Basic Training 101: The Attack Table, Mists of Pandaria Ed.

Updated for Mists of Pandaria (Patch 5.*)


This is a quick "Beginner Theorycraft" article to explain the basics of the Attack Table to new players. I began with Theck's pre-5.0.4 tutorial, and updated it to include the changes to the attack table introduced in patch 5.0.4.

This is designed to be a "tutorial," which means "readable by someone who's just getting their feet wet in tanking theory."

Any comments/questions/suggestions are welcome, especially those that further enhance the readability of the guide and the quality of information it contains. And things like telling me I got stuff backwards or totally botched numbers.

For a TL;DR version, scroll on down a couple posts, or go here

Basic Training 101: The Attack Table

To understand some of the gearing choices a paladin has to make, we first need to understand how WoW resolves combat events. The fundamental mechanic is something called the Attack Table.

Every time a player or boss attempts to make an attack, the game performs a series of two rolls to determine whether the attack succeeds or fails. The Attack Table is the scale which determines the results. To illustrate this, let's look at a few simple examples. First, let's consider a lvl 90 mob attacking an un-specced lvl 90 human paladin with base avoidance/block (3% miss, 3% dodge, 3% parry, 3% block). The attack table is then constructed like this:

Table 1

Code: Select all

00.01-03.00 = miss
03.01-06.00 = dodge
06.01-09.00 = parry
09.01-12.00 = crit (200% damage), roll on Table 2
12.01-100.0 = hit, roll on Table 2


Table 2

Code: Select all

00.01-03.00 = block (30% damage reduction - ignore critical blocks, since only warriors can do that)
03.01-100.0 = hit (100% damage or 200% damage if rolled due to crit)


the server then does a /roll 100 to determine the outcome (though it does this to more decimal places than a standard /roll 100). If it rolls a 12, you parry and no secondary roll is performed; a 16 is potentially a blocked hit, and a secondary roll is performed. If that secondary roll is 3 or less you block the attack, otherwise you are hit.

If instead we're considering a level 93 boss attacking the same un-specced level 90 paladin, we have to make a few modifications. For each level above the player's level, the NPC gains an additional 1% chance to crit the player and reduces the chance for his target to avoid or block an attack by 1.5%. Thus, we've now got a 6% chance to be crit, and our base chance to dodge, parry, block, and miss are now all 0% since you can't have a negative value.
This is another major difference from previous expansions, because it means that a level 92 (typical level 90 5-man boss) has the same chance to hit our hypothetical player as a level 93 raid boss. The resulting attack table looks like:

Table 1

Code: Select all

---------- = miss (0% chance for boss to miss)
---------- = dodge (0% chance to dodge the attack)
---------- = parry (0% chance to parry the attack)
0.01-06.0 = crit (200% damage), roll on Table 2
6.01-100.0 = hit, roll on Table 2


Table 2

Code: Select all

--------- = block (0% chance to block the attack)
00.01-100.0 = hit (100% damage or 200% damage if rolled due to crit)


Before moving forward, let's briefly discuss how the table is arranged. As you see, avoidance is at the very top of the first table followed by crits and normal hit chances, while the second determine if you block the attack or if it hits. Note that blocks and crits are on different tables - this means that you can block a crit.

As you increase the chance to do any one of these things, it "pushes" the things below it further down on the same table. As an example, if we let our example paladin chose the protection tree as his specialization, he will gain an additional 18% block chance (from our baseline mastery, and an additional 10% block chance from Guarded by the Light), a 6% reduction in critical strike chance (from Guarded by the Light) as well as an additional 2% dodge (from Sanctuary).

