EDIT: Big changes for Rangers and Monks!
I compare the melee effectiveness of different classes in the spirit of Stalker0’s work. I look at all non-full spellcasting classes in melee combat at levels 3, 7, 11, and 15. I’m mainly interested in how these classes perform for non-optimizers, so I use vanilla builds (Human, default ability array, no feats). I tried to use weapons which synergize with the class features. I model the damage dealt and taken on a turn-by-turn basis across 7 encounters. I assume each encounter has 3 rounds for simplicity and to emphasize the value of front-loaded damage.
There are a few reasons you might be interested. First, it satisfies the basic curiosity on which classes deal the most/least damage in melee. Second, it helps inform GMs on adjusting encounters for your party composition. Lastly, for those who want to tinker with the system, this can be used to adjust classes for balance and to benchmark homebrewed classes and archetypes. For those who are interested, I present houserules which could improve balance between these classes. I’d love to hear any thoughts you have on them.
Three disclaimers. First, while I think it says a lot about class balance, this math analysis can’t fully capture the experience and effectiveness of characters during gameplay. There are lots of abilities which affect combat that I did not include in my model, and I didn’t quantify out-of-combat effectiveness. I list the abilities I didn’t model, and tried to roughly consider them in my commentary. This analysis also can’t capture how fun different classes are. Still, large differences in effectiveness can affect players’ enjoyment of the game.
Second, as with any model, the numbers are sensitive to the modelling assumptions. I call them out when I make them. The largest assumptions come from the effect of defensive abilities. In most cases, I model defensive abilities as reducing a percentage of damage based on a rough estimate of how likely I think the ability would be used and its numerical effectiveness. Different abilities will have different value depending on the types of enemies you face.
Third, the spreadsheet I made is complicated and it’s likely I made mistakes. I might also have made typos copying from the spreadsheet into the post or forgot to properly update something in the spreadsheet. Please let me know if you find any mistakes.
The characters:
Main Findings:
Metrics:
General assumptions:
[sblock]
Level 3 modeling assumptions:
[sblock]Barbarian
Paladin
Ranger
Monk
Rogue
Level 3 unmodeled abilities:
[sblock]
Level 3 commentary:
[sblock]I think things are fine at level 3 for the most part. Some things I noticed were:
Level 7 modeling assumptions:
[sblock]Barbarian
Paladin
Ranger
Monk
Rogue
Level 7 unmodeled abilities:
[sblock]
Level 7 commentary:
[sblock]Monks get a lot better thanks to Stunning Strike. The TWF Ranger drops significantly in effectiveness. It barely does more damage than the S&B Ranger! The Battlemaster gets closer to the benchmark Champion.[/sblock]
Level 11 modeling assumptions:
[sblock]Barbarian
Paladin
Ranger
Monk
Rogue
Level 11 unmodeled abilities:
[sblock]
Level 11 commentary:
[sblock]Level 11 is a power spike level for many classes. Fighters get a third attack and Paladins get their Improved Divine Smite. Rogues get Reliable Talent, a significant increase of their out-of-combat utility. Barbarians, Rangers, and Monks who don’t get a powerful level 11 ability fall behind.
Level 15 modeling assumptions:
[sblock]Barbarian
Paladin
Ranger
Monk
Rogue
Level 15 unmodeled abilities:
[sblock]
Level 15 commentary:
[sblock]The largest changes at level 15 are most of the poorer-performing classes recovering somewhat, with the Paladin and Monk dropping a bit. While the Paladin is now significantly less effective than the Champion, the Paladin has some powerful auras which are not modeled.
By the way, the Monk’s level 15 is probably the worst dead level in the game.[/sblock]
I hope you find this useful. Let me know what you think, especially if you think I’m undervaluing a class feature. For those who are willing to tinker with the system and are concerned with class balance, I’ll post some houserules which bring the melee classes closer in line with one another in the next post.
