser version of Power Attack.
- Action types need to be relatively balanced among classes. One of the big things for some classes is that they never get anything useful for their swift actions, for example.
- There should be fewer categories of modifiers, and modifiers from spells and such should be designed in such a way as to either last for a while or to be strong but very short-term. Overall, there needs to be a smaller gap between what an unbuffed group is capable of and one that is buffed to the teeth.
- There should be fewer resources and uses per day to track. Give classes point-based pools for greater flexibility (but probably fewer overall uses).
- The feat trees should be pruned and streamlined. Don't need dozens of different feats that just give increasing bonuses to the same thing or fighting style. That just locks characters in to be one-trick ponies.
- Spells should just have a static effect, not have tons of variables scale by caster level. And variables should be meaningful on typical game scales. Have them scale by spell level instead, and make the basic metamagic effects available just by increasing the spell level, while the metamagic feats provide something else. A lot of individual spells need additional tweaking either for balance or to not be dull.
- Get rid of piddly stuff like traits and maybe favored class bonuses.
- Get rid of a lot of the circumstantial stuff like +1 to hit vs goblins on a moonlit night while riding a walrus and wielding a torch in your off hand.
- Martial classes need more options at higher levels. They need flexibility, and done in a way that doesn't involve getting locked down spending most of their feats on a single weapon. Something similar to the stamina system would be a good start.
- Archetypes should be a standard feature, and ideally implemented in a way that causes less bloat. There should be some thinking done as to what goes in the archetype box, and what goes in the feat box.

Lurker Underneath wrote:

The group as a whole will benefit much more from critical feats if their effects happen more often as well. So that should really encourage other players to say "hey, if you're going to use critical feats, don't use axes; scimitars are better for the group."

It's not necessarily a bad thing that some options favor some weapons over others, as long as there are other options that favor the other weapons more.

In that case, do you think that feats like Power Attack, Cleave, etc., which seem to favor weapons with a higher damage die and/or high critical multiplier, serve to balance the "critical status effect" feats? Or are more/different feats needed?

Those options aren't really more favorable to one type of critical threat range than the other. Power Attack adds the same damage on average, for longswords and battleaxes. It could be possibly be argued that old school 3e Cleave might be slightly better with a ×3 crit, as that would perhaps slightly increase the odds of one-shotting a foe, yielding an additional attack. However, the opposite could also be argued; outputting additional damage more often would increase the odds of that damage occurring at a time when a foe was low enough that a critical hit for double damage would suffice to bring him down.

What I would rather see is that options were designed taking this balance into mind to begin with. The critical feats could either feature some element that would scale with the multiplier, or they could be made to always trigger off a set roll, regardless of the threat range (i.e. they would only trigger on a natural 20, 19-20 or something like that, regardless of their critical threat range). That would make those feats more agnostic with regards to what weapons the wielder chooses.

For a feature like the fighter's weapon mastery that increases the critical multiplier, I would rather see it apply proportionally (though the ×7 criticals generated with scythes would be a bit much) or they could state that the critical multiplier only increases by +1 on a natural roll of 20 (19-20 with Improved Critical, or 18-20 with Improved Critical if the base threat range is 18-20). The latter solution might be a bit clunky, but not much more so than the current solution, and making for variability in how powerful the critical hits you can score isn't necessarily opposed to the system, even if there's not much precedence for it. I think most people could be trusted to keep more than two distinct types of outcome clear.

As I stated, I don't think it's a big issue. It's just that it's not elegant design as far as I'm concerned.

I actually don't consider most of those feats - the ones that cause status effects - to really be part of the equation regarding the question of how much criticals balance weapons. This is because these apply a penalty to an enemy that all of the PCs benefit from, regardless of what weapon they're using.

Yes, all the PCs benefit. However, my point is that whereas the crit system as it stood in 3e/3.5 is generally balanced in terms of average damage (i.e. 19-20/×2 produces as much average damage as 20/×3, making a battleaxe and longsword similarly viable), many of the new feats and features introduced by Paizo are set up so that they favor one type of weapon over another. If a character wants to use the critical feats, he's much, much better off with a longsword or scimitar than with a battleaxe, since the effects would occur 2-3 times as often, and be the same effects anyway. This means that wielding those weapons is a much better choice if you plan to take those feats. And those weapons are also a better choice because of how the end-game fighter weapon mastery ability turns out (17-20/×3 with a longsword produces more damage on average than the 19-20/×4 of a battleaxe).

The group as a whole will benefit much more from critical feats if their effects happen more often as well. So that should really encourage other players to say "hey, if you're going to use critical feats, don't use axes; scimitars are better for the group."

It's not necessarily a bad thing that some options favor some weapons over others, as long as there are other options that favor the other weapons more. It's not even a big issue; mostly one that deals with elegance in design. I think Paizo has done some potentially elegant things, but in other areas, what they've done is pretty clunky. For some reason, effects related to critical hits tend to be clunky and not mathematically sound in precisely the kind of equality-based ways SKR was ranting about way back when.