Table 1

Code: Select all

--------- = miss (0% chance for boss to miss)
0.01-2.00 = dodge (2% chance to dodge the attack)
2.00-2.00 = parry (0% chance to parry the attack)
--------- = crit, Roll on Table 2
2.01-100.0 = hit, Roll on Table 2


Table 2

Code: Select all

0.00-18.00 = block (18% chance to block the attack)
18.01-100.0 = hit (100% damage, no crits possible)


So now our level 90 protection paladin cannot be critically hit by a level 93 or lower mob. He also has a reduced chance to get hit due to increasing both dodge and block chances. Note however, just because we see an 18% block chance on table 2, this does not mean we should expect an overall 18% of attacks blocked in our combat logs.
Here's where it gets a bit more tricky: Because of the 2-table system, our overall chance to block is dependent on our overall chance to avoid the blow (miss + dodge + parry) since that is determined first. Therefore, our overall block chance looks more like (1-A)*B, where A is (miss+dodge+parry chance) and B is our block chance. For our example, we'd expect to see closer to 17.6% of attacks blocked.

How would this look if we had a bit more realistically geared lvl 90 protection paladin? Well, let's assume they've gotten geared up so they have 8% dodge, 17% parry and 25% block on their character sheet.
For a level 90 mob, their attack table would look like:
Table 1

Code: Select all

00.01-03.00 = miss
03.01-11.00 = dodge
11.01-28.00 = parry
----------- = crit, Roll on Table 2
28.01-100.0 = hit, Roll on Table 2


Table 2

Code: Select all

00.01-25.00 = block
25.01-100.0 = hit


So while they'd expect to parry and dodge the same percentage of attacks as on their character sheet (8% and 17% respectively), they'd only see about 18% ((0.72*0.25)*100%) of the incoming attacks being blocked.
This is a major change from previous expansions, in that now avoidance chance affects block chance.

Let's take a look at the attack table for our paladin against a level 93 boss:
Table 1

Code: Select all

----------- = miss
00.01-03.50 = dodge (8% - 4.5% from boss level difference, since we are above 4.5% dodge total)
03.51-16.00 = parry (similar reasons to above)
----------- = crit, Roll on Table 2
16.01-100.0 = hit, Roll on Table 2


Table 2

Code: Select all

00.01-20.50 = block (25% - 4.5% from boss level difference)
20.51-100.0 = hit


So now we'd expect to see 3.5% of attacks dodged, 12.5% parried, and 17.22% blocked. Compared to our naked paladin earlier, we actually see fewer blocked attacks despite having a higher block chance because we have a higher avoidance chance.
The net result of this is that the interaction of avoidance and block is more complex than in previous expansions. Additionally, the old ideas of pushing un-mitigated/un-avoided hits off the attack table by stacking the right mix of block and avoidance is essentially impossible (or at best, extremely unlikely). In order to do that, you'd either need 100% avoidance (not going to happen) or 100% block (extremely unlikely, given gear levels and diminishing returns being added to block chance)


2-Roll vs 3-Roll system
Another variation on the attack table system is used for some special attacks, in which crit vs. normal hit is determined by a third roll. So going back to our base level 90 player facing a level 90 mob, our table would look like:
Table 1

Code: Select all

00.01-03.00 = miss
03.01-06.00 = dodge
06.01-09.00 = parry
09.01-100.0 = hit, roll on Table 2


Table 2

Code: Select all

00.01-03.00 = crit (200% damage), Roll on Table 3
03.01-100.0 = hit (100% damage), roll on Table 3


Table 3

Code: Select all

00.01-03.00 = block (30% damage reduction - ignore critical blocks, since only warriors can do that)
03.01-100.0 = hit (100% damage or 200% if due to a crit)


And for our geared protection paladin, facing a level 93 boss the table would look more like:
Table 1

Code: Select all

----------- = miss
00.01-03.50 = dodge (8% - 4.5% from boss level difference, since we are above 4.5% dodge total)
03.51-16.00 = parry (similar reasons to above)
16.01-100.0 = hit, Roll on Table 2


Table 2

Code: Select all

----------- = crit, Roll on Table 3
00.01-100.0 = hit, Roll on Table 3


Table 3

Code: Select all

00.01-20.50 = block (25% - 4.5% from boss level difference)
20.51-100.0 = hit (100% damage or 200% if due to a crit)


