Despite the impending crap tsunami, I believe there are at least a few people out there who will show a genuine desire to make good stages that will properly balance fun, challenge, and creativity. If you're one of those people and you've never tried your hand at level design, here are some kernels of advice that hopefully will help with that.
1. The time gate is there for a reason.
Nintendo recently revealed that not all of the toolbox will be available for you to use the first time you play the game, and that the unlocking of content will be staggered over the course of nine days. Of course, this went over like a lead balloon, and already too many butthurt crybabies to count are registering their disgust on the Internet as we speak. People think they're ready to dive right in and start showing Nintendo why they should have been the president of EAD1 this whole time, but I'd be willing to bet nine out of ten of them don't even come close, and even that figure may be generous. Nintendo knows what they're doing with this feature, and I find it rather rude of people to presume they know more about what goes into a good side-scrolling platform level than the people who have been doing it professionally for three decades.
Imagine it as a pottery class. You're not going to make a beautiful vase on the first day. You're not even going to make a decent ashtray. Relinquish control; let Nintendo show you to work the wheel, shape the clay. It's like Ghost, and you're Demi Moore, and Nintendo is Patrick Swayze, and it's getting kind of warm in there, and crap this got weird.
2. Your first 30 to 50 levels will be garbage.
3. Your first 10 to 25 levels will be hot garbage.
4. Your first five levels will be flaming, fetid, radioactive, unapproachable garbage.
Just so you don't get a big head.
5. Give your land forms weight.
For a counterexample of what I mean by this, check out Jeremy Parish of USGamer as he makes his way through the nine-day intro. Parish knows his levels are terrible, and he's right, but he's not homing in on a key reason why. I hesitate to say his levels are slapdash, because he's the editor-in-chief of a major gaming journalism website and he's always busy writing articles and pumping out videos and he doesn't have all the time in the world to invest in this, but by the same token you can't help but think the guy who has written entire books on the anatomies of certain game series (including Mario!) would have a bit more of a penchant for video game architecture.
Parish lays down a bunch of thin, meatless rows of blocks with little regard for dimension and spacing and then throws it up and says "yep, that's a bad level". Well, no kidding. It doesn't feel anything like a natural world that was there before life as we know it. Blocks usually don't obey the laws of gravity in the Mario universe, but there are also large structures and walls and mountains. The land should have contiguity and weight. Don't be afraid to create hulking land masses made of lots of 1x1 tile blocks. In fact, your first few levels should probably be mostly that, with only the occasional floaty bits. It will help anchor your level and give it a comforting feel of semi-reality, as opposed to looking like a terrible Mario Paint drawing.
6. Generally, shorter is better.
Something I've seen in Mario editors that predate Mario Maker is that when people get this blank canvas in front of them with a wealth of possibilities, their inclination is go big. Incredibly big. Stupid big. They make these enormous, byzantine levels with tons of secrets and nonlinear pathways and side tracks, and in the process they forget that while the best levels of Super Mario Bros. 3 and Super Mario World did have a bunch of bonus material and Easter eggs tucked away, they could also often be beaten in only a minute or two. Furthermore, the secrets and goodies were usually not mandatory, nor did the level rely on you finding them to make it an effective and/or passable level. Don't get carried away and make your level some 15-minute slog with terrible midpoint placement. Have some restraint. It might be the single most important quality in solid level-making.
7. Center your level around only one or two core mechanics.
Although it has a (mostly overblown) reputation for extreme difficulty, "Tubular" from Super Mario World is one of the most perfect video game levels ever created, and an exemplar of both this tip and the previous one.
The level foreshadows its looming deviousness with a Clappin' Chuck. The enemy's early presence tells you they're going to factor into the level in a big way, but you don't yet know how. Soon you get to the P-Balloon powerup, which inflates Mario's body and lets him float through the air. A whole level devoted to it is a decent idea at face value. What propels "Tubular" from superficially clever to Pantheon-level excellence is how it bakes the limitations the P-Balloon imposes on Mario's movement into the challenge factor. The P-Balloon only lasts a limited amount of time before Mario deflates and falls to his death. It's easy enough to get to the second one, though the lack of solid ground nevertheless creates an encroaching sense of danger and urgency. After that, however, the level introduces projectiles. Now you must maneuver Balloon!Mario below and between and around Volcano Lotus's fireballs and Puntin' Chuck's footballs. The projectiles move and spread out at a speed and range that makes it difficult to pass by them, because they are often situated somewhat below you, and Mario has a difficult time lowering his altitude while inflated. In addition, a lot of them are placed with pixel-perfection so that they often nail you from above as well. Getting to the third and final P-Balloon can be quite a challenge indeed, though far from an insurmountable one.
