In a recent blog post about the problem of roleplaying games becoming more like video games, James Maliszewski wrote:

Tabletop RPGs aren’t about reaching 10th level. They’re about entering and exploring an imaginary world through an equally imaginary character. What matters isn’t how many hit points your fighter has, but what you do with them. Success might mean founding a colony, retiring in disgrace, making a terrible bargain with an otherworldly power, or changing the course of an empire. These are the kinds of outcomes that emerge from choices, consequences, and collaboration with the referee and other players, not from ticking boxes on a character sheet. Advancement in a tabletop RPG is ultimately about meaning, not math.

He contrasted this sort of “mechanical advancement” with games that don’t use levels, such as…

…Call of Cthulhu, where the main arc of a character’s life isn’t defined by rising power but by gradual decline—into madness, death, or at best, retirement from delving into the Mythos. He might get better at Library Use or Spot Hidden, but he’ll never become an investigator resistant, never mind immune, to cosmic horror. That’s not the point of the game. Even RuneQuest, though it includes skill advancement through use, eschews levels entirely. A seasoned Gloranthan character is still vulnerable, still mortal.

And finally ends with:

These are the kinds of outcomes that emerge from choices, consequences, and collaboration with the referee and other players, not from ticking boxes on a character sheet. Advancement in a tabletop RPG is ultimately about meaning, not math.

It’s almost an Emily Litella moment. James’s post completely misses one of the biggest advantages of level-based systems: that they aren’t about “ticking boxes on a character sheet”. Most non-level based systems, especially his examples of Runequest and Call of Cthulhu¹, handle character advancement by literally ticking boxes.

James misses this, I think partly because he seems to use “mechanical advancement” as if its definition excludes Runequest-style advancement. This makes no sense to me (so I may be having an Emily Litella moment of my own right now). I think the traditional differentiation is between direct advancement (at the level of skills) and abstract advancement (such as levels), but both are mechanical. They both improve the character’s/player’s chances of succeeding according to the mechanics of the game.

And they are both something to strive for that is more game mechanics than game world.

Some games don’t have any mechanical advancement but they’re rare. The one that comes to mind for me is Everway, and even there the author later published a potential mechanical advancement method.

Most games mediate their mechanical advancement through adventuring rather than in-game training and study.² For example, adventuring in Dungeons & Dragons provides experience points. Investigating mysteries in Call of Cthulhu and engaging in various kinds of adventure-related activities in Runequest provide the opportunity to tick skills for advancement.

Another way of thinking about mediated advancement is that it is advancement by doing what the game was made for.

James praised Traveller as a game without advancement; it’s one of his favorite games. His praise came with a caveat, but that caveat is a very important one: Traveller does have mechanical advancement. What it doesn’t have is mediated advancement. Advancement isn’t achieved by doing what the game was made for, that is, adventuring among the stars. If you want mechanical advancement in Traveller your character has to divert their attention from what the game was designed for and dedicate it to training or study.³

A lot of attention. A Traveller character “devotes himself to a four-year program of self improvement, dedicating his endeavors in something like obsession” if they want a mechanical advancement. After their character’s four-year-program, the player must roll 8 or higher on 2d6, or the time spent was wasted.⁴ That’s a 58% chance of failure.

John Carter visibly goes up a level (or ticks off the last swordsmanship box in Runequest) in Edgar Rice Burroughs’s Warlords of Mars.

I think even the game’s author recognized that this was more than a bit disincentivizing for a role-playing game. In another short paragraph in Traveller Marc Miller suggested other methods for mechanical advancement that did not remove or partly remove the character from play for four+ game years on a low chance of success.

Such methods could include RNA intelligence or education implants, surgical alteration, military or mercenary training, and other systems. Alternatives to the above methods must be administered by the referee.

Notably lacking from this list is experience, that is, improving abilities by using them in adventure-related activities. Interestingly, each of Miller’s alternative methods could be handled in-adventure.⁵

From his complaints, however, it sounds like what James is really discussing is not levels or advancement, but the reduction of goals to what is stated in the rules or the adventure, reducing player choice to finding ways of ticking off the accomplishment of those goals. That’s what happens when a tabletop game comes to resemble a video game. And it has nothing to do with levels. It’s about the importation of video game methods to roleplaying games.

It’s similar to so-called cinematic roleplaying. The real problem isn’t that video games imported levels from RPGs and then RPGs imported levels from video games, but that RPGs (the exemplar being the modern forms of D&D) have imported the mindset of video games and cinematic gaming: reduce choices so that roleplaying becomes storytelling. In cinematic roleplaying, the goal is to discover the director’s story. In video game roleplaying, the goal is to discover the programmed cues or paths.

