The Classes Aren’t Balanced
3.5 was notorious for its game balance issues. Wizards, clerics, and other spellcasters were the kings of town, and martial classes were the peasants beneath their feat. Fifth Edition has the same problem. Spellcasters are still by far the most powerful, and martial classes are still sad. The wizard and druid are locked in a battle for first place,* with the cleric sitting comfortably in third. In fact, because of some silliness in the way multiclassing works, it’s even possible for wizards to use magic in heavy armor and have access to the entire cleric spell list! That certainly sounds balanced.
Game Mechanics: Wizard Cleric Multiclass
Spellcasting levels from different classes are added together when determining spell slots per long rest (page 165), so a wizard who takes 1 level of cleric loses no spellcasting progress. The way cleric spells are worded, they can choose any spell from their list to use for the day. Their choice isn’t limited by what slots they can actually cast. A first level cleric can choose a ninth level spell if they like; they just won’t be able to cast it (page 58).
A wizard with one level of cleric gets the same ability and will gain new spell slots as they level up. So a level 17 wizard/1 cleric can cast ninth level spells from both classes. In heavy armor.
Edit: There’s been some confusion of how this works, so here it is with more detail.
Page 164 of the mutliclassing rules states: “You determine what spells you know and can prepare for each class individually, as if you were a single classed member of that class.”
It then gives an example of wizard/ranger, neither of which prepares spells like a cleric does.
Back on page 54 of the cleric class rules, after explaining how you choose a list of cleric spells to be able to cast, the book reads “the spells must be of a level for which you have spell slots.”
Note that it does not say “cleric spell slots.”
Then back on page 164: “You determine your available spell slots by adding together all your levels of bard, cleric, druid, sorcerer, and wizard…”
So when you’re preparing your spells as a first level cleric, you can pick spells of any level for which you have slots, which in this case will be levels 1-9 because when you combine your class levels together, those are the slots available to you for casting.
The paladin is also pretty high in the ranking but only because it can cast spells. Even the least powerful spellcasting classes, warlock and sorcerer, are far above everyone else. Within the martial classes, there’s a clear ranking as well. Fighter and barbarian are the best of the worst, and ranger is at the absolute bottom.
5E’s attempt to balance spellcaster and martial classes is its rest system. In brief, most spellcasters recharge their abilities after an eight-hour long rest, while martial types recharge after a one-hour short rest. This assumes you are playing D&D as an endurance contest, exhausting your party’s resources over a large number of low level encounters.
However, a lot of people don’t play that way. For many groups, it’s more fun to have a smaller number of more difficult encounters. Fights aren’t fun when the players know there’s no chance of losing. Even if you play the game exactly as intended, clever spellcasters will keep their best abilities in reserve and then completely overshadow their martial companions when it matters. The difference in rest lengths is also tricky to manage. If the party can afford to rest for an hour, they can usually rest for eight hours.
Fourth Edition tried to solve the problem by standardizing all class abilities so they could be more easily balanced. This created a new problem of every class feeling the same, but at least it was an effort. Fifth Edition seems to have given up entirely.
The Class Specializations Aren’t Balanced
Each class in 5E has two or three specializations to give more variety. Rogues can be thieves, assassins, or arcane tricksters; fighters can be champions, battlemasters, or eldritch knights. This should be a good thing, as it allows for a wider range of characters without waiting for new supplemental books.
Unfortunately, like the classes themselves, few of these specializations are in any way balanced. Nearly every class has one option that is clearly the best. If you play a sorcerer, you get to choose between dragon blood and wild magic as the source of your power. Both sound really cool, but the dragon blood is unquestionably superior. It gets a powerful breath weapon and a sweet armor buff. The wild mage gets a random effect that happens only when the GM thinks it should, and then only when a 1 is rolled. Most of the effects aren’t even beneficial.
The lowest of the low is the beast master ranger. This pitiable class/specialization combo gets a weak animal companion, which it must give up its action in order to use. In many cases, the beast master is better off attacking on their own, so their main class feature goes completely unused. This is a major flashback to 3.5, when the ranger got an animal companion in the same way druids did, but at one half the level progression.
The lack of balance here is especially baffling. With twelve classes in the 5E player’s handbook, balancing them against each other would have been a challenging task. But there are only two or three specializations for each class.* Making them comparable to each other shouldn’t have been that difficult. I don’t know which scenario is more troubling: that the beast master wasn’t playtested at all, or that it was and they said “yeah, this looks fine.”
Update: Some time after this review was published, Wizards of The Coast put out a new version of the Ranger to address concerns that the class was underpowered. The beast master in particular got a huge boost.
The Gear Grind Is Worse Than Ever
D&D is infamous for being a game as much about magical accessorizing as it is about epic adventure. Gear is a huge part of a character’s advancement, and we can all remember long hours spent pouring over 3.5 magic item lists, looking for something that would give us a bit of an edge. Not getting the right equipment could render a character useless, and the endless quest for better loot would overshadow the adventure itself.
Fifth Edition is just as bad, but it’s in denial. The Dungeon Master’s Guide (DMG) section on magic items gives no indication how much gear a character should have at each level. None. Magic items also have hugely variable costs. An Amulet of Health can cost anywhere from 501 to 5,000 gold pieces. Talk about market fluctuations!
These factors, taken together, mean that GMs have no idea how much gear they should be giving their PCs, and gear really matters. A well equipped martial class can actually hold their own with the spellcasters. But what is “well equipped,” exactly? At what point does it become “over equipped” or “under equipped?” Fifth Edition has no interest in telling you.