Note that under the basic assumptions here (no modifying buffs/debuffs, assuming player-level + 3 or below mob, etc.), the 2 and 3 roll system become the same thing because the crit chance gets removed. However, if a debuff was introduced that say increased the critical hit chance on the player from 6% to 10%, there would be a difference (looking at the 90 protection paladin and 93 boss again):
2-Roll:
Table 1

Code: Select all

----------- = miss
00.01-03.50 = dodge (8% - 4.5% from boss level difference, since we are above 4.5% dodge total)
03.51-16.00 = parry (similar reasons to above)
16.01-20.00 = crit, Roll on Table 2
20.01-100.0 = hit, Roll on Table 2


Table 2

Code: Select all

00.01-20.50 = block (25% - 4.5% from boss level difference)
20.51-100.0 = hit (100% damage or 200% if due to a crit)


3-Roll:
Table 1

Code: Select all

----------- = miss
00.01-03.50 = dodge (8% - 4.5% from boss level difference, since we are above 4.5% dodge total)
03.51-16.00 = parry (similar reasons to above)
16.01-100.0 = hit, Roll on Table 2


Table 2

Code: Select all

00.01-04.00 = crit, Roll on Table 3
04.01-100.0 = hit, Roll on Table 3


Table 3

Code: Select all

00.01-20.50 = block (25% - 4.5% from boss level difference)
20.51-100.0 = hit (100% damage or 200% if due to a crit)


So in the 3-roll case, the overall critical hit chance would see the influence of dodge and parry but in the 2-roll case it would not (since in the 2-roll case avoidance would only reduce the chance of normal hits because they are the 'filler' in the table). To put some numbers to this, the 2-roll case would see an overall 4% critical chance, while the 3-roll would see 3.36% for the player considered above. If we bumped the player's avoidance (let's say dodge) up by 5% to 13% total, we'd see a 3-roll table like:
Table 1

Code: Select all

----------- = miss
00.01-08.50 = dodge (8% - 4.5% from boss level difference, since we are above 4.5% dodge total)
08.51-21.00 = parry (similar reasons to above)
21.01-100.0 = hit, Roll on Table 2


Table 2

Code: Select all

00.01-04.00 = crit, Roll on Table 3
04.01-100.0 = hit, Roll on Table 3


Table 3

Code: Select all

00.01-20.50 = block (25% - 4.5% from boss level difference)
20.51-100.0 = hit (100% damage or 200% if due to a crit)


Which would result in an overall critical hit chance of 3.16% (though some of those may only be 170% damage attacks, rather than 200% due to block chance).

Appendix: Crushing Blows
A crushing blow is an auto-attack that hits for 150% damage. It can only occur when an NPC attacks a player, and should now only occur if the NPC's level is 4 or more levels above your own. As a result it is irrelevant for end-game tanking in Mists of Pandaria. It can, however, occur with non-level capped characters against much higher level NPCs. The chance for a crushing blow to be received is a base 15%, even if the NPC is more than 4 levels higher than the player. There may be some buffs (such as The defense debuff in the WotLK Thorim encounter) that may still impact this mechanic.

This tutorial has focused solely on incoming attacks from bosses, but the attack table system is also used for our offensive attacks. We'll discuss that in the next section.

The Attack Table for Player Offensive Abilities

In the previous section, we covered how the attack table works for NPCs attacking the player. Things get a little different in the reverse case, but the general idea is the same.

Let's consider an un-specced and naked level 90 paladin, auto-attacking a level 90 mob from the front. This is important because NPCs cannot parry or block attacks behind them, but they can still dodge them. The basic attack table for the mob looks something like:
Table 1

Code: Select all

00.01-03.00 = miss
03.01-06.00 = dodge
06.01-09.00 = parry
09.01-15.00 = glancing (75% damage), Roll on Table 2
15.01-20.00 = crit (200% damage), Roll on Table 2
20.01-100.0 = hit (no block), Roll on Table 2


Table 2

Code: Select all

00.01-03.00 = block (30% damage reduction)
03.01-100.0 = hit (damage based on Table 1 result)


When Blizzard moved to the 2-roll system, it appears that NPCs block ability moved to 2-roll as well.