I compare the melee effectiveness of different classes in the spirit of Stalker0’s work. I look at all non-full spellcasting classes in melee combat at levels 3, 7, 11, and 15. I’m mainly interested in how these classes perform for non-optimizers, so I use vanilla builds (Human, default ability array, no feats). I tried to use weapons which synergize with the class features. I model the damage dealt and taken on a turn-by-turn basis across 7 encounters. I assume each encounter has 3 rounds for simplicity and to emphasize the value of front-loaded damage.
There are a few reasons you might be interested. First, it satisfies the basic curiosity on which classes deal the most/least damage in melee. Second, it helps inform GMs on adjusting encounters for your party composition. Lastly, for those who want to tinker with the system, this can be used to adjust classes for balance and to benchmark homebrewed classes and archetypes. For those who are interested, I present houserules which could improve balance between these classes. I’d love to hear any thoughts you have on them.
Three disclaimers. First, while I think it says a lot about class balance, this math analysis can’t fully capture the experience and effectiveness of characters during gameplay. There are lots of abilities which affect combat that I did not include in my model, and I didn’t quantify out-of-combat effectiveness. I list the abilities I didn’t model, and tried to roughly consider them in my commentary. This analysis also can’t capture how fun different classes are. Still, large differences in effectiveness can affect players’ enjoyment of the game.
Second, as with any model, the numbers are sensitive to the modelling assumptions. I call them out when I make them. The largest assumptions come from the effect of defensive abilities. In most cases, I model defensive abilities as reducing a percentage of damage based on a rough estimate of how likely I think the ability would be used and its numerical effectiveness. Different abilities will have different value depending on the types of enemies you face.
Third, the spreadsheet I made is complicated and it’s likely I made mistakes. I might also have made typos copying from the spreadsheet into the post or forgot to properly update something in the spreadsheet. Please let me know if you find any mistakes.
The characters:
- Fighter (Champion): Greatsword; Great Weapon Fighting and Defense
- Fighter (Battlemaster): Longsword and Shield; Dueling
- Barbarian (Berserker): Greataxe
- Paladin (Oath of Devotion): Longsword and Shield; Dueling
- TWF Ranger (Hunter): 2x Short Sword; Two-Weapon Fighting, Colossus Slayer, Multiattack Defense, Uncanny Dodge
- S&B Ranger (Hunter): Rapier and Shield; Two-Weapon Fighting, Colossus Slayer, Multiattack Defense, Uncanny Dodge
- Monk (Open Hand): Quarterstaff; Ki spent on Stunning Strike then Flurry of Blows
- Rogue (Assassin): 2x Short Sword
Main Findings:
- Before level 11, the Battlemaster outstrips the Champion in both raw numbers and utility. Their effectiveness is roughly equal after level 11, but the Battlemaster still brings more in-combat utility. This makes the Battlemaster the most effective character in melee combat.
- The Barbarian is roughly equal with the Champion before level 11, then falls behind.
- The Paladin is mostly in line with the Champion and is in a good spot.
- The Ranger is maybe a bit too effective given their utility at level 3 and level 7. They fall off to match the Rogue after level 11, which is probably OK given their level of utility.
- The Monk is weak until level 5, when he gets Stunning Strike. Stunning Strike seems pretty awesome. I modeled it as a 25% damage reduction (accounting for the save and single-target), but it’s probably better because it grants advantage to attack the poor target. I think the Monk is in good shape after level 5.
- Rogues seem to be in a good spot. They’ve got 70%-85% of a Fighter’s in-combat effectiveness and have great mobility and very effective boosts to their skill checks. I have one small concern: their contribution in combat is limited on turns when they can’t get Sneak Attack, eventually doing less than a cantrip.
- More generally, Ability Score Increases make a big difference. This is true even for the 3rd+ ASI boosting Constitution.
- Two-Weapon Fighting seems a bit weak. It’s nice for Rogues but doesn’t even add damage for the Ranger.
Metrics:
- DPR: average damage per round.
- DTPR: average damage taken per round.
- RIS: rounds I survive (HP/DTPR)
- EFF: average damage dealt before dying (DPR*RIS)
- %EFF: fraction of Champion’s effectiveness (EFF/Champion EFF)
General assumptions:
[sblock]
- 7 encounters per day
- 2 short rests per day
- 3 turns per encounter
- Fighting against one CR=Level enemy with stats from table on DMG p.274
- Human, default ability array (15/14/13/12/10/8)
- No feats
- Ability Score Improvements go to primary statistic, then to Constitution
- 100% of healing per short rest and 33% of healing per long rest added to HP.