And it doesn't wear out its welcome; the player in the video below beats it in about 45 seconds. Even as SMW levels go, it's incredibly short. It's sparsely populated, but everything has a clear reason for being where it is. It doesn't need to be any longer than it is, and it knows it. Making it any longer would just make it exasperating and enervating. It has a clear message and a clear objective, it has an interesting combination of ideas, it blends and communicates those ideas with crystal clarity, and it gets the heck out of there. Perfect.
8. Consider how you interact with enemies (and how they interact with you).
There is nothing inherently scary about a Goomba. Even the most novice player is not afraid of it. It walks at you very slowly, it does not attack, and it does not fire any projectiles. Therefore, to you, as a designer, Goombas might seem very boring and not worth using. You may be inclined to use enemies that take a more offensive tack against Mario or that have a more obvious set of qualities that make them a threat. And that's fine. But if you try to brainstorm ways to make Goombas threatening, then you're onto something, because if you're trying to make a series of connected levels, it gives you a curve to work along. A Goomba walking at you isn't a big deal, but what about a Goomba above you about to drop onto your head? What about a Goomba that's shot at you out of a pipe? What about a flying Goomba? a bouncing Goomba? a Goomba wearing a Spiny helmet? a horde of giant Goombas? If you can make simple foes work in the early going, then later on when you fill a room six levels later with giganto-Thwomps you'll be able to look back and see a more natural progression.
9. Distribute powerups evenly and fairly.
A mushroom should not be an oasis. Instead of putting a powerup at the end of a gauntlet, put it in the middle. Give players a chance to work their way up from mushroom to flower to leaf to Kuribo's Shoe and beyond. Creating an artificial scarcity of powerups is a cheap, tacky way of adding challenge to a level, and it will exasperates players to no end if they end up playing skillfully and make it deep into a level and then get punished and sent back to the beginning by one tiny mistake because there was no help in sight. Be reasonable.
10. Avoid enemy spam.
Speaking of artificial challenges, here's another one. If you can't figure out a way to make Koopa Troopas interesting without filling the screen with them, you're not a good level designer. It can also be relative to the space you're inhabiting; if the area is tiny and claustrophobic, one or two enemies might be all you need to push it to the proper degree of difficulty. Just don't go overboard. Enemy spam can be executed reasonably, if you make sure you give the player a fair and fun way to counter it, and if it makes sense in the context of the level. If you can't justify it in one or both of those ways, don't do it.
11. Kaizo traps aren't funny or clever, and neither are you for using them.
Kaizo traps are named after a famous ROM hack of Super Mario World called Super Kaizo World, which was, naturally, chock full of them. In case you're unfamiliar with the term or the concept, a kaizo trap is a death (or on very rare occasions, a non-death setback) triggered by the performance of an otherwise innocuous action that you normally wouldn't think twice about. The classic example is an invisible coin block over a normal-looking pit. You jump over the pit like you've done your whole life, except now there's a block there that you couldn't see; you hit it with your head, and you go tumbling into the death hole. There's also the type where you die by not defusing the trap before triggering a condition where it's impossible to avert it, most often manifested as dying during a victory outro. Below is the all-time classic hall-of-fame number-one über-example of being fooled by one such trap. (Skip to 7:26 for the money shot. NSFW, as he does spew forth a torrent of colorful language, though considering what he went through before that happened, it's warranted.)
There is a time and a place for the unfair troll move, but shenanigans like this almost always cross a line in the hands of an amateur. If they are the crux of an entire set of levels, fine, go for it. Some thrill-seekers go actively looking for that kind of thing. But if you pull that business in an otherwise normal level, people will turn on you, and your level will get one-starred into oblivion. Rest assured you will be the only one laughing.
12. Murder your darlings.
This is a quote from a 19th-century British writer named Sir Arthur Thomas Quiller-Couch, who I'm sure would be extremely proud of me for taking it out of context and applying it to video game design. In case you've never heard it, it means that you shouldn't be afraid to edit or omit something you made even if you personally think it's awesome. It might be too self-indulgent or arcane for a general audience. It might be too hard. It might just be too long. Although it's technically possible to clear that spike pit, the precision required might be excessive given the obstacles you've placed ahead of it. Don't be afraid to check yourself and say "that's too far". People will be more pleased with a delicious morsel than with a giant hunk of meat they can't eat. Remember that, and everything else I've said here, and you might just stand a chance of creating a halfway decent level.