Just like with hitting the player, for each level higher the mob gains 1.5% chance to be missed, and to dodge, parry and block the player's attack. They also become 1% harder to crit per level and gain 6% additional chance per level for player auto-attack hits to be glancing blows. Now, for our hypothetical level 90 paladin auto-attacking a level 93 raid boss the boss's attack table looks something like:
Table 1

Code: Select all

00.01-07.50 = miss (7.5% miss chance)
07.51-15.00 = dodge (7.5% dodge chance)
15.01-22.50 = parry (7.5% parry chance)
22.51-46.50 = glancing (75% damage), Roll on Table 2
46.51-48.50 = crit (200% damage), Roll on Table 2
48.51-100.0 = hit, Roll on Table 2


Table 2

Code: Select all

00.01-07.50 = block (30% damage reduction)
07.71-100.0 = hit (damage based on Table 1 result)


In order to reduce the miss, dodge and parry chances, players must use the combat ratings of hit and expertise. This can come from racials, gear, consumables and buffs/debuffs. Hit reduces the miss chance of attacks against the mob. There are 3 types in the game: melee, ranged and spell. Currently, melee hit = ranged hit. Spell hit is still obtained from the hit combat rating, but is really only relevant to spell-casting classes and has a different rating->percentage formula for reasons we will discuss shortly. At 7.5% hit, our paladin will have eliminated their chances to miss the level 93 mob.

Expertise, on the other hand deals with dodge and parry. A major change in 5.0.4 was that now expertise first reduces the dodge chance until it is 0, then reduces the parry chance of the enemy until it is 0. After that, it has no effect. Currently there is no combat rating that a player can get that reduces the enemy's chance to block. Therefore block, like glancing blows, is just something you have to deal with. At 7.5% expertise (below that some dodge would remain), our paladin has pushed the level 93's dodge chance to zero:
Table 1

Code: Select all

----------- = miss (0.0% miss chance)
----------- = dodge (0.0% dodge chance)
00.01-07.50 = parry (7.5% parry chance)
07.51-31.50 = glancing, Roll on Table 2
31.51-33.50 = crit, Roll on Table 2
33.51-100.0 = hit, Roll on Table 2


Table 2

Code: Select all

00.01-07.50 = block (30% damage reduction)
07.51-100.0 = hit (damage based on Table 1 result)


Any additional expertise would reduce the parry chance, until at a total of 15% expertise our paladin has eliminated both dodge and parry from the level 93's attack table:

Code: Select all

----------- = miss (0.0% miss chance)
----------- = dodge (0.0% dodge chance)
----------- = parry (0.0% parry chance)
00.01-24.00 = glancing (75% damage), Roll on Table 2 
24.01-26.00 = crit, Roll on Table 2
26.01-100.0 = hit, Roll on Table 2


Table 2

Code: Select all

00.01-07.50 = block (30% damage reduction)
07.51-100.0 = hit (damage based on Table 1 result)


This is why most DPS only wish to get 7.5% expertise and hit: at that point, they've knocked dodge and miss of the boss's attack table and since they are usually situated behind a mob, they don't need to worry about parry (most ranged attacks similarly don't have to worry about it, regardless of position). Meanwhile For protection paladins, we need to ensure we hit in order to generate Holy Power, our main resource for active mitigation (since mana is largely a non-issue)... but that's another topic all together.

Earlier I mentioned spell hit having a different conversion rate than melee and ranged hit. This is because spells cannot be parried, blocked or dodged. This leaves hit, miss and crit on the table and to compensate for the radical difference in combat ratings, the hit chance required for spells is increased. The idea being that all things equal, a caster would need to allocate as many points toward hit as a melee DPS would need to allocate to hit + expertise. For most casters this rating comes from hit rating but it can also be from hit and expertise, which is the case for Paladins.