In terms of crunching the numbers, this is demonstrably not so. Is there some other way in which it could be considered too powerful?

I don't think it's too powerful per se; I allow keen and Improved Critical to stack in my own campaigns.

However, in terms of analyzing numbers, just looking at averages overlooks the impact of the higher values when they occur. A critical for ×4 damage has a much higher potential to swing the outcome of a fight than two for ×2. In terms of game dynamics, elements like this works against players. Since player characters are usually favored to win, more random elements in fights are to their disfavor.

I think this is also why they opted to change the default orc weapon from a battleaxe that did ×3 on a crit to falchions.

I don't necessarily think that's a legitimate reason for nerfing critical stacking either, but it's definitely more destabilizing for a game when used by NPCs than by PCs. PCs winning due to criticals is fine, but getting randomly murdered by criticals isn't necessarily a fun time. (And given that PCs will get a lot of attacks against them, it will happen. Only with less critical threats it happens a little less often.)

Ultimately, I don't think the nerfing took place because of balance concerns as much as perhaps a feeling that it devalued the impact of criticals if they occurred too frequently. As SKR's rant alludes to. If crunching numbers and making for numeric balance had truly been an overriding concern, the changes made in 3.5 should have looked different overall.

Of course, Paizo's record with regards to feats involving criticals is also annoying. In spite of SKR's rant about numbers, the various critical feats favor weapons with high crit ranges over those with high multipliers. Similarly, effects like the fighter's weapon mastery feature, which increases the critical multiplier by +1 favors weapons with higher threat ranges over those with higher multipliers. While the idea may be to avoid excessive swinginess in battle, the fact that they work the way they do means that overall balance between weapon types for certain characters is actually less under Pathfinder rules in the long run than they were under 3.5 rules. A 20th-level fighter with a keen longsword and a critical feat will do better on average than a 20th-level fighter with a keen battleaxe and a critical feat. Not to mention that scimitars or other 18-20 weapons benefit most of all from those combinations. It depends on which end you view the system from (it's not a low-level issue), but overall, balance between weapon types has gotten worse due to issue relating to critical hits rather than better under Pathfinder.

Kthulhu wrote:

Heladriell wrote:

I would really like a book to support levels 21 to 50 (and beyond). I think Paizo has all that is needed to make a working epic system.

Level 50, working, in a d20 system? * snorts *

As for me, I've seen level 50 play in 3.0 work, and I don't doubt Paizo is capable of making a better system. You need players with a minimum of maturity and willingness to cooperate with the DM (and a DM with the same), but then, immature and uncooperative players are quite capable of messing up the game at level 1 as well.

There are two different considerations there: The game as a social activity between players, and the game as a purely mechanical thing, built around the game statistics.

The game actually breaks in a mechanical sense at higher levels, once bonuses and values start to diverge sufficiently. It can still be playable given the right group of players, as long as they agree not to break it, or play in such a fashion as to not be too abusive of certain things; in other words, a social contract to minimize the effects of mechanics coming apart. The mechanical elements can also be band-aid fixed with some house rules, magic item distribution and judicious GM fudging.

It's sort of similar to how we are all here despite the existence of nuclear weapons. There's basically an agreement not to break the game by blowing each other up too much. This is usually the case in games too. People generally share a mutual interest in the game going along for as long as it has to, so they find ways to cope. That doesn't mean the system itself doesn't break apart, even without going into the kinds of terribly cheesy things you could do with the 3.0 epic rules as they were written.

There's also the matter of observer bias. What some people think "works" might be an awful state of affairs for others.

I'd prefer that they maintain a reasonable degree of backwards compatibility in terms of core mechanics while improving on a few issues of playability, balance, and overall design.

