Baldur's Gate 3 weapon damage math

The recurring BG3 weapon questions, in one click each

The build-choice math players keep re-asking. Each card links to either the live comparison tool or the worked-out concept pillar.

• Greatsword (2d6+3) vs greataxe (1d12+3)?

  Half a point of mean separates them; greatsword's tighter variance wins at low HP, greataxe's fat tail wins above the mean. The crossover sits right at HP 10.

• Longbow (1d8+3) vs heavy crossbow (1d10+3)?

  Heavy crossbow wins on per-shot mean by 1 — but loses Extra Attack interaction. At level 5+ longbow doubles its DPR; heavy crossbow stays on one shot unless you have crossbow expert.

• Should I take Great Weapon Master?

  Closed-form rule: break-even AC ≈ attack-bonus + 8 for Greataxe-class. With +9 to-hit and a magic weapon, GWM is on against everything except plate-and-shield.

• Sharpshooter break-even AC?

  One AC point lower than GWM: attack-bonus + 7 for Longbow. And at long range, Sharpshooter is correct almost regardless of AC.

• Is the +1 weapon worth the slot?

  The damage shift is a flat +1 per swing (1d8+3 → 1d8+4, mean 7.5 → 8.5). The +1 to-hit is usually the bigger deal — it's a property of the attack roll, not the damage roll, and lives outside this engine's scope.

• GWF reroll — placebo or real?

  Real: ~+0.83 mean on Greataxe, ~+1.33 on Greatsword for the canonical reroll-on-1-or-2 (≈+11% on Greatsword DPR). Plus a quiet variance reduction that helps above the mean and hurts below.

• Paladin Divine Smite — when do I burn the slot?

  Smite-only-on-crit delivers exactly 1.875× more damage per slot spent than smite-first-hit, across every slot level. Champion 3 multiclass is the build that actually clears its slot bag.

• Champion Fighter (crit on 19+) — does it pay?

  Sometimes. Two builds with identical expected DPR can have very different per-target kill rates — the variance pillar applied to one weapon's crit profile.

• Elven Accuracy worth the +2 ASI?

  Yes if you're at advantage on more than ~30% of attacks. Mean of 3d20kh1 is exact: 1239/80 = 15.4875, and crit rate jumps to ~14% on advantage rounds.

Common martial weapons

All means below assume a +3 ability modifier (STR or DEX 16, the typical mid-game value before stat bumps). For a +4 or +5 modifier, add 1 or 2 to every mean. Click any expression to open its full distribution in the engine.

Ranged weapons

Build-choice deep-dives

Greatsword vs greataxe

Both two-handed, both standard 5e fare. Greatsword (2d6+3, mean 10, range 5–15) has slightly higher mean and substantially lower variance than greataxe (1d12+3, mean 9.5, range 4–15). Most build guides stop at "greatsword is +0.5 damage on average, therefore better." That's true if you're chip-damaging trash where any swing finishes them — and if you're swinging at a target whose HP you can't reach on a typical roll, the higher-variance greataxe lands more kills.

Worked out at the variance-pillar level: Why variance helps when your mean is below the threshold . Try it live: /vs/2d6+3/1d12+3 — drag the HP slider and watch the winner flip around HP 10.

Dual-wield (two attacks at 1d6+3) vs two-handed (one at 2d6+3)

Two attacks of a light weapon (shortsword, scimitar) versus one big swing. Once both attacks land, the cumulative damage of two 1d6+3 rolls (mean 13, variance ~5.83) outperforms 2d6+3 (mean 10, variance 5.83) on every threshold — the modifier counts twice.

The interesting question is when the second attack misses. Dual-wielding's bonus action attack is no-ability-mod by default (Two-Weapon Fighting style adds it back); without that style the comparison flips: 1d6+3 + 1d6 (mean 6.5+3.5=10, variance ~6.7) vs 2d6+3 (mean 10). Same mean, dual-wield slightly higher variance — and now the reliable-vs-nuke question answers it: dual-wielding (slightly higher variance) wins above the mean, two-handing wins below. The main difference is to-hit chance, which the math above doesn't capture but which usually dominates in practice.

Longbow vs heavy crossbow

Longbow (1d8+3, mean 7.5) vs heavy crossbow (1d10+3, mean 8.5). Heavy crossbow wins on mean by 1 per shot, but loses Extra Attack interaction — at level 5+ longbow gets two attacks per round (mean 15) vs heavy crossbow's loading property limiting it to one (mean 8.5). The damage math says heavy crossbow is dominated past level 5 unless you have crossbow expert. Try them side by side: /vs/1d8+3/1d10+3.

Should I take the +1 weapon?

A +1 weapon shifts every swing's damage by 1 (1d8+3 becomes 1d8+4, mean 7.5 → 8.5) and adds +1 to-hit. The damage shift directly changes which HP thresholds you can reach reliably: /vs/1d8+3/1d8+4. The +1 to-hit is usually the bigger deal but lives outside this engine's scope — it's a property of the attack roll, not the damage roll.

BG3-specific notes

• Magical weapons stack their bonus on damage and to-hit; the damage piece behaves like the "+1 weapon" section above with larger numbers.
• Some uniques add additional damage dice of a different type (e.g. a flaming sword: 1d8+3 + 1d4 fire). Resistance and immunity are computed per damage type; a fire-resistant target halves only the +1d4 portion, not the base 1d8+3.
• Crits in BG3 follow the 5e doubling rule: roll the dice twice, add modifiers once. So a crit on 1d8+3 becomes 2d8+3, not 2(d8+3). Try /strike/1d8+3 next to its crit form /strike/2d8+3.
• Champion fighter (level 3+) extends the crit range to 19–20. That's the main case where crit chance beats base damage — specifically against high-HP solo bosses where normal hits chip but rarely close out.

Run any expression

↦ Comparison tool — type your own weapons ↦ Concepts — counterintuitive damage math ↦ /strike/2d6+3 — greatsword's distribution ↦ /strike/1d12+3 — greataxe's distribution ↦ /vs/2d6+3/1d12+3 — the choice