So for a spell being cast by our naked unspecced level 90 human paladin against a 93 level mob, we have an attack table like the following (most spells use a 2-roll system. Some only roll for crit chance.)
Table 1

Code: Select all

00.01-22.50 = miss (22.5% miss chance)
----------- = dodge (no dodge chance)
----------- = parry (no parry chance)
----------- = glancing (no glancing blows for spells)
22.51-100.0 = hit (no block), Roll on Table 2


Table 2

Code: Select all

00.01-00.38 = crit (3.38%-3% crit for level difference)
00.38-100.0 = hit (no block)


So if we gear that paladin up to have 7.5% hit and 15% expertise (=22.5% spell hit), we knock spell miss off the spell attack table as well as eliminating miss/dodge/parry from the normal melee table:
Table 1

Code: Select all

----------- = miss (0.0% miss chance)
----------- = dodge (no dodge chance)
----------- = parry (no parry chance)
----------- = glancing (no glancing blows for spells)
00.01-100.0 = hit (no block), Roll on Table 2


Table 2

Code: Select all

00.01-00.38 = crit (3.38%-3% crit for level difference)
00.38-100.0 = hit (no block)


So which attacks are 'normal melee' attacks and which ones are spells and what does that even mean? We'll get into that in the next section!

Offensive Abilities: Attack Table Types

Offensive abilities have different miss rates depending on the type of ability and target level. A boss is always considered level player+3, or 93 for a lvl 90 player. Here are the melee and spell miss chances for different level targets (assuming max weapon skill):

Code: Select all

    Target        Miss %         Dodge%   Parry%    Block%     Glancing%
    Level     Melee     Spell
     90        3.0       9.0        3.0      3.0       3.0           6.0
     91        4.5      13.5        4.5      4.5       4.5          12.0
     92        6.0      18.0        6.0      6.0       6.0          18.0
     93        7.5      22.5        7.5      7.5       7.5          24.0


Here's a complete list of our abilities, sorted by the type of attack table and avoidance they're subject to on a lvl 93 raid boss[9]:

*=procs seals

Melee Abilities - 7.5% miss chance, 7.5% dodge, 7.5% parry (unless otherwise noted), crits for 2x damage

One-Roll:
Seal of Insight/Truth/Righteousness procs (automatically succeeds if the triggering attack connects, only rolls for crit)
Censure applications (applying new stacks of Censure, obviously can't crit)
Censure ticks (automatically succeeds, only rolls for crit)

Two-Roll:
Auto-attacks*
Hammer of the Righteous* (main target component, physical damage)
Crusader Strike*
Shield of the Righteous*

Ranged Abilities - 7.5% miss chance, cannot be dodged/parried, crit for 2x damage

Two-Roll:
Judgement*
Hammer of Wrath

Spell-like Abilities - 22.5% miss chance, cannot be dodged/parried, 2.0x damage on crits

Two-Roll:
Consecration (each target gets a separate roll per tick)
Exorcism
Holy Wrath
Avenger's Shield (each bounce is a separate roll)
Light's Hammer
Execution Sentence
Hammer of the Righteous nova effect (AOE component automatically connects if the triggering attack connects, only rolls for crit)
Weakened Blows (from Hammer of the Righteous, affects all targets if the triggering attack connects and cannot crit)
Holy Prism (cannot miss, only rolls for crit)

Taunts - Automatically connect (0% miss/dodge/parry), cannot crit
Hand of Reckoning


So what does this all mean?