That is, in video game-style playing or cinematic-style playing, it is the job of the players to find the plot at the end of the script. It is not the job of the players to Kobayashi-Maru the adventure and make a completely new solution or generate their own narrative that may differ not just from that expected by the game master but even that envisioned by other players.

That’s what leads to players treating the game as “ticking off boxes” as a replacement for progress. They’re not just ticking off boxes on advancement, they’re ticking off boxes everywhere in the game. It’s how such games are meant to be played.

Even in its earliest published forms, Dungeons & Dragons was about players using the results of their adventures to make their character’s way in the world. That is, using the money, knowledge, and experience they received to build kingdoms, to inspire religious fervor, to interact with a world in such a way that the world changes. None of that was mechanical, which is why D&D needed a gamemaster.

James is discussing a symptom rather than the problem: that levels in modern D&D have become a system of ticking off accomplishments. The solution is not games that are often literally about ticking boxes on the character sheet. Runequest, Call of Cthulhu, even my own old-school favorites Flashing Blades and Daredevils⁶, literally have checkboxes that you check next to each skill and/or attribute in order to advance in them.

The player is, in other words, always looking to the character sheet to see how this particular situation will help them advance. Levels abstract that away—there’s nothing on an early D&D character sheet to tell you how to roleplay your way through the current situation to maximize character advancement.

Levels, in other words, are a great way to get away from ticking boxes on a character sheet. Done right, levels move away from mechanics and return to a game of meaning rather than math. The problem James described in his post is one that permeates the entirety of modern D&D, not just character advancement.

• The inability to use more than a specific number of special magic items unrelated to any physical limitation of the character? That’s from video games. It’s pure programming.
• Reliance on a grid for movement, reach, range, and special effects? Video games.
• That characters are almost literally indistinguishable except for superficial reskinning, so that warriors use weapons and magic-users use at-will power blasts to do the same thing? That’s from video games. Again, it’s the easiest way to program player characters. They all do the same thing and the programmer merely has to change the descriptor.
• Adventures designed with a specific path that the characters are expected to take, checking off progress along the way? Adventures that almost literally, if not literally, are interspersed with cut scenes to reward reaching those scripted milestones? This is, literally, video games.

The player always should be ahead of the game rules.

These are just off of the top of my head. None of it has to do with levels. All of it would remain video-game-like if mechanical advancement were handled at the skill and attribute level.

It’s hidden by his classification of some mechanical advancement forms as not mechanical, but what James seems really to be asking is not “are levels necessary” but “is any form of advancement necessary” and “should advancement be mediated through doing what the game was meant for”? I think, in the end, advancement—not necessarily defined as a raw power increase—is necessary. And any such advancement should be mediated through doing what the game was meant for. This is something I’ve thought about a lot.

Should a character with twenty adventures under their belt have a better chance of survival and more options for success, than a character with one adventure? I think they should. My usual example, paradoxically given my dislike of cinematic roleplaying, is that protagonists are more likely to die in the first few chapters or scenes than three-quarters of the way through a book or movie.

It does happen, of course. But when it does it’s news. And imagine if, say, Marion Crane in Psycho had died at 14% of the movie remaining instead of 57% remaining?⁷ This technique tends to be powerful partly because it is rare and partly because it’s an unnatural narrative. Examples range from just about any movie where the character has an incurable disease, to movies based on real life adventurers such as Disney’s Davy Crockett.

I don’t see this as contradicting my abhorrence of cinematic roleplaying. We expect late removals to be rarer in any narrative, not just in scripted storytelling forms such as books and movies. By making character removals rarer, what removals do occur gain more impact. This is exactly what we’d want for the death of an established player character.⁸

What levels make easier is defining advancement less as a means of gaining specific skills or abilities and more as revealing the character’s place in the world. One of the commenters called this a “pacing mechanism”, but I’d argue that it’s more than that. It’s a link between the mechanical interactions of the character and the social interactions of the players. One of modern D&D’s big problems in my view is that it flattens the “experience” aspect of level advancement—hit points, success rolls—in favor of a predefined progression of special abilities.

There will always be choices that maximize character advancement in any game that has character advancement. But with levels this doesn’t usually involve ticking boxes—or ticking anything—on the character sheet. Play is solely about interacting with the world and the advancement generated by that interaction is applied later.

My own view is not only that levels are the best way to avoid complicated advancement methods that require ticking boxes on a character sheet, but that a simple acquisition of experience points is the best way to measure advancement. Especially when playing with people who aren’t necessarily immersed in the source material, character advancement rules provide a simple guide to what players ought to be focusing on. Is it about adventuring in dungeons? Solving mysteries? Saving people from supervillains? Swashbuckling among the stars? Experience acquisition ought to reflect those goals.

This solves both the problem of ticking boxes and the problem that mechanical advancement diverts players from the “meaning” in favor of the “math”. Instead of the math providing the meaning, the meaning provides the math.