⚠️ Note: This page was written in 2017. The concept is still interesting, but the skill levels and player examples should be considered only for historical value.

I occasionally run into situations where I want to describe someone’s skill quickly to a 3rd party, but I don’t have an easy way to do it. The current popular system seems to be to use the “D1 D2 D3 D4” results of waterfall brackets to categorize players, but I’m not a huge fan of this system because division results for the same player can vary widely from tournament to tournament.

For that reason and because i’m bored at work, I’ve made a set of 5 (technically 6) groups that I think roughly describe player skill, as follows:

Group 0. Complete casuals and people that don’t play smash.

Group 1. Can’t reliably execute even normal b&bs, have very little control over their character, frequently SD or do the complete wrong move by mistake, have little to no concept of spacing. Generally are aware of some but not all of the fundamental ATs, and are usually poor at implementing them in their play.

Examples - Random people that show up to your locals, people that go 1-4 in placement pools at regionals, your brother you taught to up-tilt and Z-cancel

Group 2. Can spam their character’s basic b&bs and are at least vaguely aware of b&bs of all characters. May know what DI is but can’t really apply it. Know normal movement and edgeguarding options, but don’t have the experience to effectively pattern-recognize and apply them. Often have 1 or 2 areas of surprising competence (good recovery mixups, good DI, good movement) but suffer in the rest of the game. Generally poor spacing, can avoid hits briefly but will mess up their spacing or do the wrong move by accident after even a bit of pressure. People that have been to a few tournaments but are generally referred to as “randoms” are usually in this group.

Examples - Bombproof, Apparition, most random people you play online, that “ham meat” guy or whatever his name was from socal, people that are happy getting a couple wins in PPs or bracket pools at nationals

Group 3. Can execute all their b&bs with no issues. Have developed recovery mixups, and can edgeguard effectively against most characters in a variety of situations. Apply pivots, shield drops, fox trots, etc. smoothly as a part of their movement and spacing. Can recognize and escape untrue combos. Rely on improvisation less and generally are aware of most common situations, and have optimized answers for a few of them. Have some understanding of spacing and are much less likely to just jump into attacks, but can still be frequently caught by common setplay and tricks in the neutral. Often have difficulties with patience, and still generally win or lose by how effectively they can “get in”. Will still be destroyed by the game’s best players, but can take stocks and generally function against the vast majority of the competitive scene. This is the typical level of long-time players that don’t have a strong desire to reach the upper levels, or ambitious players that have been playing for a couple of years. Regions that are considered “deep” or “well-developed” are generally because they have a lot of players at this level. This is a big group, with a substantial range in skill.

Examples - Wookiee, SotoH, Cobr, Daniel, Finio (RIP)

Group 4. Have mastered most of their character’s “standard” play. Neutral at this level is more nuanced, with an understanding of both their and their opponent’s goals and tools, how they change with % and DI, how they can be countered, etc. Players at this level are much less reliant on “getting in” and getting a big setup hit, and are much more capable of edging an opponent out by repeatedly beating them in neutral or repeatedly exploiting a particular flaw in their opponent’s play. Players at this level make better use of all their available options, and know both sides of almost all common situations. Fluid movement, recovery mixups, DI, solid spacing, etc. are all a given at this level and players begin looking for more nuanced strategies and more specific optimizations in order to beat the competition. These players are usually able to put up at least some sort of fight even against the best players in the game.

Examples - sHEERmADNESS, BarkSanchez, Banze, Shalaka, Fireblaster, Firo, Hero Pie

Group 5. Play at this level is extremely refined and accurate. Techniques that 3s would consider difficult or hard to execute are hit almost 100% of the time. Players at this level can maintain extremely good spacing at all times, often giving them the aura of being impossible to hit. Knowledge of rare and specific situations is pervasive at this level. Faced with a difficult situation, 5s will often find a solution most players have never seen before. These are the best players in the world, and those close to taking their spots.

Examples - SuPeRbOoMfAn, Alvin, Gerson, Wario, Jouske, Wangera, Fukurou, K Y S K


A few notes:

I think 5 groups is a good compromise between being specific enough to be useful and being general enough to have very meaningful differences between the groups. These categories being large, it may be useful to describe a player as a “high 2” or a “low 5” when relevant. That said, things always get blurry on the boundaries, and what one person considers a low 3 another might call a 2.

In general, the amount of time it takes to rise a group increases as the levels rise. Many players rise from 1 to 2 in a few months, but 2-3 generally takes longer, usually after at least a year or two. 3 to 4 is a longer climb still, and reaching 5 usually takes even the most motivated players the greater part of a decade.

About scaling: I chose these categories to reflect what I see as the rough groups in the current meta. As players get better and the skill gap between the best in the world and a completely new player gets wider, the scale needs to be adjusted to still be useful. There are two obvious ways to do this: rescale everything, so a 5 still represents the best in the world, or add new numbers, so 5 represents the same skill level and the best players are now 6 or 7 or whatever. Both ways have their merits, in my opinion the second option is probably slightly better but rescaling might also be necessary if players begin improving drastically faster, making the gaps in the lower ranks less important.