D&D 5E 2017 Class Rank Survey: results are in!

[snafuy]

2017 D&D 5E CLASS RANK SURVEY

There's been only a single vote in the past three days, so I'm calling it done. Thank you to the nearly 200 participants.

https://docs.google.com/spreadsheet...68OKeKxKyM3prgvHqx1k7acRdTKO0ZR6XXLOp/pubhtml

I hope data nerds appreciate the graphs, especially the Power vs Enjoy bubble chart.

A+: outstanding far beyond the rest, with over 75% S or A votes. They deserve nerfs.
Hexblade Warlock, Divination Wizard, Lore Bard, Moon Druid

A: still well ahead of the B classes, but only half as much as A+.
Totem Barbarian, Vengeance Paladin, Abjuration Wizard, Divine Soul Sorcerer (also UA Revised Ranger and Mystic)

B+: the line separating B & B+ is fuzzy.
Life Cleric, Ancients Paladin, Tempest Cleric, Swashbuckler Rogue, Arcana Cleric, Zealot Barbarian, Oathbreaker Paladin, Battlemaster Fighter, Shadow Sorcerer, Gloom Stalker Ranger, Forge Cleric, Ancestral Guardian Barbarian

B: most classes are in the middle, where they should be.
Bladesinger Wizard, Fiend Warlock, Shepherd Druid, Devotion Paladin, Open Hand Monk, Conquest Paladin, Swords Bard, Evocation Wizard, Illusion Wizard, Light Cleric, Cavalier Fighter, Death Cleric, Draconic Sorcerer, Celestial Warlock, Arcane Trickster Rogue, Grave Cleric, Horizon Walker Ranger, War Cleric, Glamour Bard, Knowledge Cleric, Scout Rogue, Conjuration Wizard, Dreams Druid, Necromancy Wizard, Eldritch Knight Fighter, Nature Cleric, Kensei Monk, Shadow Monk, Enchantment Wizard, Valor Bard, Monster Slayer Ranger, Land Druid, Samurai Fighter, Drunken Master Monk, Thief Rogue, Whispers Bard, Assassin Rogue, Hunter Ranger, Arch Fey Warlock, War Magic Wizard, Great Old One Warlock, Long Death Monk, Crown Paladin, Storm Sorcerer

B-: just barely escaped median C.
Transmutation Wizard, Inquisitive Rogue

C+: just barely fell short of B-.
Redemption Paladin, Sun Soul Monk, Champion Fighter, Mastermind Rogue, Storm Herald Barbarian, Arcane Archer Fighter

C: clearly mediocre at best.
Trickery Cleric, Battlerager Barbarian, Undying Warlock, Berserker Barbarian (also UA Artificer)

C-: over 75% C or D votes.
Wild Magic Sorcerer, Purple Dragon Knight Fighter

D: overwhelming agreement that Mark Singer, Aang, & Korra deserve so much better.
Beastmaster Ranger, Four Elements Monk


COMMENTARY

The 1st survey question offered five ranking options (plus blank for no opinion):

• S (+2): Superior, Overpowered, Deserves a Nerf
• A (+1): Above Average, Mighty
• B (+0): Balanced, Good, Just Right
• C (-1): So-So, Passable, Niche
• D (-2): Defective, Inferior, Deserves a Buff

I used slightly different wording than in the race survey. With the phrase "Deserves a Nerf" appended to the S description, its vote frequency fell dramatically. No published class reached even 30% S votes. I think a lot of people like "overpowered" things, but they don't like the implication that overpowered is bad. (Yes, it is a bad thing.)

Nevertheless, the final tally was balanced almost perfectly. Out of more than 10000 individual votes cast, the total deviated from zero by only a handful of points.

The voting was pretty consistent for most classes, even the Xanathar ones. Tempest Cleric and Fiend Warlock share the Best Consensus trophy; their votes were over 95% B & A, with no Ds.

At the other end of the spectrum, Artificer and Mystic from Unearthed Arcana were a mess. All of those archetypes (and only those) had statistical variance scores higher than 1. But the votes for UA Revised Ranger didn't suffer from this problem. Why the difference?

