|
| 6 people marked this as a favorite. |
I'm not big on theorycrafting, in general, but I like benchmarks. Benchmarks help me to understand things and can be used to inform discussions about things. On the FB group, I got deep into a discussion of monks and armor class was being discussed. Out of curiosity, I ran some numbers and it ended up becoming a more robust analysis.
TL;DR version: The more armor you wear, the better off you are. The overall range of ACs is fairly small. Shields can help a lot and push critical range more meaningfully than being-hit range.
I ranked Dexterity in four grades. For analysis, I've assumed that PCs will maximize their dexterity to the extent that it benefits their AC. In practice, this is the number that will vary the most among players, though it probably won't vary by more than a couple of points.
A: A class with Dexterity as a Key Ability and the intention of pushing it as high as possible
B: A class without Dexterity as a Key Ability that still intends to push it as high as possible
C: A class intending to use medium armor with no other main use for Dexterity
D: A class intending to use heavy armor with no other main use for Dexterity
There are only three classes that get any special training in defense, though two of those have two different progressions for different types of armor (well, fighters technically have three). It is beyond the scope of this analysis to cover characters that gain proficiency through anything other than base class features (such as the Fighter Dedication feat).
A: Most classes with no special increases beyond Trained.
B: Fighters with heavy armor (E at 11, M at 15)
C: Fighters with medium armor (E at 15)
D: Monks without armor (E at 1, M at 13, L at 17)
E: Paladins with heavy armor (E at 7, M at 13, L at 17)
F: Paladins with light or medium armor (E at 13, M at 17)
In all cases, I assume that the PC will obtain an item when their character level is one greater than its item level. There will obviously be some variation in this in play, but it feels like a reasonable assumption. Mage Armor (as a heightened spell) is presented as an option, but the steep cost of using this option makes it unlikely that most characters would want to go that rotue.
B: Bracers of Armor
S: Mage Armor Spell (fully heightened)
L: Light Armor (Studded Leather or Chain Shirt, 5 max dex)
M: Medium Armor (Breastplate or Chainmail, but it is viable to switch to hide at level 5 by bumping Dex to compensate)
H: Heavy Armor (Half-Plate is the general consideration here, though full plate works the same)
For all classes, I'm assuming that the PC is not using a "bad defense" option. For instance, I'm ignoring the strength-based monk and the unarmored barbarian in this analysis. You can fairly easily determine what their progression would look like by taking the closest progression and subtracting 1-3 points from it. I also include a column called The Worst, which represents a Dexterity 10 character that doesn't wear armor, which is most useful to show the consequences on the to-be-hit and to-be-crit charts.
Alchemists: B/A/L
Barbarian: B/A/L or C/A/M
Bard: B/A/L
Cleric: B/A/L or C/A/M
Druid: B/A/L (assumed Studded Leather is non-metal) or C*/A/M* (Assumed Dex bump at level 5)
Fighter: A/A/L or C/C/M or D/B/H
Monk: A/D/B
Paladin: B/F/L or C/F/M or D/E/H
Ranger: A/A/L or C/A/M
Rogue: A/A/L
Sorcerer: B/A/B or B/A/S
Wizard: B/A/B or B/A/S
AC: https://i.imgur.com/nDQCxRf.jpg
To-Be-Hit: https://i.imgur.com/yWXTf1p.jpg
To-Be-Crit: https://i.imgur.com/J5ybDfE.jpg
Among armored characters (and monks), the AC divide is about two points at low levels and three at higher levels. Shields widen this gap to four and five, respectively. In play, I would expect this gap to be as much as two points wider, though I suspect a single point is more likely (which would account for a 16 on a class with Dexterity as a Key Ability or a 14 on anyone else).
In terms of being hit, most characters will get hit 60-75% of the time by an of-level monster's first attack in a round. Shields mitigate this somewhat, but overall hit rates go up slightly as levels increase with occasional weird hiccups (probably caused by the limited dataset for monsters).
Critical hits were the most interesting chart for me because it showed one thing very clearly. If you don't want to be crit, raise a shield.
Obviously, there are some pretty big limitations in this analysis. I didn't consider absolutely every case that could be out there (the wizard with the Fighter Dedication or the Animal Barbarian fighting in a loincloth or the strength-based monk are notable examples). I also don't have numbers that conclusively say what Dexterity ends up being in play. Finally, the hit and crit charts relied on data collected on this forum by LuniasM which can only be as good as the monsters we currently have to go from.
Please note any errors in assumptions or math, and I'll discuss and/or correct them.
This is legit, nice job!
Lines up with what I thought, you need to keep AC at max to just not get killed by enemies.
I'd love to see this for attacks and how that looks.
Shields and other circumstance AC bonuses may not be something you want in the base analysis. Shields stack with cover, screening, prone, crane style, giant bane, parry(and variants), shield(and variants), deflect arrows, nimble dodge, and I think that's it. There's probably a spell or two.
So it may be or accurate to say; if you don't want to be crit, get to cover or carry a shield.
The lack of higher proficiencies in light and medium armor is annoying. I'm sure its a balancing point somewhere, but I quite like lighter armors and would prefer there be some way to be really good with them.
| 1 person marked this as a favorite. |
You're missing the +2 DEX @14th level from Anklets of Alacrity, which would apply to any DEX-maximizing class that doesn't hit a max dex cap. That'll bump up monks a little at higher levels.
