Non-AC Defenses

[eamon]

When analyzing these things, it's worth keeping in mind that an improvement to a low defense matters less than an improvement to a high defense. Roughly speaking, raising your defense by 10 from hit-on-2+ to hit-on-12+ halves your damage and thus doubles your life expectancy. Then raising your defenses by just 4 or 5 again halves your damage and doubles your life expectancy.

Assuming a monster crit is about 50% more powerful than a hit, eventually, raising your defense from hit-on 19 [expected damage: 5% * a hit + 5% a crit ~ 12.5% a hit] to hit on twenty [7.5% of a hit] is a 40% damage reduction for just one point. Raising it to miss-on twenty [5% of a hit] decreases damage taken by another 33% for one point of defense.

Basically, it just doesn't matter all that much whether a monster hits you on a 4 or a 7 - and it matters even less whether he'll hit you on a 2 or a 3, or, worse yet, on a 1(effectively 2) or a 2 which is truly no change.

In short, there's no point in bothering with the lowest defense unless you can somehow manage to raise it by enough to make it competitive.

The House Rule
What does this mean about the house-rule which raises FRW at 5/15/25? Well, that house rule is good and works well for your strongest two defenses, which thus remain competitive. It doesn't hurt your weakest defense either, but it won't be significant; that would require even further raises.


Fixing the lowest defense
There's three choices here. (1) don't fix it. (2) use feats. (3) use stat raises. WotC basically does (1). The feat fixes (without above house-rule) are needed to just keep the highest defenses competitive - they're not enough for the lowest defense, and in fact, a player would be wasting a feat if he tried raising his lowest defense with such a feat (so the feats are both overpowered and underpowered simultaneously - great).

Using (2) is hard - and some might object to a feat tax. It's hard because any feat you make will be much much more attractive to use on the high defenses - where such a feat would be overpowered. I don't think that's a good idea.

Using stat raises is comparatively easy; just raise all stats on levels 4/8 and (in combination with the general 5/15/25 raise) players will lose only 1 point vs. the monsters over their career, which looks OK to me.

Such house rules have other nice properties as well. Earlier in this thread, there was some dislike of 18/14/11 builds. Well, if you raise all three stats, all of the sudden your other stats actually have a chance of remaining relevant. After the change, it might actually be worth it to consider raising a tertiary stat to help your weakest defense (and skills, and maybe init or hitpoints). Right now, stat levelling rules basically mean that you never ever should bother investing in non-core stats unless you absolutely need to for feat prereqs. That's pretty lame, and it's a nice side effect that that's no longer the case when you raise all stats. I'm also not too thrilled about the way feat-prerequisites really punish people that don't pre-plan their character 20 levels in advance. Getting rid of those stat prerequisites is probably a good thing - unless it breaks game balance (I can't, off the top of my head, think of any examples thereof, however).

It also ameliorates the need for "fixer" feats like heavy armor for Str/Wis Rangers (still attractive, but no longer absolutely required) or like the primal con-to-AC feat (also still attractive for a non-Dex/Int build, but again, not absolutely required).

That's a bunch of nice side effects, IMNSHO .

 

[UngeheuerLich]

Yeah, maybe the 3rd stat increase is not that bad at all. But something has to be done to make monsters hit harder then.

Maybe add another damage dice at paragon and another one at epic r something along that line. (or just add half level to damage)

what is the profit? Less status effects, but more damage. So fights maybe remain more dynamic.
And suddenly V-shaped could become more attractive than A-shaped classes.

 

[KarinsDad]

Yeah, maybe the 3rd stat increase is not that bad at all. But something has to be done to make monsters hit harder then.

Maybe add another damage dice at paragon and another one at epic r something along that line. (or just add half level to damage)

what is the profit? Less status effects, but more damage. So fights maybe remain more dynamic.
And suddenly V-shaped could become more attractive than A-shaped classes.

Interesting. I did exactly that in my house rules. Increase 3rd stat each time. +1 NADs at 5/15/25. And increase Paragon monsters by one die and Epic monsters by two dice (and change D20 damage dice to 2D10 instead).

 

[keterys]

When analyzing these things, it's worth keeping in mind that an improvement to a low defense matters less than an improvement to a high defense. ...

The entire premise is flawed under the assumption that all you cared about is damage.

Almost invariably status effects are the things I'm far more worried about. At which point going from a 2 to 7 instead of a 10 to a 15 is still all good

The suggested house rule seems fine though.

 

[UngeheuerLich]

Interesting. I did exactly that in my house rules. Increase 3rd stat each time. +1 NADs at 5/15/25. And increase Paragon monsters by one die and Epic monsters by two dice (and change D20 damage dice to 2D10 instead).

Does it work out good?

(note, that i have yet to see if theorie is as bad in actual game, as it sounds on paper... but it is nice to have an ad hoc fix, which works)

 

[eamon]

The entire premise is flawed under the assumption that all you cared about is damage.

Almost invariably status effects are the things I'm far more worried about. At which point going from a 2 to 7 instead of a 10 to a 15 is still all good

The basis of the analysis is the observation that life expectancy is a useful metric for comparing situations (for instance, it also permits comparing outgoing DPR with defenses quantitively).

All worthwhile combat effects I can think of impact this; and generally the results are the same as for straight damage: the difference between having a 90% or 95% chance of being effected is trivial, but the difference between having a 5% or 10% chance is not. The absolute difference for one round is the same - but of course, you don't play combat for one round, you play it for several rounds, and in particular, if you're almost always hit, you'll die faster and thus the same absolute difference adds up fewer times.

D&D combat, in extremely abstracted form, resembles a negative binomial distribution. Whether you're pushing someone over a cliff, dealing damage, reducing his defenses to later have more impact, or dazing him to give yourself more time to kill him; effectively you're in a race against the clock: can you kill them before they kill you?