- Studded Leather, Scale Mail, and Chain Mail at level 3
- Half Plate and Splint Armor at level 7
- Plate Mail and +1 magic weapons/armor at level 11 (including Bracers of Defense +1 for the Monk)
| Level 3 | Champion | Battlemaster | Barbarian | Paladin | TWF Ranger | S&B Ranger | Monk | Rogue |
| DPR | 9.4 | 9.0 | 10.0 | 7.9 | 12.0 | 9.2 | 10.2 | 13.6 |
| DTPR | 10.6 | 8.2 | 9.1 | 8.2 | 10.6 | 8.2 | 11.2 | 11.8 |
| RIS | 3.4 | 4.3 | 3.4 | 3.9 | 2.6 | 3.3 | 2.1 | 2.0 |
| EFF | 31.5 | 38.8 | 34.1 | 30.7 | 30.6 | 30.3 | 21.0 | 26.6 |
| %EFF | --- | 123% | 109% | 97% | 97% | 96% | 67% | 85% |
[sblock]Barbarian
- Resistance to bludgeoning, piercing, and slashing damage is a 33% reduction in damage.
- Uses Frenzy twice during the 7 encounters. This results in 2 levels of exhaustion, with disadvantage on ability and skill checks and halved speed. The third level of exhaustion imposes disadvantage on all attacks.
- Uses Reckless Attack on turn 3 of every encounter “for free”. I assume the Barbarian takes no additional retaliation for doing so, and using it on other turns results in enough additional damage taken to not be worth it.
Paladin
- All spells are used for Divine Smite
- For spells of the highest spell level, Divine Smite damage is multiplied by 1.25 to account for critical hits
Ranger
- Colossus Slayer is applied 75% of the time.
- Colossus Slayer cannot crit (tiny effect which makes math much easier).
- All level 1 and 2 spells used for Hunter’s Mark.
- Hunter’s Mark lasts 2 rounds.
Monk
- All Ki is used for Flurry of Blows.
- Deflect Missiles is a 5% reduction in damage.
Rogue
- Sneak Attack is applied 67% of the time
- The advantage bonus of Assassinate always works on round 1 and raises Sneak Attack chance to 100%.
- I ignore the auto-crit bonus of Assassinate for surprise.
Level 3 unmodeled abilities:
[sblock]
- Champion: --
- Battlemaster: 3x Maneuvers, Student of War
- Barbarian: Danger Sense
- Paladin: Spells 3*/0/0/0/0, Divine Sense, Divine Health
- Ranger: Spells 3*/0/0/0/0, Extra Skill, Favored Enemy x1, Natural Explorer x1
- Monk: Unarmored Movement +10 ft, Open Hand Technique
- Rogue: Extra Skill x2, Expertise x2, Thieves’ Cant
Level 3 commentary:
[sblock]I think things are fine at level 3 for the most part. Some things I noticed were:
- The Battlemaster is way more effective than the Champion (the Champion doesn’t gain much changing to Longsword + Shield), and has more utility to boot.
- The Monk is really far behind on raw combat effectiveness, and have roughly the same utility as the Battlemaster.
- This is more of a minor point, but the Ranger and Paladin (to a lesser extent) are about as effective as the Champion and have more utility.
| Level 7 | Champion | Battlemaster | Barbarian | Paladin | TWF Ranger | S&B Ranger | Monk | Rogue |
| DPR | 23.2 | 21.2 | 20.2 | 19.1 | 20.4 | 20.1 | 17.0 | 20.0 |
| DTPR | 23.8 | 19.0 | 19.2 | 19.0 | 21.4 | 17.1 | 15.0 | 17.0 |
| RIS | 2.9 | 3.6 | 3.4 | 3.6 | 2.7 | 3.3 | 3.7 | 2.9 |
| EFF | 67.0 | 76.3 | 68.3 | 69.1 | 54.4 | 67.0 | 63.6 | 62.5 |
| %EFF | --- | 114% | 102% | 103% | 81% | 100% | 95% | 93% |
[sblock]Barbarian
- Resistance to bludgeoning, piercing, and slashing damage is a 33% reduction in damage.