- Tidier presentation of mechanics. More concise and precise use of rules language and better layout of the books in some cases. Walls of words can be tough to get through, especially if the writer isn't using mechanical terms with sufficient accuracy.
- Rework a lot of things in the system with a focus on playability and taking less work to play out. Particularly at higher levels, the math gets absurd, and bogs down the game. Character sheet design that takes into account all potential bonus types and that "flows" the values in a natural pattern would help playability a bit.
- The math behind the system should get looked at in a few places. Saving throws should all scale by the same value. Give characters with good saves a +2 bonus to that value, and make all saving throws scale by half level. Everything doesn't need to scale by the same amount, but there should be a standardized way of integrating mechanics that scale differently (i.e. skill checks vs saving throws; maybe half the value of a skill check + 5 as a saving throw DC would be a good starting point).
- Iterative attacks in their current form should go. When characters get multiple attacks, it should follow the same model as Rapid Shot; more attacks, all at the same penalty.
- Short-term effects and conditions should never modify ability scores; it can get tedious to rework all the modifiers.
- Rework the content of spells and special abilities so that specific types or levels of effects start becoming available only at specific levels, and hardwire that into the system. I.e. only spells of level X or above can inflict conditions that paralyze or hold, only spells of level Y or greater can inflict death effects. Make counters of some kind available to all characters prior to the effect itself.
- Spells shouldn't trump the options available to other characters in their own niches, unless the spellcaster pays a price to be able to do so. For example, a knock spell should require a check.
- Spellcasters should have more limited access to options in general. Limiting spells known per level for divine casters and wizards would work (and would be in line with editions prior to 3e). Another option would be to separate basic spells from "rare" spells, and make the rare spells require some kind of buy-in before they can be accessed. Maybe similar to how cleric domains work, but for all casters, and the selection being expandable through feats or the like. Makes for more flavorful spellcasters too, rather than generic do-all wizards. Some may be able to summon monsters, some can turn invisible, and some can fly, but very few can do all of it (and if they can, they'll have other weaknesses).
- Design that makes for more basic options for characters and fewer potty training feats. If a new mechanic is something all character should reasonably be able to do, it should be a new basic option, not a feat. Feats can be improved versions thereof, not required to do something that seems pretty basic. Overall: Give non-spellcasters more options to play with, and fewer feat chains they need to work their way up through to do cool stuff.
- Along the lines of more basic options, there should be fewer sub-optimal and required auto-select options, and there should be fewer options which tie characters down to only one weapon or similar very specific area of expertise.
- The skill system needs to be expanded upon a bit. Add more basic options for skills, and add better mechanics, particularly for social skills. Make skills more useful, particularly at higher levels. The solution to options involving skills isn't to make the skill modifiers ever higher (i.e. what was done with Skill Focus and the like) but to make skills more useful overall. If Craft checks were required to craft magic items, that would make it more valuable to player characters, for instance.

There's a lot that can be done to polish the current system while essentially remaining the same game.

With recent discussions of caster power within Pathfinder, I found myself contemplating ways to somewhat downplay their power without completely nerfing them. Anyways, a thought from older editions came to me.

I seem to recall, waaaay back in I think 1st edition, that casting a spell required a certain number of "segments" -- and that as initiative counted down, a caster began a spell when his or her initiative came up, but didn't finish said spell until the casting time had been satisfied.

This meant that a caster who rolled high on initiative could almost certainly get off a low-level spell, but that a higher-level spell ran the risk of being interrupted as opponents went "Oh crap, a caster!" and directed hostilities towards them as said opponents' initiative counts came up.

A simple way to adjudicate this might be to have any spell take one "segment" / initiative count to cast per spell level. Spells which display a casting time of one swift action or one immediate action could be cast whenever, and would always go off on the caster's rolled initiative count.

A caster could take a move action before starting to cast the chosen spell, or after he/she finished casting. This would, of course, involve complications on the initiative count though.

Thoughts?

A few considerations: Initiative in 3e/Pathfinder is cyclic. In 1e/2e, it was rerolled each round. This meant you could never be certain of the order you would go in, and thus, casting a spell with a high casting time ran a high risk of disruption.

However, with cyclic initiative, information does not get to be so difficult to come by. Depending on how things are resolved (how much info the GM withholds), no later than after the first round, a spellcaster should have sufficient information to game a "tick-based" approach. For example, if the caster rolls a 21 for initiative, he knows his ally has a 16, and no enemy acts before that, the caster would know all spells with a casting time of "-5" or less would be safe to cast on his next turn -- thus making the issue of the casting time for such spells non-existent for that battle. (Enemies would then have to ready or something to disrupt him, as normal.) Conversely, given an enemy acting at 20, that enemy would be able to disrupt any spell that's cast at "-1" or more -- which would be just about any spell above level 0, if you use 1/spell level. Unlike 1e/2e, that situation would remain static for the rest of the battle. Unless the caster chooses to delay. By delaying until everyone else has acted (or just until a "safe spot" in the initiative order), the incremental count becomes a non-factor. If there's a space of five points in the initiative order, there's no practical difference between a casting time of 1 or 4.

Essentially, counting increments like this introduces an additional complication that would sometimes simply not matter at all due to how initiative spreads and the cyclic initiative work. It would only work against casters at random battles. It would also be within the ability of casters to control -- by taking initiative-boosting options and by delaying. Whereas with rerolling initiative each round, and having to declare your action ahead of that, it was always a consideration when choosing a spell to cast in 1e/2e -- the balance lay in managing that risk. With cyclic initiative, that does not work.

The way to implement the same concept under the current set of mechanics would be to move casting times up the action hierarchy: Troublesome save-or-suck and save-or-die type spells should be given casting times of 1 round (completed at the start of the caster's next turn). That would yield completely predictable results each and every time (it would either always be a penalty or never), and doesn't require making any modifications to the existing system. Simple damaging spells and the like aren't generally problematic, so they can safely remain standard actions.

Another point to this is that it doesn't require tracking additional information. Having to essentially track two initiative counts for each caster in a fight, one of which might vary from round to round is additional work for the GM -- for something that may not even accomplish its intended purpose.

On the other hand, if you really want to use a numeric casting time mechanism, that would probably require a more thorough revision of the entire initiative system. Either making all actions incremental, or using the 1e/2e initiative standard of rerolling each round, after declaring actions for the turn. And both would be more cumbersome than the cyclical system.