There's something deeply satisfying about becoming stronger. Video games have capitalized on this especially well, often taking care to provide very visual systems of levels and xp. Due to the commonly numeric nature of video games, it's possible to figure out far in advance exactly how much stronger you will become. This is far better than the feedback-less nature of real life exercise and possibly why I find it so hard to do consistently.
I've been thinking about this a lot lately because I'm trying to understand my favourite games on a deep level, so that I can iterate on the things I like. And levelling up is one of my favourite parts of a game. But it can also be my least favourite. So I want to talk through a couple of systems for in game character growth and a couple hypothetical ones I've been thinking about.
The first and most obvious way to handle this is to not have any form of in game character growth or change. This is very common in shorter games. This is the simplest to design and also has the advantage of not providing a distraction from other more relevant mechanics.
Another approach is to pile most of the actual character change onto equipment, and then provide a very bare minimum level up system. This is very common in metroidvanias and classic roguelikes, but has also been implemented in Zelda and Terraria. In this case levelling might slightly increase your damage or hp, but overall doesn't provide significant long term changes in your capabilities. A first level character would be roughly equivalent to a max level character given the same equipment. I still remember watching a friend of mine, new to classic roguelikes, excitedly spending several hours grinding to level 30 on reproducing worm masses while learning Angband. Convinced he had broken the game, he descended to the next dungeon level and was immediately eaten. His armour was bad and he'd forgotten to bring enough escape scrolls, so being 25 levels higher than expected earned him nothing.
Breath of the Wild used this approach to great effect. By flattening the level up benefits, the game could become truly open world. You could go anywhere and do anything because all fights could be overcome with sufficient determination and player skill. One of my biggest criticism of RPGs like Divinity: Original Sin is that while they often seem to masquerade as somewhat open world (with respect to each area) the sheer power gap between levels often leaves you a very linear path available.
Of course, the downside to this approach is that combat can get to be very samey. Breath of the Wild avoids this by focusing on making puzzle solving and exploration large parts of the game, and ensuring that extra skills do come regularly in the form of equipment and treasures (which goes back to motivating exploration).
Here's a question: What's the difference between levelling up and finding a new item? Both increase your characters effective power or capabilities. The way I see it, the difference is the source of the power gain. Level ups are intrinsic to your character. Regardless of what activities you do and where you do them, it all contributes to the same system. You can choose to navigate the world in whatever order you wish, but level 2 must be reached before level 3. Equipment drops are tied to location, however. You have to go out and find equipment. Another key difference is that usually equipment is easily changed to react to situations, perhaps motivating maintaining multiple sets of different equipment. By contrast, once set, level ups are very hard to change (although it's often possible).
Of course, having both an equipment system and a levelling up system can occasionally knock things off the rails. One of the leading causes of unbalanced encounters in, for example, DnD is over or under-geared characters. This is why 5e specifies exactly how many magic items of each rarity a party should have for each level of their career. DnD relies on the GM to casually drop in the appropriate items at appropriate times, which can be a tall order. You could go a step further with this, of course, and treat equipment as part of the levelling up process, as I did when I wrote a TTRPG (which I may blog about someday, but probably won't post). It has obvious immersion disadvantages (how did "levelling up" cause you to spontaneously manifest a bigger gun?) but can provide a much tighter game balance. The immersion problems can go away with plenty of downtime baked in ("oh you went out and bought that gun last night while everyone was sleeping").
The downsides are that this method discourages exploration and can rob powerful emotional moments, like the recovery of an ancient and powerful magic artifact. The advantages are tying everything into one system and keeping the balance tight. Also, playing Batman becomes plausible. Think about it! Batman is actually very difficult to make in any RPG system, because his power is his use of items and wealth. But if all characters are assumed to have the same level of money available, then how do you represent Batman? Sure, there have been attempts (that usually focus on making him very skilled), but they always miss the mark to me. Why not just treat wealth as another form of character power and balance it as something you can specialize in?
That brings us onto proper classic level up systems, where your character has a single numeric level representing their worth in the eyes of the universe. This has a number of advantages and flaws. Advantages are being simple to implement and understand, and providing decent feedback for players on their progress. But a levelling system is pointless without decision making somewhere. That's why often, each level might come with a choice of different power ups. This also allows the abstraction of the concept of a "level", such as in Hades. You could call your level in Hades the total number of power ups you've received, but due to the variance between them, it wouldn't make as much sense.