Compared to the 2015 survey, these ratings show a lot of downward movement. Most of the previous A ranks have been demoted: Land Druid, Valor Bard, lots of Cleric & Wizard options. Transmutation Wizard suffered the biggest drop, from solid A down to B-. Many Bs became Cs, and the bottom Cs became Ds.

Swimming against that tide, Totem Barbarian moved upward from A- to surpass everyone except the A+ group. Is this due to Elk & Tiger totems from Sword Coast, or a high reassessment of the PHB options?

IMO, every archetype from B- rank on down should be given buffs (or at least nice ribbons).


SUBJECTIVE RATINGS

https://docs.google.com/spreadsheet...Hqx1k7acRdTKO0ZR6XXLOp/pubhtml?gid=1854970113
and
https://docs.google.com/spreadsheet...Hqx1k7acRdTKO0ZR6XXLOp/pubhtml?gid=1830653425

The 2nd survey question offered four options (plus Not Interested / no opinion, equivalent to blank).

• Played & Liked (+1)
• Want to Play (+0.2, because wanting isn't the same as doing)
• Never / Hated (-0.5, since at least the desire & action match)
• Tried & Disliked (-1, actual experience is worth more)

I then compressed to a relative scale with the highest total set to 100%.

Battlemaster Fighter, Totem Barbarian, and Lore Bard were the clear favorites.

The Xanathar classes tend to have lower totals, since fewer people have played them yet, but not by as much as I expected.

Tempest Cleric wins the Miss Congeniality trophy, as the only class with no negative votes at all.

Looking only at official published classes, there's an unmistakably strong relationship between power and happy gameplay. Nearly all of the well-liked classes ranked B+ or higher, while Elements Monk, Beastmaster, and PDK ruled the bottom of both lists.

The UA Mystic classes were glaring exceptions, all with high power but low preference. On the other hand, the UA Revised Ranger is beloved, in 5th place overall. Why the difference?

I think the world would be a better place if more people preferred fairness, shunning brokenly overpowered choices just as much as brokenly weak ones. Instead we're about to see vast numbers of Hexladins, Hexbards, and other Hex-dip multis.

 

[snafuy]

Does the Revised Ranger overcompensate and make the class too good?

Is Beastmaster objectively mechanically bad, or is it just unfun?

Is Hexblade overpowered as a single-class option, or is it mainly a problem due to multiclassing?

 

[mellored]

Does the Revised Ranger overcompensate and make the class too good?

No.

Is Beastmaster objectively mechanically bad, or is it just unfun?

Both.
You get a squishy beast, and it can't attack without sacrificing one of your own.

Is Hexblade overpowered as a single-class option, or is it mainly a problem due to multiclassing?

Neither. It just opens up some options.

 

[Swarmkeeper]

2017 D&D 5E CLASS RANK SURVEY
C-: over 75% C or D votes.
Wild Magic Sorcerer, Purple Dragon Knight Fighter

Aw - as a DM, I love me some Wild Magic Sorcerer. Such wacky fun can ensue, especially if the player is willing to use Tides of Chaos often. For example, in a recent session, the sorcerer rolled a 2 on the Wild Magic table - yep, he then had to roll on the table EACH ROUND for a minute. He summoned a Unicorn first, then became frightened of said unicorn next round. Doing so caused him to run from the Unicorn, who was less interested in fighting the Abominable Yeti and more interested in getting the food in the sorcerer's pack. Running away put the sorcerer out of range of the Yeti's cold breath, which killed two other characters and knocked two more out. Wild Magic saved his life!

I get it, some classes (or class-race-background combos) are not optimized for "best" combat results. But it is those characters that often bring the most memorable moments to the table - which is really such an important part of why we play.

 

[ccs]

Looking only at official published classes, there's an unmistakably strong relationship between power and happy gameplay. Nearly all of the well-liked classes ranked B+ or higher, while Elements Monk, Beastmaster, and PDK ruled the bottom of both lists.