There are a few points to take away from this:

• Attacks are not all treated the same by the game, and this impacts how important different combat ratings are for given abilities and even specs/classes.
• Just because an attack can be made from range and looks like a spell in-game, doesn't mean that it is resolved the same way as 'true' spells.
• Attacks vs. different level targets have different requirements to hit (i.e. a lvl 93 NPC is harder to hit than a lvl 90 NPC for a lvl 90 player)
• Attacks by NPCs against the player are resolved a bit differently than those of the player against the NPC. Also note that Player vs Player attacks are resolved differently than the NPC vs Player examples here (i.e., the attacking player doesn't have fixed combat ratings based on level like an NPC)
• NPCs can block now. No combat rating reduces the amount of their block, much like glancing blows.
• The block chance you see on your character sheet will be higher than your overall block chance (i.e. the one you'd see in logs) unless you have 0 avoidance (dodge+parry+miss) as a result of the 2-roll (and 3-roll) system.
• Outside of very peculiar circumstances, we have extremely predictable attack tables in PVE encounters. This makes both offensive and defensive theorycrafting possible.
• The discussion here ignored hit chances for auto-attacks made by dual-wielding specs. Their auto-attack and special attack hit ratings are a bit different than single-weapon situations.

Yeah, TL;DR... is there a short version?

If you didn't want to wade through all the above, here's the Cliff's Notes version:

Attacks get resolved at the server by performing a series of 'rolls'. These rolls are numbers taken from a uniform random number generator (from 0-100, inclusive), and use several decimal places (i.e. not integers) and so are not simply the same as a '/roll 100' (which uses integers 1-100 inclusive). The outcome of the attack is then determined by a series of tables: one set for a player attacking an NPC, and another if it is an NPC attacking the player.

The NPC vs. Player tables boil down to one of the following (numbers here are base numbers for a level 90 vs lvl 90 mob, mobs get 1.5% reduction to miss, dodge and parry per level above you):

2-Roll (normal melee table)
Table 1

Code: Select all

00.01-03.00 = miss
03.01-06.00 = dodge
06.01-09.00 = parry
09.01-12.00 = crit (200% damage), roll on Table 2
12.01-100.0 = hit, roll on Table 2


Table 2

Code: Select all

00.01-03.00 = block (30% damage reduction - ignore critical blocks, since only warriors can do that)
03.01-100.0 = hit (100% damage or 200% damage if rolled due to crit)


For some special melee abilities and spells the crit vs. normal hit is moved to a separate table, and then goes to the block/hit table (Table 2 above). This is called a '3-roll system'. Spells generally ignore block, as well.

If instead, we have a base lvl 90 player attacking a lvl 90 mob from the front, the tables look like the following
2-roll, normal melee
Table 1

Code: Select all

00.01-03.00 = miss
03.01-06.00 = dodge
06.01-09.00 = parry
09.01-15.00 = glancing (75% damage), Roll on Table 2
15.01-20.00 = crit (200% damage), Roll on Table 2
20.01-100.0 = hit (no block), Roll on Table 2


Table 2

Code: Select all

00.01-03.00 = block (30% damage reduction)
03.01-100.0 = hit (damage based on Table 1 result)


Just like with hitting the player, for each level higher the mob gains 1.5% chance to be missed, and to dodge, parry and block the player's attack. They also become 1% harder to crit per level. NPCs also have the ability to be struck by a glancing blows, which are sort of like blocks in that they are 'hits' but with a reduced damage done amount, but they are on Table 1.

2-roll system for spells (lvl 90 human paladin vs lvl 90 mob):
Table 1

Code: Select all

00.01-09.00 = miss (9.0% miss chance - goes to 22.5% for lvl 93 mobs)
----------- = dodge (no dodge chance)
----------- = parry (no parry chance)
----------- = glancing (no glancing blows for spells)
09.01-100.0 = hit (no block), Roll on Table 2


Table 2

Code: Select all

00.01-03.38 = crit (3.38% spell crit chance - goes to 0.38% for lvl 93 mobs)
03.39-100.0 = hit (no block)


There are a few points to take away from this:

