Stats Explained: What STR, INT and Luck Actually Add to Your Damage

✍️ Alice・愛麗沙

⚠️ This page follows Taiwan’s current server data (Season 0), labelled throughout. The stat-to-substat conversion table is Nexon’s own official mapping, published only on the help site rather than anywhere in-client. The damage-formula bucket structure and every divisor below are Korean-community-maintained reverse engineering — Nexon has never published the formula itself — cross-checked against two independent measurement threads where noted.

Nexon published the mapping — just not anywhere you’d look

This table is genuinely official, but it lives only on the help-center guide page — nothing in the client shows it, which is exactly why Korean players have complained about it for a year and why this exact question has come up dozens of times on Taiwan’s own board. The rule: every point of a primary stat adds 0.25 to each of its three dedicated sub-stats (not 0.25 split three ways):

Mapping is official (site guide only — not visible in-game); the +0.25/point figure is cross-verified by two independent KR measurements.

Worked example: +100 Luck becomes +25 Critical Hit, +25 Additional Hit, +25 Vital Evasion — run the first two through their own divisors into final damage and you land around +0.36%. That’s what “indirect” means for a stat point, and it’s also why stacking primary stats never feels as good as stacking secondary stats directly. The +0.25-per-point figure is corroborated by two independent Korean measurement threads — one derives it arithmetically, the other backs it out from a final-damage table, and the two close on the same number, which is about as high-confidence as community data on this site gets. One trap worth flagging: an older “+100 stat = +X%” conversion table (Strength 100 = +0.5% Additional Damage Enhancement, roughly) still circulates and is a completely different, incompatible model — later re-measurements in the same thread side with the +0.25 model, and this article uses that model throughout. Don’t mix the two.

The table’s shape also tells you each stat’s personality: Strength and Skill both feed offense-leaning sub-stats, Intelligence spans both offense and speed, Luck is the only stat feeding two independent damage-multiplying buckets at once (Critical Hit and Additional Hit) — the mechanical reason DPS classes lean Luck on their upgrade cards — and Will feeds three purely defensive/ultimate-charge sub-stats, making it the healer/tank stat rather than a damage number.

One honest gap: Vital Evasion and damage-mitigation sub-stats have no reliable public conversion formula, so this page doesn’t invent numbers for them — the divisor table below only lists offense-side values with a published formula behind them.

21 classes’ primary/secondary stats — find your row first

Primary/secondary mapping compiled from the KR official guide and per-class data; coefficients measured by players on lv20 blacksmith weapons and grow with weapon level — roughly 3× by endgame.Stat efficiency

The rule that turns a stat point into Attack: elemental Attack = stat points × weapon panel coefficient × 0.2. The coefficient is printed on the weapon’s own tooltip and scales with weapon level — a measured level-20 blacksmith weapon runs single-stat classes at ×1.5 on their main stat, dual-stat classes at ×1.0 main / ×0.5 secondary, and late-game weapons land around +0.6 Attack per point of main stat, +0.4 for secondary — roughly a 3x amplification over the raw formula. Three practical takeaways:

• A secondary stat’s Attack efficiency is a fixed two-thirds of the main stat’s — not a decoration you can ignore. Dual-stat classes’ secondary-stat gear rolls are real value, not filler.
• Upgrade cards are balanced around this. A main-stat card gives N, a secondary-stat card gives 1.5N — the two come out equal in raw Attack. So the right way to pick a card isn’t Attack at all, it’s which sub-stats the +0.25 path feeds — DPS classes picking Luck cards in high-crit-value level brackets (feeding both Critical Hit and Additional Hit at once) is standard Korean-community practice for exactly this reason.
• Food buffs don’t feed weapon elemental Attack (measured) — cooking buffs inflate your panel numbers but shouldn’t be counted at face value for actual damage math.

One more thing worth saying once: armour material determines which primary stat you get — heavy armour feeds Strength, light armour feeds Skill, cloth feeds Intelligence. Wearing the wrong material means literally missing your own primary stat; the auto-equip button in the top-left of the gear screen hints at your class’s preferred material for exactly this reason.

The bucket rule: same bucket adds, different buckets multiply

The damage formula (Korean-community-maintained; never published by Nexon) is a chain of multiplying buckets: Attack × Attack% × Additional Damage × Enhancement bucket × Gem tags × Critical × Additional Hit × Defense mitigation.

Sources inside the same bucket add together; different buckets multiply against each other — this one rule matters more than any single number here, because it’s the entire logic behind “when a bucket gets too tall, switch buckets instead of stacking further.” Here’s the exchange rate each value uses to enter its bucket:

Divisors are KR community-model constants (identical across two independent implementations), not official; the per-+100 column is a marginal value from zero — returns diminish as a bucket fills.Stat efficiency

Sorted by “value per +100,” the same numbers as a chart make it obvious at a glance which value in a shared bucket is actually worth more:

Three mechanics worth memorising on their own:

1. Additional Hit Enhancement and Critical Strike Enhancement share the same pool. The cleanest Korean-community test on this: fix the total, swap the split between the two, and damage matches to the last digit. Mechanically only the sum matters; concentrating on one in practice is about triggering rune/emblem synergies tied to that specific stat, not because one side of the pool is inherently stronger.
2. Critical has a closed-form solution. The crit-chance curve is 0.5×CritStat/(CritStat+2000), asymptoting toward 50%; the multiplier side starts at 1.4x and adds +0.2x per thousand points — combine both and they simplify to expected damage = 1 + CritStat/10000, which this site cross-checked against a community-measured reference table (3000→1.3x, 6000→1.6x) with zero error. This is the single most rigorously verified line in the entire model.
3. Defense mitigation is a hyperbola: 1 ÷ (1 + Defense/14828) — there’s no subtraction term, so it never hits zero. That’s the actual mechanical reason “deal damage during break state” isn’t folklore — the formula routes break-state damage through an entirely separate multiplying bucket that defense’s hyperbola can’t touch.