- Uses Frenzy twice during the 7 encounters. This results in 2 levels of exhaustion, with disadvantage on ability and skill checks and halved speed. The third level of exhaustion imposes disadvantage on all attacks.
- Uses Reckless Attack on turn 3 of every encounter “for free”. I assume the Barbarian takes no additional retaliation for doing so, and using it on other turns results in enough additional damage taken to not be worth it.
Paladin
- All spells are used for Divine Smite
- For spells of the highest spell level, Divine Smite damage is multiplied by 1.25 to account for critical hits
Ranger
- Colossus Slayer is applied 75% of the time.
- Colossus Slayer cannot crit (tiny effect which makes math much easier).
- All level 1 and 2 spells used for Hunter’s Mark.
- Hunter’s Mark lasts 2 rounds.
- Multiattack Defense is a 10% reduction in damage.
Monk
- Ki is used for Stunning Strike first and leftovers are spent for Flurry of Blows
- Deflect Missiles is a 5% reduction in damage.
- Evasion is a 15% reduction in damage.
- Stunning Strike is a 25% reduction in damage.
Rogue
- Sneak Attack is applied 67% of the time
- The advantage bonus of Assassinate always works on round 1 and raises Sneak Attack chance to 100%.
- I ignore the auto-crit bonus of Assassinate for surprise.
- Uncanny Dodge is a 20% reduction in damage.
- Evasion is a 15% reduction in damage.
Level 7 unmodeled abilities:
[sblock]
- Champion: Remarkable Athlete
- Battlemaster: 5x Maneuvers, Student of War, Know Your Enemy
- Barbarian: Danger Sense, Fast Movement, Mindless Rage, Feral Instinct
- Paladin: Spells 4*/3*/0/0/0, Divine Sense, Divine Health, Aura of Protection, Aura of Devotion
- Ranger: Spells 4*/3*/0/0/0, Extra Skill, Favored Enemy x2, Natural Explorer x2, Primeval Awareness*
- Monk: Unarmored Movement +15 ft, Open Hand Technique, Slow Fall, Ki-Empowered Strike, Stillness of Mind
- Rogue: Extra Skill x2, Expertise x4, Thieves’ Cant, Cunning Action
Level 7 commentary:
[sblock]Monks get a lot better thanks to Stunning Strike. The TWF Ranger drops significantly in effectiveness. It barely does more damage than the S&B Ranger! The Battlemaster gets closer to the benchmark Champion.[/sblock]
| Level 11 | Champion | Battlemaster | Barbarian | Paladin | TWF Ranger | S&B Ranger | Monk | Rogue |
| DPR | 37.3 | 33.5 | 26.9 | 33.0 | 29.7 | 29.6 | 25.2 | 27.3 |
| DTPR | 32.2 | 28.6 | 29.3 | 28.6 | 35.4 | 29.0 | 21.6 | 25.6 |
| RIS | 3.6 | 4.0 | 3.4 | 3.7 | 2.5 | 3.0 | 4.0 | 3.4 |
| EFF | 132.6 | 134.2 | 90.7 | 121.6 | 73.1 | 88.9 | 100.2 | 91.8 |
| %EFF | --- | 101% | 68% | 92% | 55% | 67% | 76% | 69% |
[sblock]Barbarian
- Resistance to bludgeoning, piercing, and slashing damage is a 33% reduction in damage.
- Uses Frenzy twice during the 7 encounters. This results in 2 levels of exhaustion, with disadvantage on ability and skill checks and halved speed. The third level of exhaustion imposes disadvantage on all attacks.
- Uses Reckless Attack on turn 3 of every encounter “for free”. I assume the Barbarian takes no additional retaliation for doing so, and using it on other turns results in enough additional damage taken to not be worth it.
