I’m in a bit of a tough spot. I love the vibe Old School Renaissance and the fantastic content coming out of it… but it’s just so much easier getting people to play Fifth Edition. To find a compromise, I tried my hand at a light set of house rules to give 5e that old-school feel.
The results are here! And they’re kind of short, to be honest. Most versions I’ve seen are much, much more complicated. I’ve even seen one clocking in at nearly 200 pages, and at that point, you’re really better off just playing a different game, or even creating one from scratch.
But that’s the point of mine. My goal was to pick house rules that could fit onto a single page, give the best return on investment for creating an OSR feel, with the ability to run those cool OSR modules for players who want to use the latest edition of the game. A lot of thought went into so few words—and in case my thought process might help you tweak your own 5e game to suit your own tastes, I’ve written that, too.
Fair warning, this article is going to be far longer than the hack itself. I’m not going into detail to convince you to do things my way, but by explaining my thought process, I hope you find something useful to bring into your games.
And before we begin, I want to point you all to the official 5th Edition conversion guide. (PDF Warning) In particular, since I think monster stats are the hardest thing to convert on the fly, the quick conversion section towards the end is a godsend. Most everything else can be handled with on-the fly judgment calls, in my experience.
Tiers of Play
The most striking difference between old-school and modern D&D is simply the power level. It’s purely a stylistic choice, but (to be brief) the game shifted from focusing on adventure to focusing on heroics. Neither is better than the other, and it’s not like any edition of D&D has done just one or just the other. Still, to get that old-school feel, we’ve got to focus on the scrappier, lower power levels.
The most obvious way of containing the power level of modern D&D is simply to lower the level cap. The idea is nothing new: Epic6, a variant of Third Edition D&D, did the same thing but stopping regular level advancement at level 6.
Basic D&D notably broke up tiers of play into different volumes of the rulebooks: Basic D&D featured levels 1-3, and Expert D&D expanded the levels up to 8 or 14, depending on class. Eventually this was expanded multiple times in a later revision of Basic D&D, with the Companion, Master, and Immortal sets.
This breakdown is mirrored by the “tiers of play” in Fifth Edition. Characters between the levels of 1 and 4 are “Local Heroes,” characters between the levels of 5 and 10 are “Heroes of the Realm,” and so on.
For our purposes, those first two tiers are probably best for an old-school feel. So, limit the levels to 4 and 10… or, maybe not.
See, when player characters advance to a new tier of play, that often comes with some significant boons. At level 5, Wizards and Clerics get third level spells, Fighters get two attacks per round, Rogues get a bump to their sneak attack and Uncanny Dodge… etc.—and everyone gets an improvement to their proficiency bonus. At level 11, Wizards and Clerics get sixth level spells, fighters get yet another attack, and Rogues get Reliable Talent.
Basically, if you decide to limit your games to a certain tier of play, I highly recommend raising the level cap by just one level, to give most classes a very impressive capstone ability.
For my purposes, I chose a level cap of 5, partly because that makes running an open table easier—there just won’t be any severe problems with parties of mixed levels. A lot of OSR content is geared towards low-level characters, so that helps, too. Finally, it simplifies a bunch of other stuff, but I’ll get to that later.
Race and Class
This one’s short. For a traditional game, you can limit the races to humans, dwarves, elves, and halflings, and limit the classes to fighters, wizards, clerics, and rogues.
…or, as I decided, don’t. Maybe your vision for a campaign really does need to exclude certain races or classes, but I think it’s a lot easier and a lot more fun for the players to lean on the side of letting them roll what they want.
(That said, I’m toying with a West Marches-style game with a developing town, and might begin with a reduced list of options until the players forge relationships and earn a name for themselves, opening up the roster of people willing to join the party.)
…okay, so maybe class options aren’t quite so simple.
The short of it is that multiclassing will inherently raise the power level of the characters, if only because it purely adds options. If it were weaker than sticking with a class, players could always just not do it, after all.
But I ultimately decided not to forbid this technically optional rule that’s usually assumed to be allowed anyway. This goes back to the level cap: at 5, there’s just not that much room to break the game, and any multiclassing is going to come at the cost of those capstone abilities. I like that it becomes a much more difficult choice.
For those who want to set a higher level cap, I would consider disabling multiclassing. You can always compromise for specific character visions by at least having the characters quest for it.
As a side note, combining a relatively small number of levels with a robust support for multiclassing reminds me of The Goblin Laws of Gaming, one of my favorite OSR systems that definitely influenced this project.
Another option that’s technically optional, but so common it’s usually just assumed.
The game just runs a little more cleanly without them, and certain powerful feats may bump the power level up a little bit. Again, I decided not to bother with disabling this one, because a 5th level character gets just one, possibly less if they multiclass. I’m fine with that once-in-a-lifetime decision for whatever boon they want.
As a side note, this is also why I would lean towards allowing Variant Humans regardless. Even if you don’t enable feats by default, it makes that one feat particularly special.
Besides, you can always let players venture forth on a quest to earn those feats, especially after reaching the level cap. Here’s a reddit comment from Matt Colville on why he doesn’t use feats, but also how he awards mechanical benefits a a function of the story. If the low level cap feels restrictive, remember that it’s certainly not the end of the story.
On that note…
Without going on too much of a tangent, I love the experience-for-gold system, as it shifts the focus of the game away from combat and towards more diverse forms of problem-solving. Not all of the experience in Basic D&D came from treasure—a small amount did in fact come from killing monsters and enemies—but, frankly, I didn’t see any reason to futz with both.
There’s just a bit of fine-tuning to do with how much gold to award. For that, I’d rather have a good rule-of-thumb than anything too complicated.