Walk the “same bucket adds” rule through a worked, measured example: 1,000 Attack, 300% skill coefficient, +20% Attack, +10% Additional Damage — the math is 1,000 × 1.2 = 1,200, multiplied by the skill coefficient gives 3,600, multiplied by 1.1 gives 3,960. Add the 20% and 10% together into 30%, or multiply the two Attack percentages against each other, and you get the wrong number either way — trace most “how much is this rune actually worth” disputes on the board back far enough and they usually turn out to be exactly this addition-vs-multiplication confusion.

Three secondary stats also carry trigger conditions — the panel number isn’t unconditionally active. Critical Strike Enhancement only applies when a single hit exceeds 40% of your Attack; Combo Enhancement needs 6+ hits within 3 seconds; Wide-area Enhancement needs 2+ simultaneous targets — so the only two sub-stats that are “just always on” are Critical and Additional Hit, which is half the reason they rank near the top of the per-point sort table. Critical’s multiplier side has one more layer: base 1.4x, +0.2x per thousand Critical points, then Critical Damage% multiplies on top of that — stack Critical Damage lines and the marginal value of your Critical stat rises with it, so these aren’t independent numbers. One more frequent question worth closing out: skill level doesn’t feed combat power, but multiplies directly into damage. Skill enhancement adds +3% per level, plus an extra +2% each at levels 2, 10, 15, 20, and 30 — max level 30 exactly doubles the skill’s coefficient (verified against two different classes’ measured data with zero discrepancy). Deciding whether your next resource should go to skill enhancement or a secondary stat is exactly what /statcalc exists to answer in one comparison.

Is “stack crit to 3000” a hard gate? The honest version

It isn’t a mechanical gate — it’s an economic crossover point. The curve for mental math: 1000 Critical Hit = 16.7% chance, 2000 = 25%, 3000 = 30%, 4000 = 33.3%, 6000 = 37.5%, 8000 = 40% — smooth throughout, no single breakpoint anywhere, and the asymptote is 50%: stat points alone can never push you past half, the rest has to come from flat-% sources on runes and passives. “Diminishing returns feel worse the more you stack” isn’t a perception — it’s the literal shape of this curve.

What 3000 actually means: under a typical Season-1-equivalent loadout, past this point every 100 Critical points’ marginal value starts losing to Additional Hit or Critical Strike Enhancement. Korea’s own already-played-through order, offered as a reference for Taiwan: stack Critical to 3000–3500 and stop, then put everything into whichever of Additional Hit or Critical Strike Enhancement your class prefers, and only return to top off Additional Hit toward roughly 2600 once Combo/Critical Strike Enhancement’s combined total passes about 4800.

Korea’s Season 2 target values have since risen to roughly 8000 — that’s propped up by new season runes and systems Taiwan doesn’t have yet; copying it today just burns resources for nothing.

Two things worth being upfront about. First, there’s a Korean per-point priority table circulating (Critical 0.00756% > Additional Hit 0.00679% > Combo = Critical Strike Enhancement 0.00669%…) that gets treated as scripture — but that’s the marginal value at one specific loadout. Every bucket has diminishing returns, so your own order won’t match theirs exactly; run your own panel through /statcalc and let /dps tell you where the next point should go.

Second, monster critical resistance is real — it isn’t in the main formula chain and no reliable number exists for it (its existence is credible; the magnitude isn’t), but measured results show Critical’s marginal value collapsing against high-resistance content. “Prioritize crit” is a content-dependent recommendation, not a universal constant.

On Combat Power itself: it’s a display formula, not the damage formula, and the two measured reconstructions of it don’t even agree on the secondary-stat weighting (0.3 flat vs. a 5:1 ratio) — this site keeps both readings on record rather than averaging them into a false-precision number. Where your next Combat Power should come from — and what to spend it on first — is a separate question with its own full per-unit-cost breakdown at /en/guides/cp-priority/; this page only covers the underlying mechanics.

Two new Korea Season 2 variables: Magic Resistance and season-skill thresholds

Taiwan will eventually meet two systems that reorder all of the above. Magic Resistance: below a zone’s pressure threshold, every 1000 points of damage shortfall is halved (a 2000 shortfall cuts it to a quarter) — a hard wall no secondary stat rescues you from, and resistance outranks everything else the moment it applies; above the threshold it caps out near +40%. Season-skill stat gates: Korea’s Season 2 season skills require a specific stat to hit 5000 before their bonus effect unlocks at all — the only genuine mechanical gate in this entire model, as opposed to every other “thousands” figure above, which is an economic recommendation, not a switch. Taiwan doesn’t have either system yet — worth knowing ahead of time, not worth pre-stacking for today.

Our conclusion

Taiwan’s current general-purpose DPS order: wear the correct armour material so your main stat fills itself in naturally (it’s a byproduct of gear, not something to chase directly); push whatever Critical Hit you can freely choose to 3000–3500; put everything else into whichever of Additional Hit Enhancement or Critical Strike Enhancement your class prefers; only circle back to top off Additional Hit toward roughly 2600 once the Combo/Critical total clears about 4800.

What would overturn this: this order depends on Korean-measured constants (8500 / 13000 / the closed-form crit solution) and the assumption that “the current Taiwan boss roster doesn’t carry unusually high crit resistance.” The moment anyone measures a live Taiwan boss showing sub-3000 Critical’s marginal return already losing to Additional Hit/Critical Strike Enhancement (under roughly 1.18% per hundred points), step two should yield to step three — or if Taiwan ever publishes an official formula that overturns any divisor here, this entire page gets recalculated.