Effects matter, but they don't change the fundamental principles underlying the defenses: small changes to low defenses have less impact than small changes to high defenses.

 

[eamon]

Yeah, maybe the 3rd stat increase is not that bad at all. But something has to be done to make monsters hit harder then.

Maybe add another damage dice at paragon and another one at epic r something along that line. (or just add half level to damage)

That's not a bad idea; but it's pretty orthogonal to the issue of defenses. Currently, FRW are not commonly targeted; and the fix does raise them to levels where monster damage (or effects) will be greatly impacted - after all, most damage still comes attacks vs. AC, and even after the fix FRW defenses won't generally be high, they just won't be trivially hit.

To avoid grind, more damage may be useful. On the other hand, raising defenses fixes something else; namely the dramatic imbalance between attacks vs. AC and attacks vs. FRW. Rather, fixing FRW defenses may well also reduce grind by avoiding situations where mono-thematic encounters happen to have single effects that are virtually impossible to shake (i.e. parties that are dazed almost nonstop because they're always hit by dazing controller powers).

So let's not mix up these two discussions. You can add extra damage even without fixing defenses, and you can fix the FRW defenses without adding extra damage; both work independently. They also both have independently complex side effects (i.e. extra damage reduces the effectiveness of leaders, but raising all stats changes class build balance), so confouding them makes it trickier to evaluate the consequences.

In terms of stat raises, I'd strongly be in favor of raising all stats rather than just three as a fix. Raising only three stats doesn't fix the entire problem (witness classes with aligned primary and secondary stats), and it does fix divergence of things like skills and init nor the requirement for extensive char-planning (which 4e was supposed to reduce). On the other hand, raising all stats hardly increases the problematic aspects of the fix; these being the possibility of imbalances made possible by avoiding feat prereq's (after all, even adding just one extra stat raise is probably enough to make 99% of all stat prereqs easily achieveable with planning). So, if you want a stat-based fix, I'd go whole hog and happily get rid of a bunch of problems related to stat modifier divergence by doing so, rather than just fix the weakest defense in some of the cases.

 

[UngeheuerLich]

Nope,

you don´t see the point.

You just have to compensate for monsters hitting less. Damage up is the preferred method. Actually when you compare xp, you could drop attack by 4 and double all damage.

edit: just to be clear: it has nothing to do with fixing anything.

edit2: of course it has the side effect that you are less often hit with status effects.

edit3: not been clear enough: my comment was based on the assumption that there is no need for fixing anything, if the DM doesn´t overuse too many status inflicting monsters.

 

[KarinsDad]

Does it work out good?

(note, that i have yet to see if theorie is as bad in actual game, as it sounds on paper... but it is nice to have an ad hoc fix, which works)

I'll find out next weekend. We wanted to get the PCs all the way up to 30th level, unfortunately, we are having scheduling issues. So, we decided to do 4-5 sessions at level 7, then take some weeks off, then 4-5 sessions at level 10, then take some weeks off, then 4-5 sessions at level 13, etc. Each time, we increase it by 3 levels.

Next weekend, the PCs start at level 10. So, they will face Paragon level foes in some cases, even though they themselves are not Paragon level yet. This will be a fairly crucial test since the PCs themselves do not get the 11th level boost (extra powers, special abilities, feat, and ability score increases) that helps against Paragon level.

 

[keterys]

The basis of the analysis is the observation that life expectancy is a useful metric for comparing situations

And I'm saying that players can cope with getting damaged. In fact, not being damaged is potentially frustrating sometimes.

But they _hate_ being permanently dazed, stunned, or whatever. So I don't think the DPR argument is the critical point of the discussion and every point, with the exception of the extremes, is just as relevant.

All worthwhile combat effects I can think of impact this; and generally the results are the same as for straight damage: the difference between having a 90% or 95% chance of being effected is trivial, but the difference between having a 5% or 10% chance is not.

Doubling the chance of not being affected vs. doubling the chance of being affected. Tomato, tomato.

The +/- 1 is the same. The difference is that you _already_ feel like you're being autohit or the DM feels like you're auto-missed. But when the 2 actually misses you, you still go 'Holy crap, you missed!?' and feel great, and when the 19 actually hits 'The DM goes, hey, wait, does a 35 actually hit you? Holy crap, I hit'

D&D combat, in extremely abstracted form, resembles a negative binomial distribution. Whether you're pushing someone over a cliff, dealing damage, reducing his defenses to later have more impact, or dazing him to give yourself more time to kill him; effectively you're in a race against the clock: can you kill them before they kill you?

Much of the combats in dnd really have no great capacity for the entire group to die, and it's actually far more a competition for 'How fast can you do this?' 'What cool stuff can you do on the way?' 'Do I manage to make you blow daily or consumable resources, or maybe kill one guy?'. There are, of course, the full on everyone might die combats, but it's very important to make sure as many of the possible combats are fun along the way.

Effects matter, but they don't change the fundamental principles underlying the defenses: small changes to low defenses have less impact than small changes to high defenses.

No, they don't. From an absolute standpoint they are identical (+/-5%). From a relative standpoint, small additions to low defenses mean you're not affected a greater percentage of the time (+1 = +100% more misses) and small subtractions to high defenses mean you are affected a greater percentage of the time (-1 = +100% more hits). Vice versa if you flip the bit. So it's all a matter of perspective for what you're looking for. Which means absolute is the most useful for discussing effects.

For DPR, sure, and that certainly does matter, but there's all kinds of things that impact on that - miss damage, extra crit, does its attack carry a defense penalty or grant CA, ongoing damage (which doesn't stack with repeat hits and failed saves), vulnerability save ends or 1 turn (where repeat hits are more important).