A newcommer to the series

[SacredCaedeus]

[INCyr]

[scy]

I posted this elsewhere so here's just a random copy/paste:

Notes:
Power = Power of Heal Spell
aTEC = Attacker TEC
Mastery Bonus = Bonus from the Spell Mastery (Healer, War Lore)
aTR, aWP = Attacker STR, [Attacker Weapon Power / 3]
dVIT, dAC = Defender VIT, [Defender AC / 3]
aAGI, aLUC = Attacker AGI, Attacker LUC
dAGI, dLUC = Defender AGI, Defender LUC
pAGI, pLUC = Highest Party AGI, Highest Party LUC
eAGI, eLUC = Highest Enemy AGI, Highest Enemy LUC
aAGI, aLUC = "Attacker" AGI, "Attacker" LUC
vAGI, vLUC = Average Enemy AGI, Average Enemy LUC.

___

Healing = [[Power * (160 + aTEC) / 160] * Mastery Bonus]

Attack = (aSTR + [aWP / 2]) * (20 + aSTR) * (20 + aWP)

Defense = (20 + dVIT) * (20 + dAC)

BaseDamage = [[Attack / Defense] * 1.7]

Accuracy = [(BaseAccuracy * (1 + aAGI/100 + aLUC/200)) / (1 + (dAGI/100) + (dLUC/200)] * [AccuracyModifiers], capped between 20% and 100%.

Party Rating = (pAGI + 10) * (pLUC + 10)
Enemy Rating = (eAGI + 10) * (eLUC + 10)
Ambush = 5 * [Party Rating / Enemy Rating]
Blindside = 7 * [Enemy Rating / Party Rating]

Escape Chance = [10 * (aAGI + aLUC + 5) / ([vAGI] + [vLUC] + 5)]

Edit: Not 100% certain that dAC is also a /3 stat for the calculations but, for the sake of the symmetry involved with the formulas, I believe it to work that way. This means that if dVIT = aSTR and dAC = aWP, the damage dealt is merely aSTR + [aWP/2]. Multipliers, if any applicable, are applied after the Attack calculation; this applies to Skill Multipliers and the Weapon Multiplier (i.e., Whips 10/10 is 111% Damage).

Both Attack and Defense use the values from stats and equipment, not the listed ATK/DEF stats. If you have 48 VIT and +107 DEF, your DEF stat will list those two combined so any number crunching with those values will ultimately be wrong. For spells, STR becomes TEC and aWP becomes aSP (Attacker Spell Power) though it lacks the /3 factor (i.e., a Weapon of 120 ATK and Spell of 120 Power are not the same; the Weapon uses [120/3] or 40 while the Spell uses the 120).

In addition, monsters all deal 80% of the calculated Damage.

Random Variation is 1-5 for BaseDamage < 5; otherwise, it is 10 * [BaseDamage / 100].

For AccuracyModifiers, this applies to all abilities that modify chance-to-hit or chance-to-be-hit with the exception of Head Bind (80%) and Blind (25%) which are used after the Accuracy is capped.

Likewise, Shiraha is only used if the Accuracy check is successful; for example, if the Accuracy of the attack is 50%, then Shiraha's 70% chance-to-be-hit is applied if and only if that 50% succeeds and the attack would land for damage. Otherwise, the attack misses without Shiraha ever triggering.

For Ambush/Blindside abilities, the percentage is merely additive/subtractive and works after the calculation. Ambush is capped at 25% and Blindside is capped at 1%.

Your Chance to Escape also has a +10% per Failed Escape; the ESC Up ability merely adds 30% to the total. Escaping is capped at 90%.

And, finally, anything in brackets [ ] means to round down the result. So, to use the Weapons as an example:

aWP = [wATK / 3]
wATK = 114, 115, 116, 117, 118, 119, 120

aWP = [114/3] = 38
aWp = [115/3] = 38
aWP = [116/3] = 38
aWP = [117/3] = 39
aWP = [118/3] = 39
aWP = [119/3] = 39
aWP = [120/3] = 40
And so on.

This means that it takes 3 ATK to make a difference to the ATK formula (assuming you're at a wATK that is divisible by 3). Beyond that, 38 and 39 are the same in the actual formula of [aWP / 2] since [38/2] and [39/2] are both 19. Or, in other words, it takes 6 ATK to make a difference there (again, assuming the wATK is initially divisible by 6).

Though, there is still the plain aWP factor so it isn't a huge issue but just something worth noting I guess. Only real useful for those instances of upgrades that are almost the exact same.

___

Skill percentages can be found in the Guide that someone else listed; there's quite a few mistakes (notably the entirety of the Alchemist section) but they shamelessly stole the rest of the numbers from elsewhere so I can't be too surprised at the lack of fixes.