For a reference point, the text of basic suggests it should take about three “adventures” (there defined rather strictly as a foray into and out of a dungeon) to level up. I’ll take that to mean about three sessions. A quick check of the 5e recommendations on awarding treasure, and we’re about good to go: it would take about three hordes of treasure for a party to advance, which should line up with three satisfying sessions of play. Awesome.
As for running modules, I just look ahead and see if the module works on the assumption that you need hundreds or thousands of gold to get to level 2, and then just use factors of ten to get close enough. For instance, Lamentations of the Flame Princess uses a silver standard, and it takes a couple thousand experience to get to level 2, so it’s actually fine as is. Official old-school D&D modules, or those for retroclones, work under an experience-for-gold system where you need thousands of XP to hit level 2, so I just divide all the reward values by 10. Skerples deserves a special shoutout for explicitly putting its assumed conversion rate (200 gold to level 2) in the introduction of Tomb of the Serpent Kings, and is at least the first time I’ve seen it done.
Naturally, this also works if you want a game where players find a lot more treasure overall—you might just multiply the 5e experience table by 10, and bump up stingier modules
Whatever you do, keep most of this on the back end. The players should just know that earning gold (or silver) will get them an equivalent amount of experience, and not worry about the rest.
Rest and Healing
Healing was incredibly slow in B/X D&D, happening at a rate of 1-3 HP per day of rest. I don’t feel obligated to be completely accurate here, but I wanted something quite a bit slower than the modern system where a good night’s sleep heals just about any common injury. I want something that makes adventuring dangerous, places importance or resource management, and makes downtime important.
Luckily, the Dungeon Master’s Guide has a perfect solution: make short rests take 8 hours, and make long rests take a week. Easy. Now the players can’t count on easy recuperation, have to manage their hit dice and prepared spells much more carefully, and have a week to spend in town between outings.
I declined to use the “slow natural healing” rule also in the DMG, since it didn’t seem to add much to the “gritty realism” rules mentioned above that make rests just take longer. I did decide to add Healer’s Kit dependency, just to add another resource they need to keep in check.
To make death even scarier, I like a variant rule where failed death saving throws don’t just go away when a downed player is saved. (I first heard of this variant via WebDM.) Near-death experiences stop being something you just spring back from as if they never happened. I don’t want them to be permanent, though, so I leaned on the generous side by having them go away when hit dice are spent. I don’t want to doom the characters, just make them want to go to bed, and consider going home.
Speaking of death…
I’m not opposed to it as a general concept, I just think death should be a bit more challenging. It makes all the other risks more meaningful if, unlike in modern D&D past a certain level, you can just decide not to be dead.
So, as a general rule, even in my more heroic games of fifth edition, I would seriously consider making any resurrection spell off-limits for the players. It’s not that I want to make death permanent, I just want to make it harder to bring someone back. It should take a challenging quest, or a ritual that demands painful sacrifice, or a bargain with a feindish creature, or a pact with some other power, or… you get the idea.
But it doesn’t come up in my page of houserules. The reason goes back mostly to the low level cap, and a bit to the long rests. The only spell in the scope of those rules would be Revivify, which (while handy at staving off death) has the most limitations, can only be cast by max-level clerics (who did not multiclass!), and only twice at most between week-long rests back in safety. Those are enough limitations and stipulations that I’m comfortable leaving it in.
This one takes up nearly half of the text on my page of rules, but the basics can be summed up in the three bold sentences: “Items take up inventory slots; you have a number of slots based on your strength or constitution; and carrying more items than that makes you encumbered.”
I am distinctly not looking for accuracy in this one. I want gear to matter, but if we start worrying about individual item weights, then let’s face it, most tables will just give that up entirely in short order. It’s just more frustrating than fun. By making inventory slot-based, then it’s a more fun kind of frustration, planning what to bring on expeditions, how to get treasure back and what to leave behind, and—importantly—whether to use beasts of burden or hire retainers. The latter has a much more classic feel to it, and restrictive inventory incentivizes it.
My system is nothing new, bearing the closest similarity to The Goblin Laws of Gaming, though similar systems exist in Lamentations of the Flame Princess, Knave, The Black Hack, etc. I also combined Fifth Edition’s two levels of over-encumbrance into one for simplicity’s sake.
Another change I ultimately elected not to use, but I think is worth considering. Character skills are often used in place of player skills, and in keeping with the principles of the OSR, I want to reward player skill.
My solution is just to let players bypass skill checks entirely if they narrate appropriately. If they check the bottom of the drawers in the cabinet specifically to find a false bottom, then they will, no roll needed. If they just roll Investigation, then they’re going to have to roll very well.
So, I decided that handling this on the back end and setting player expectations worked better for me than to change the skill system itself.
However, the Dungeon Master’s Guide presents some great alternatives. Background and Personality Trait Proficiency are solid narrative options which other games have used to excellent effect (off the top of my head, wildcard skills in GURPS and backgrounds in 13th Age). For an OSR game, I might lean towards Ability Check Proficiency, which actually results in something quite close to (but much more elegant than) the SEIGE Engine from Castles & Crusades, a forerunner of the OSR.
…I really overthought this, and thoroughly enjoyed doing it.
The OSR isn’t some monolothic thing, where you have to meet certain requirements before the gatekeepers will let you pass. (Not that those gatekeepers don’t exist, they’re just more aptly called trolls. Just ignore them and they’ll slink back under the bridge.) It’s a style, and at the risk of making it sound like more than a bunch of people with a certain take on their own personal elf games, a movement. I went into detail about the decisions I made in the hope that you, the reader, might find it compelling, or helpful in making your game just a bit more fun.
So just be on the lookout for compelling ideas, and be bold in making them your own. If nothing else, that is definitely in the spirit of the OSR.