Correction: There is a guide to how much gear higher level PCs should start with, on page 38 of the DMG,* but it’s very general and doesn’t provide the hard numbers a GM would find most useful.
The DMG indicates that magic items should be rare and wondrous, not a simple commodity to be traded at market like common equipment. That’s nonsense. Martial classes need gear to hold their own. Without magical equipment, PCs have nothing to spend their gold on, and earning gold is the default motivation for adventuring.* Fifth Edition is as much a game about magical accessorizing as 3.5; it’s just not as honest.
Some magic items are blatantly overpowered as well. Daern’s Instant Fortress is essentially a supercharged, reusable fireball* that also summons a tower for you to hide in. It costs the same amount as a quiver of +2 arrows. Then there are the items like the Gauntlets of Ogre Power, which set one stat* to 19 rather than giving a bonus. They create a strange incentive to start with some stats super low, because a character with 8 strength will benefit far more from the Gauntlets than a character with 18 strength.
It’s Easy to Fail Character Creation
Creating a character in D&D has always been quite the ordeal. You need to pick the right race/class combination, the right feats, the right everything. Get too much wrong, and your character will be completely unplayable. Since many people get into roleplaying games through D&D, that would be a major priority for Fifth Edition, right? Nope!
As mentioned, picking the wrong class or specialization can ruin your character right from the start. I shudder whenever a new player tells me they want to play a beast master ranger, and that’s just the start of it. There are six saves, but only three of them matter. Anyone who invests in Strength, Intelligence, or Charisma saves will be sorely disappointed.
Many of the base stats themselves are now traps. For example, sorcerers have no reason to raise Intelligence. In 3.5, it at least gave you more skill points. Now, all Intelligence does is give you a small bonus on some skills. So if you want to play a smart sorcerer,* you’re shooting yourself in the foot.
By the same token, armor class (AC) is way more important than it used to be. Many spells and other effects target AC, so being easy to hit is a death sentence. If you decide to leave Dexterity at 10 because your class doesn’t traditionally need it, your character will soon be a pincushion. In 3.5, you could mitigate having a low AC, but not anymore.
Physics Issues Persist
3.5 D&D tried to simulate an entire world. There were rules for exactly how many feet a person could jump based on their height and rules for what happened when you put a one dimensional folding device inside another. This led to some… interesting results. Fourth Edition was much more abstract, with most of its rules only pertaining to the exchange of damage in combat. Fifth Edition tries to walk a middle ground, not having rules for every little thing but also being more than a white room in which fights take place.
Unfortunately, it does not always succeed. The rules don’t cover a number of situations that are very likely to come up. One is the Sleep spell. It will typically knock out a low-level target with a single casting. To balance this, the spell states that any damage immediately wakes the target up. But what about snapping a pair of manacles on them while they sleep? Does that count as an attack, and if so, do they wake up before or after the manacles are on? The game gives no indication. For that matter, what about lifting a really big rock over the target and dropping it on them?
Invisibility is another problem. An invisible character should be, by definition, invisible. But by a strict reading of the rules, they are only a bit harder to hit. Enemies can still walk up to them and attack, and it’s not clear how they’re located. For extra silliness, the penalty to attack an opponent you can’t see is the same as attacking while prone.
Then there are owls. Giant owls, specifically, brought forth by the Conjure Animal spell. The spell is powerful enough at face value, as the critters it summons are quite strong. But giant owls are the worst, thanks to the grapple rules. While grapple has been simplified, it still allows you to drag enemies across the map. Giant owls can drag enemies straight up and then drop them, doing a disproportionately high amount of damage in any battle with an open ceiling, to say nothing of those bottomless caverns GMs are so fond of.
Owls can lift quite a lot, it turns out. Up to 195lb, and any creature Huge sized or smaller.* Nothing in the rules makes it harder to move something straight up than to drag it across the ground. PCs can also ride the owls, which means Conjure Animal can bestow flight on a party of up to eight characters for an hour. If they don’t like an encounter, they can skip over it.
Building Encounters Is Really Hard
Dungeons and Dragons, no matter what the edition, is all about the encounter. Balancing difficulty so a fight is challenging but not impossible is a lot of work for the GM. In 3.5, the encounter builder was almost useless, because it was based on level only. A PC’s capability could vary wildly without ever leveling up.
Fifth Edition has the same problem. According to the book, an encounter worth 20,400 experience points (XP) should be “deadly” for a group of four level 13 PCs. That means death is likely for at least one character. In one game run by a friend of the blog, a group of four level 13 PCs defeated a 150,000XP encounter without a single fatality.*
How can that possibly work? It has to do with how optimized the characters are for their level, and these were some damn optimized characters. A more casual group might indeed have been threatened by a 20,400XP encounter. A group of beast master rangers would have been wiped out. Because character ability varies so wildly, following the encounter builder will get you nowhere.
The encounter builder also doesn’t compensate at all for magic items. The group in question had what I thought was a lot of powerful gear, but I have no idea how it would have rated in the designer’s eyes.
If the GM can’t effectively build encounters, they’ll have a hard time building an effective campaign, or even a single session. This problem isn’t insurmountable, but it is disappointing. That sums up Fifth Edition pretty well: disappointing. A few improvements have been made, but in general it has all the same problems 3.5 did. At least Fourth Edition was willing to try something new, for all its flaws. If you’re looking to play an improved version of 3.5 Dungeons and Dragons, stick with Pathfinder. It’s better supported, and you probably already have the books.