- Relentless Rage is 33% damage reduction for 1 round worth of attacks each encounter while Raging.
Paladin
- All spells are used for Divine Smite
- For spells of the highest spell level, Divine Smite damage is multiplied by 1.25 to account for critical hits
Ranger
- Colossus Slayer is applied 75% of the time.
- Colossus Slayer cannot crit (tiny effect which makes math much easier).
- All level 1 and 2 spells used for Hunter’s Mark.
- Hunter’s Mark lasts 2 rounds.
- Multiattack Defense is a 10% reduction in damage.
- Whirlwind attack is 50% chance of an additional attack.
Monk
- Ki is used for Stunning Strike first and leftovers are spent for Flurry of Blows
- Deflect Missiles is a 5% reduction in damage.
- Evasion is a 15% reduction in damage.
- Stunning Strike is a 25% reduction in damage.
Rogue
- Sneak Attack is applied 67% of the time
- The advantage bonus of Assassinate always works on round 1 and raises Sneak Attack chance to 100%.
- I ignore the auto-crit bonus of Assassinate for surprise.
- Uncanny Dodge is a 20% reduction in damage.
- Evasion is a 15% reduction in damage.
Level 11 unmodeled abilities:
[sblock]
- Champion: Remarkable Athlete, Indomitable x1
- Battlemaster: 5x Maneuvers, Student of War, Know Your Enemy, Indomitable x1
- Barbarian: Danger Sense, Fast Movement, Mindless Rage, Feral Instinct, Intimidating Presence
- Paladin: Spells 4*/3*/3*/0/0, Divine Sense, Divine Health, Aura of Protection, Aura of Devotion, Aura of Courage
- Ranger: Spells 4*/3*/3/0/0, Extra Skill, Favored Enemy x2, Natural Explorer x3, Primeval Awareness*, Land's Stride, Hide in Plain Sight
- Monk: Unarmored Movement +20 ft, Open Hand Technique, Slow Fall, Ki-Empowered Strike, Stillness of Mind, Purity of Body, Tranquility
- Rogue: Extra Skill x2, Expertise x4, Thieves' Cant, Cunning Action, Infiltration Expert, Reliable Talent
Level 11 commentary:
[sblock]Level 11 is a power spike level for many classes. Fighters get a third attack and Paladins get their Improved Divine Smite. Rogues get Reliable Talent, a significant increase of their out-of-combat utility. Barbarians, Rangers, and Monks who don’t get a powerful level 11 ability fall behind.
- The Champion finally evens up with the Battlemaster in raw numbers, but the Battlemaster still has lots of utility.
- The Barbarian falls behind. They simply can’t keep up with the damage output. The Barbarian takes less damage, but the Champion’s higher AC, higher Constitution, and Second Wind means the Barbarian falls greatly behind in HP.
- The Ranger falls in line with the Rogue. Given the Ranger’s level of utility, it’s probably OK.
- The Monk falls behind on damage dealt and taken, but gets 12 more uses of Stunning Strike across the day. Again, I think the model assumptions underweight Stunning Strike, so the Monk is probably OK. The Monk also gets a long list of “utility” features, but don’t be fooled -- most have very little effect in-game. For example, at level 11, the Monk gains Tranquility. This is worse than Magic Initiate (Sanctuary), which is way worse than an Ability Score Increase.
- The Rogue falls to their lowest combat effectiveness, but gains Reliable Talent. I think their utility makes up for their lower combat effectiveness in general. My concern is they still only deal 1d6+5 damage (average 8.5) with an action if they can’t sneak attack. It’s almost not worth it! Compare this with a 3d10 damage (average 16.5) Fire Bolt cantrip. A Rogue might lose if trapped in a featureless room against a Wizard who’s completely out of spells (assume the Wizard still has Mage Armor up).
| Level 15 | Champion | Battlemaster | Barbarian | Paladin | TWF Ranger | S&B Ranger | Monk | Rogue |
| DPR | 38.7 | 34.4 | 33.1 | 35.1 | 29.7 | 29.6 | 27.0 | 34.0 |
| DTPR | 43.0 | 38.2 | 35.9 | 38.2 | 36.8 | 30.1 | 26.3 | 34.1 |
| RIS | 4.2 | 4.8 | 4.1 | 4.1 | 3.6 | 4.4 | 4.4 | 3.8 |
| EFF | 164.3 | 164.5 | 136.6 | 144.1 | 106.7 | 129.8 | 119.4 | 130.3 |
| %EFF | --- | 100% | 83% | 88% | 65% | 79% | 73% | 79% |
[sblock]Barbarian
- Resistance to bludgeoning, piercing, and slashing damage is a 33% reduction in damage.