Additionally, people not using light,med,or heavy armor can still raise a shield (particularly relevant is the shield cantrip).
Otherwise very nice analysis. High level combat looks like its going to be a crit-fest vs. players, as even perfectly optimized characters will be crit 20% of the time or more on the first attack delivered by a boss enemy. Ouch!
|
|
| 1 person marked this as a favorite. |
You are basing your calculations on even-level enemies, but 75% of the time you'll be fighting enemies at least a couple levels below yourself.
I hear that said a lot, but my main playtest experience has been running Doomsday Dawn where there tend to be several equal level and usually a level +2 or +3 critter in each scenario. It's those fights that usually end up with dead PCs, which makes them the ones more interesting from a design standpoint. If we're talking about number of encounters, I don't think the number is anywhere near 75% of the time that the highest level monster is of lower level to the PCs.
As for whether the analysis holds up... sort of? To actually do that analysis is another thing. You need to look at each class and each armor case, individually. It probably isn't useful when discussing between classes, which is how this chart started out.
However, it's a good question, and one I spent some time on just now. The answer seems to be that the numbers will change by 5-10% between each level. It takes a whole chart to do one case, though, so I only did two:
Paladin in Heavy Armor to be Hit
Paladin in Heavy Armor to be Crit
Rogue in Light Armor to be Hit
Rogue in Light Armor to be Crit
(Note: I didn't actually look up the rules on the bottom end of crits. I assumed it was that a 20 always hits but only crits if it also succeeds, but I didn't confirm that.)
Shields and other circumstance AC bonuses may not be something you want in the base analysis. Shields stack with cover, screening, prone, crane style, giant bane, parry(and variants), shield(and variants), deflect arrows, nimble dodge, and I think that's it. There's probably a spell or two.
They weren't part of the initial analysis, but I ended up putting them in here because they're more reliable and less situational than other things. The way they changed the crit numbers was very interesting, too, and prompted me to leave them there, but it's also why I left them pulled out as their own thing in separate sections. A criticism that I'll agree with is that the analysis column on the right side of the AC chart should have an "Armor Only" and "Armor + Shields" column. Shields somewhat skew the answers to the current one and you have to mentally back them out when looking at the delta.
So it may be or accurate to say; if you don't want to be crit, get to cover or carry a shield.
That's a fair statement, but the availability of cover isn't assured. There are several fights in Doomsday Dawn where a melee character cannot expect to find cover, for instance. There are a few where no one can find cover.
The lack of higher proficiencies in light and medium armor is annoying. I'm sure its a balancing point somewhere, but I quite like lighter armors and would prefer there be some way to be really good with them.
I've come to the conclusion that the "default" case should be hide armor. Hide armor is the highest armor bonus you can get without any kind of movement penalty and it doesn't require a dexterity higher than 18 to get the full benefit from it. Armor that gives you more penalties than hide should provide better statistics in some way. Armor that provides fewer penalties than hide (either because it requires a lower grade proficiency or has a lower check penalty) provides less. I'm very fine with heavy armor getting you a couple of points of AC over the guy in hide.
You're missing the +2 DEX @14th level from Anklets of Alacrity, which would apply to any DEX-maximizing class that doesn't hit a max dex cap. That'll bump up monks a little at higher levels.
I discarded it in my initial analysis because under the (item level +1) rule it doesn't show up until the other classes were already capping dex to armor. I'd feel comfortable adding it to the monk, I think, but that's the only class that would be likely to reliably take it and have it affect their AC.
Additionally, people not using light,med,or heavy armor can still raise a shield (particularly relevant is the shield cantrip).
That's true, but I had to draw the line on cases somewhere ;)
|
|
(And for reference, a natural 20 on a Strike is a critical hit regardless of AC)
You made me go look it up... and I don't think that's right.
However, if you succeed and rolled a 20 on the die (often called a “natural 20”), or if your result is equal to or greater than the DC plus 10, you critically succeed.
...
If your enemy is far more powerful than you or a task beyond your abilities, you might roll a natural 20 and still get a result lower than the DC. In this case, you succeed instead of critically succeed or fail. If you lack the proficiency for a task in the first place, or it’s impossible, you might still fail on a natural 20.
Unless I'm missing something, if you roll a 20 and it's not enough to hit (but you are proficient in your attack), you just get a regular hit.
That's a fair statement, but the availability of cover isn't assured. There are several fights in Doomsday Dawn where a melee character cannot expect to find cover, for instance. There are a few where no one can find cover.
That's a little disappointing. Still, screening should be readily available through ally positioning, parrying weapons are common enough, monk and rogue have fairly reliable effects like crane wing and nimble dodge. There's also the ranger dual parry feats.
But your work here is great and the charts are clear enough that I can just scan over to the new number. I shouldn't nitpick. Thanks for the work.
|
|
After some consideration and discussion, I went back and added an additional +2 to DEX at level 15 (where the DEX boosting item could reasonably be obtained). This had no effect on anything except for the monk.
As was noted in the Facebook group, this chart does not include the potential for the level 20 Bracers or Armor or the use of Parrying Weapons /Stances which could conceivably push the Monk to the same level as the Paladin with shield. I think the base numbers are just fine as they are (and even including shields goes a bit far).
Armor Classes: https://i.imgur.com/e7XDWSi.jpg
To Be Hit: https://i.imgur.com/3cqYVFy.jpg
To Be Crit: https://i.imgur.com/5TguOrv.jpg