These are common and exist in many forms. An important distinction then, is linear vs non-linear. A linear system has stats being relatively linear functions of your level (perhaps weighted by your class or type). This is common in JRPGs and makes more sense when you want to have more levels available. By making it a calculable function, it's easy to just put a very high level cap in and go. On the other hand, this sort of system can also become unbalanced at higher levels, due to the less fine control. Non-linear systems work in other ways, like Hades does, or by providing very different levels of power at each stage, such as tower upgrades in Bloons Tower Defence.
Linear systems have one other major pitfall: the gaps between levels change as you progress through the game. Recall the case study of Divinity: Original Sin. A 1st level character is exactly half as strong as a 2nd level character. This creates a very rough early game where you have to follow the listed level of enemies exactly to figure out how they compare to you. But the difference between a 2nd level character and 3rd level character is only 1.5 times. And it gets lower the further you go! By level 10, an 11th level opponent is only 1/10 stronger than you, a difference that is negligible next to an effective build and good use of tactics! This can feel frustrating and limiting in the early game while leading the player to make conclusions that don't hold up in the late game. There is an easy way to make the early game smoother, of course, and that's to pull a Pokemon. Ever wondered why Pokemon starts you at level 5 instead of level 1? I bet this is partially why (the other reason is to counter some of the problems of making a symmetric game, but that's a whole other blog post).
The total number of levels is an interesting aspect to consider as well. TTRPGs rarely go higher than 30 because of the necessity of modifying paper stats every time you level, as well as the pacing requirements. If you only meet once a week or less, play for a few hours, and want to get all the levelling down during downtimes, then it would take a whole year to get to level 50. Longer if you wanted each level to last longer than a single session. Collapsing the level space also makes each level feel more impactful which can be very satisfying (although it does come with immersion problems. How does fighting a group of goblins for a day give me the ability to survive an axe to face without even a scar?) Longer systems are only viable in video games where the bookkeeping is done for you.
So can we say non-linear systems require more work, but generally provide a better expirience? Perhaps, but only in the sense that something with more careful and personal design will always provide a better expirience than something more generated. And, as always, generated content tends to have longer playability which can also be advantageous depending on what kind of game you're making.
But I think the real strength is when you pair the two systems. Consider Pokemon again. Levelling, which controls basic stats, is a very linear system. But, because there's so many levels and no choices are required, they often fade into the background as a kind of passive upgrade. Evolution and move choices, two nonlinear upgrade systems which connect to the level systems are much more exciting and provide the desired degree of personalization. While each are very spaced out across the levels, having 6 mons collapses it and they all occur at a regular enough pace that it all works out.
Choice making is probably the key to getting all of this right. A concept seen a lot in conceptual TTRPG design talk is that of the "dead level", a level where the player only gets slightly higher numbers. Consider as a case study the DnD 3.5 cleric and the DnD 5e cleric. Seriously, look at those level up tables and tell me what you notice. Dead levels plagued early DnD for a number of reasons and one of the reasons Pathfinder took off was a concentrated effort to eradicate them (although their clerics didn't properly take off until they introduced archetypes, in my opinion).
While more levels might seem on the surface level to create more choice, dead levels, levels without anything interesting, space the game out and often feel empty and disappointing. And the best way to get rid of dead levels? Reduce the level cap and compress more. Which reduces playtime.
A final issue worth noting is that getting a big "level up" popup can be quite immersion breaking if you're trying to focus on a specific mood. While not an issue for games that are "tactics with an excuse plot" type, it can be serious for very emotional games. It might be possible to counter that by doing some kind of diegetic level up process? Perhaps something like when your character gets some down time you're offered a chance to practice "magic, combat, or lockpicking" and that determines what new skills and stats you get next? I can't think of any real game examples like that off the top of my head, but I'm sure someone has to have done it. It could make for quite an interesting game if your character scaled like a typical RPG protagonist, but nothing was ever given any tight numerical description. No, "use fireball for 2 * magic stat + 34 fire damage", just "use fire spell". It would require good use of fluff and art design to work right, but it could make for a very impactful and more uncertain expirience.
So we're right back where we started. A carefully thought out and deployed system will always win. Bare minimum systems have advantages in simplicity and keeping players focused on other areas of the game while complicated ones can become games in and of themselves. Having multiple interlocking systems is a reliably good approach, but much more work.
I have more I want to say about this, but I also don't want to do any more writing on this today, so I'm going to stop here for now. Maybe this will be my first blog post to get a part 2? Let me know what you think down in the comments. Haha, just kidding, there are no comments! Suck it, losers.