• Attacks are not all treated the same by the game, and this impacts how important different combat ratings are for given abilities and even specs/classes.
• Just because an attack can be made from range and looks like a spell in-game, doesn't mean that it is resolved the same way as 'true' spells.
• Attacks vs. different level targets have different requirements to hit (i.e. a lvl 93 NPC is harder to hit than a lvl 90 NPC for a lvl 90 player)
• Attacks by NPCs against the player are resolved a bit differently than those of the player against the NPC. Also note that Player vs Player attacks are resolved differently than the NPC vs Player examples here (i.e., the attacking player doesn't have fixed combat ratings based on level like an NPC)
• NPCs can block, and use a 2-roll system like players. No combat rating reduces the amount of their block, much like glancing blows.
• The block chance you see on your character sheet will be higher than your overall block chance (i.e. the one you'd see in logs) unless you have 0 avoidance (dodge+parry+miss) as a result of the 2-roll (and 3-roll) system.
• Outside of very peculiar circumstances, we have extremely predictable attack tables in PVE encounters. This makes both offensive and defensive theorycrafting possible.
• The discussion here ignored hit chances for auto-attacks made by dual-wielding specs. Their auto-attack and special attack hit ratings are a bit different than single-weapon situations.

placeholder

placeholder

Some comments:

1) It is worth making a clear distinction between auto-attacks and special attacks. For your examples, I'm assuming you want to demonstrate auto-attacks, which use a two-roll system. For special attacks, a three-roll system is used.

2) In the two-roll (auto-attack) system, crit is on the first table, as are glancing blows (on offense, only relevant to auto-attacks). In other words, the generic two-roll combat table is:

Miss
Dodge
Parry
Crit
Glancing (only relevant for players auto-attacks against bosses)
Hit

followed by a 2nd roll (if hit) for

Block
Hit

3) In the three-roll system, crit is moved to a third table:

Miss
Dodge
Parry
Hit

Followed by

Crit
Hit

Followed by

Block
Hit

I'm not 100% sure what boss special attacks use in general. I'm assuming the 3-roll table, but I've seen no evidence/testing to indicate one way or the other. For practical purposes it's irrelevant thanks to crit-immunity talents until we get another Thorim-style boss that applies an increased critical chance debuff.

I'd planned on sticking to 2-roll (thanks for the clarification on crit) attacks, but it's probably worth putting the 3-table variant in (and I will for player offensive abilities like the section yours had in it).

Also is it still true that crits can't be blocked for 2-roll attacks? So it's only a normal hit that forces a block/hit roll?

updated the first post to include the 3-roll table and a bit of discussion on it.

I'll add in the offensive sections when I get some more time.

You can definitely get blocked criticals. Hit and Crit results on table #1 will still trigger a roll on table #2 for blocks. Similarly for the 3-roll system.

theckhd wrote:You can definitely get blocked criticals. Hit and Crit results on table #1 will still trigger a roll on table #2 for blocks. Similarly for the 3-roll system.

Thanks - it seemed a little odd that they weren't counted as hits, but part of my brain convinced me it made sense

I tried backing some things out of the matlabadin code - but may be easier to just ask: does expertise *actually* reduce dodge then parry, or are there caps in place? (i.e. does it only reduces dodge up to 7.5%)

Also - did the glancing chance change with the changes to miss, dodge and parry chances? The posts I've found mentioning it didn't really say much other than the same 24% value I've seen pop up in posts made over several years.

Yes, expertise applies to dodge first, and then any leftover gets applied to parry. If you're up against a 92, for example, the first 5% expertise eliminates dodges and the next 5% eliminates parries. If you had 15% expertise and attacked a level 94, the first 9% would eliminate dodges and the next 6% would apply to parry (leaving a net of 3% parry).

As far as I know the glancing chance hasn't changed. Should be easy to check though, as there's no mechanic to eliminate it. Find a log where you're always attacking from the front (Stone Guard, maybe?) and you should see about 24% glancing blows.

updated with new info on block. I was able to verify glancing blows as well.

Right, you can has stiky.

Worldie wrote:Right, you can has stiky.

Thanky much!