- Uses Frenzy twice during the 7 encounters. This results in 2 levels of exhaustion, with disadvantage on ability and skill checks and halved speed. The third level of exhaustion imposes disadvantage on all attacks.
- Uses Reckless Attack on turn 3 of every encounter “for free”. I assume the Barbarian takes no additional retaliation for doing so, and using it on other turns results in enough additional damage taken to not be worth it.
- Relentless Rage is 33% damage reduction for 1 round worth of attacks each encounter while Raging.
- Retaliation adds 1/2 extra attacks every round (does not benefit from Reckless Attack).
Paladin
- All spells are used for Divine Smite
- For spells of the highest spell level, Divine Smite damage is multiplied by 1.25 to account for critical hits
Ranger
- Colossus Slayer is applied 75% of the time.
- Colossus Slayer cannot crit (tiny effect which makes math much easier).
- All level 1 and 2 spells used for Hunter’s Mark.
- Hunter’s Mark lasts 2 rounds.
- Multiattack Defense is a 10% reduction in damage.
- Whirlwind attack is 50% chance of an additional attack.
- Uncanny Dodge is a 20% reduction in damage.
Monk
- Ki is used for Stunning Strike first and leftovers are spent for Flurry of Blows
- Deflect Missiles is a 5% reduction in damage.
- Evasion is a 15% reduction in damage.
- Stunning Strike is a 25% reduction in damage.
Rogue
- Sneak Attack is applied 67% of the time
- The advantage bonus of Assassinate always works on round 1 and raises Sneak Attack chance to 100%.
- I ignore the auto-crit bonus of Assassinate for surprise.
- Uncanny Dodge is a 20% reduction in damage.
- Evasion is a 15% reduction in damage.
Level 15 unmodeled abilities:
[sblock]
- Champion: Remarkable Athlete, Indomitable x2
- Battlemaster: 9x Maneuvers, Student of War, Know Your Enemy, Indomitable x2
- Barbarian: Danger Sense, Fast Movement, Mindless Rage, Feral Instinct, Intimidating Presence
- Paladin: Spells 4*/3*/3*/2*/0, Divine Sense, Divine Health, Aura of Protection, Aura of Devotion, Aura of Courage, Persistent Rage
- Ranger: Spells 4*/3*/3/2/0, Extra Skill, Favored Enemy x2, Natural Explorer x3, Primeval Awareness*, Land's Stride, Hide in Plain Sight
- Monk: Unarmored Movement +25 ft, Open Hand Technique, Slow Fall, Ki-Empowered Strike, Stillness of Mind, Purity of Body, Tranquility, Tongue of the Sun and Moon, Diamond Soul, Timeless Body
- Rogue: Extra Skill x2, Expertise x4, Thieves' Cant, Cunning Action, Infiltration Expert, Reliable Talent, Imposter, Blindsense, Slippery Mind
Level 15 commentary:
[sblock]The largest changes at level 15 are most of the poorer-performing classes recovering somewhat, with the Paladin and Monk dropping a bit. While the Paladin is now significantly less effective than the Champion, the Paladin has some powerful auras which are not modeled.
By the way, the Monk’s level 15 is probably the worst dead level in the game.[/sblock]
I hope you find this useful. Let me know what you think, especially if you think I’m undervaluing a class feature. For those who are willing to tinker with the system and are concerned with class balance, I’ll post some houserules which bring the melee classes closer in line with one another in the next post.
Last edited:
. I'd prefer Rangers as alternate Fighters, but I don't think they can get there with only a simple set of houserules.