If power = happy gameplay (and thus presumably happy players/customers).... Then the answer is obvious. Make more classes comparable to those that ranked the highest.
The answer is NOT nerf what people like.

The UA Mystic classes were glaring exceptions, all with high power but low preference. On the other hand, the UA Revised Ranger is beloved, in 5th place overall. Why the difference?

Because the Mystic only has a relatively tiny fan base & so far hasn't really nailed it for them. Or inspired the rest of us.
The Ranger on the other hand.... Everyone likes Rangers. And (practically) no one liked seeing the Beastmaster hamstrung by leftover 4e crap concerning how they & their beast interact. So anything that fixed that interaction would rank high.

I think the world would be a better place if more people preferred fairness, shunning brokenly overpowered choices just as much as brokenly weak ones.

I hope you're always in the minority on this.

Instead we're about to see vast numbers of Hexladins, Hexbards, and other Hex-dip multis.

I don't think my tables in too much danger of this. But if that's what the players decide to play? (shrugs) Then that's what I'll write my adventures to adequately challenge. Same as if they all said "Let's play Rogues!" or "Let's head into the depths of the Great Swamp looking for the Bullywug King!" Ok.

As for what gets chosen in other peoples games & how often/why? Might be interesting to hear about, but in the end it'll have little (if any) bearing on the games I play in/run. So it boils down to "So what?".

 

[Harzel]

If power = happy gameplay (and thus presumably happy players/customers).... Then the answer is obvious. Make more classes comparable to those that ranked the highest.
The answer is NOT nerf what people like.

Um, no? The logical extension of this is to give all PCs infinite HP, infinite spell slots, etc. I could be wrong, but I don't think that would make most players happy. There has to be a line someplace. You appear to be asserting that there should be no such line.

 

[CTurbo]

Not surprised about the Tempest Cleric as I've found it to be not only extremely fun to play, but also extremely effective.

What am I missing with the Divination Wizard for it to be ranked so good? (excuse my ignorance)

I haven't played one yet, but the Undying Warlock does not seem to be so bad. If anything, the UA Undying Warlock was OP.

I think the Long Death Monk should maybe be a little higher probably B+

and I still think the Paladin class in general needs some kind of nerf

But mostly I don't see anything that I really disagree with too much.

 

[Ancalagon]

I think the chart that fascinates me the most is enjoyment vs power. Look at the eldrich knight vs hexblade. According to this survey, the hexblade is extremely powerful (is it really?) while the EK is "average". However, the EK is considered more fun to play... Old ones vs Hexblade is another interesting comparison - same fun level, wide gap in power. Scout and arcane trickster - same power level, wide disparity in enjoyment.

There is definitely a trend, but the edge cases are very interesting.

Finally, I think I have to take these results with a grain of salt, as the "final" version of many of these subclasses has just been released. Which version of the Arcane Archer is underpowered exactly?

 

[flametitan]

I think the chart that fascinates me the most is enjoyment vs power. Look at the eldrich knight vs hexblade. According to this survey, the hexblade is extremely powerful (is it really?) while the EK is "average". However, the EK is considered more fun to play... Old ones vs Hexblade is another interesting comparison - same fun level, wide gap in power. Scout and arcane trickster - same power level, wide disparity in enjoyment.

There is definitely a trend, but the edge cases are very interesting.

Finally, I think I have to take these results with a grain of salt, as the "final" version of many of these subclasses has just been released. Which version of the Arcane Archer is underpowered exactly?

Additionally, 200 participants seems like a pretty small sample size to conclude trends upon player preferences, I would think. Using the 5th edition subreddit as a reference point, there's almost 70K subscribers. The D&D Facebook page dwarfs this at 737K likes and 673K of them receiving regular updates.

 

[Parmandur]

Additionally, 200 participants seems like a pretty small sample size to conclude trends upon player preferences, I would think. Using the 5th edition subreddit as a reference point, there's almost 70K subscribers. The D&D Facebook page dwarfs this at 737K likes and 673K of them receiving regular updates.

Yeah, this is fairly meaningless junk data. No more statistically significant than any random forum poll.