Controversies over Smash Bros. characters like Diddy Kong and Little Mac illustrate a larger concern that people have with Nintendo's famous fighting game: is it properly balanced? One of the most compelling arguments I've read says no. But then it asks a compelling follow-up: So what?

Forrest Smith, a software engineer at Uber Entertainment and Smash fan, has been picking apart the balance (or lack thereof) in Smash Bros. for months now in a series of highly regarded essays posted on his personal blog. All of his essays have made a big splash in the Smash community once he's published them—sparking lengthy discussions on forums like Reddit. Using data gathered from the SmashWiki, Smith examined the performance of individual Smash characters in competitive tournaments to track how their individual rankings (he calls these "power rankings") and the overall composition of character-based tiers changed over time. He finally completed this ambitious and very cool endeavor last week with a third essay that examines the tiers for Melee and Brawl specifically.

The distinction between power rankings and tiers is a tad confusing, but he finally got around to explaining that in the third essay (emphasis mine):

Tiers are a layer on top of power rankings. A tier is a group of characters that are considered roughly equal in terms of strength. A tier list simply an ordered power ranking list where all characters are also categorized into a tier.


So: "power rankings" refer to the performance of Smash characters based on their performance in competitions (i.e., how many times they win or lose). "Tiers," then, are the groups that characters are placed in based on their relative power rankings. Fighters like, say, Fox and Marth have always been considered "top tier" characters for Melee because they've consistently been at the top of that game's power rankings.

Here are the tier lists Smith pulled together for Melee:


...and the "power rankings" from the same game:

(Note: these are screencaps I took of Smith's graphs. They all have snazzy interactive features that make them easier to view in their original forms on his website, which you should go to anyways since all three essays are well worth your time.)


For evidence of Smash being unbalanced, you need look no further than Smith's charts to see that not all Smash Bros. fighters are created equal. If the game wasperfectly balanced, rankings and tiers based on characters, rather than actual players, wouldn't make as much sense in the first place. Over the course of several years, then, the status of some characters changed for better or, in many cases, worse. Smith charted the decline of Melee characters who seemed like promising contenders early on, but didn't live up to that promise:

Only some characters changed in the ranks, though. That's key to seeing how the game isn't perfectly balanced. Others stayed relatively stable:


A small group started strong as top-ranked characters, meanwhile, and never stopped being the most powerful in the group:

Ah, Fox.

It's tempting to see Smash's unbalanced design as a symptom of the game's two separate lives: as a whacky party game for four (or more) players and, simultaneously, a fiercely competitive 1-v.-1 fighting game on par with Street Fighter or Mortal Kombat. But what's fascinating about Smith's analysis is that he doesn't see either the tiers or power rankings as indications of quantifiable "truths" about how Smash Bros. works as a game. "Tier lists are not an objective statement on what the game is," he argues. "They can't be. The lists continuously change even though the game stays the same! Quite surprisingly they aren't even a subjective statement on what the game is. That's what I assumed they were all along, but it's not quite right. Quite surprisingly they aren't even a subjective statement on what the game is. That's what I assumed they were all along, but it's not quite right."


"Tier lists are an objective review of how the game was played," Smith concludes. "They're a reflection of the past!" He likens the competitive Smash community's ranking and categorization of characters to the way that "any professional sports league" measures the standing of relevant athletes. "They're merely a system for rating how well characters have done in tournament play since the last list. They don't define objective truth on game balance nor are they a predictor for the future."

Looking to the future, then, the obvious question is how Smith's analysis applies to the new Super Smash Bros. games for the Wii U and 3DS. Things are slightly different this time around since Nintendo has released updates that specifically address character balance. But Smash creator Masahiro Sakurai also said in an interview last year that he felt he was done balancing his latest game. So if the new Smash has the same lifespan as Melee or Brawl, it will presumably end up producing the same sort of data that Smith was able to use in his work.

Smith doesn't have that data for the new Smash Bros. since the game hasn't even been around for a full year yet. But there's still a crucial takeaway from his work on previous Smash games that's worth mentioning. Because rankings and tier lists aren't set in stone, they change over time. Often dramatically so, in cases where players only start to realize a character's true potential over the course of playing with him or her for months, even years.


In Smith's second essay, for instance, he highlights the "diamonds in the rough" from Super Smash Bros. Brawl—characters that jumped in the competitive rankings from 2008 to 2013:

Here's the chart for all Brawl characters, for a point of reference:


Notice how Diddy Kong started out at a respectable but modest point in the top ten of Brawl's rankings, and ended up in the top five by 2013. Donkey Kong, on the other hand, fell hard and fast over the same time period:

What caused these changes? Smith argues that each Smash game changed, often in dramatic ways, over time. Players discovered new ways to use the best characters. Weaker ones were unable to withstand the added pressure of refined techniques, and fell to the bottom.


It's important for Smash players to keep this type of historical evolution in mind. While Diddy Kong is the reigning champ in the new Smash Bros. right now, there's really no telling what could happen over the next 5 or 10 years. Who knows: he could end up suffering the same fate as the older Kong did. I, for one, am still hoping that Captain Olimar has his day in the sun. He probably won't, but can't a Pikmin fan dream?

Pointing out that rankings and tier lists for a competitive game change over time isn't revolutionary or surprising. So why did Smith go through so much trouble to make a seemingly basic point? It goes back to his original suggestion that Smash is an unbalanced game. In his view, the lack of balance and Nintendo's unwillingness to change or correct that is what makes Smash such a great game. From the first essay, emphasis added:

These discoveries didn't just take a little time. It wasn't days or weeks or even months. It took years to be discovered. In the internet age gamers flip their shit if a given character is considered overpowered or underpowered after a single week. Meanwhile in Smash Brothers it took half a decade for game changing depth to be found in Ice Climbers.

I think a large part of what enables this depth to be found in Smash Brothers is that the game isn't balanced. It's not a small roster of perfectly tuned characters. It's a big game with lots of characters that creates a huge and unexplored problem space. It's up to the players to explore the nooks and crannies and see what treasures they can find. I find that incredibly exciting and compelling.


Smith compares Smash to League of Legends, another popular competitive multiplayer game with a key difference: it's constantly changing thanks to never-ending updates that make some champions more powerful and leave others hopelessly nerfed. There are pluses and minuses to taking either approach as a game developer, of course. But Smith suggests that leaving a game alone the way Nintendo has with Smash ultimately gives the game more of an opportunity to evolve than it otherwise would, since it gives players ample room to explore for themselves:

Riot changes LoL all the time. Characters are regularly buffed. The nerfhammer is swung with reckless abandon. Ability sets are completely redesigned. The meta is meticulously influenced between seasons. I just can't help but wonder how the game would evolve if it was left alone like Smash. What would it look like after five or ten years? What amazing team comps would be found? What metas would develop? We'll never know, and that makes me a little sad.

The counter-argument in defense of an always-updating approach like League of Legends is that Riot can refine the game, or even experiment with new ways to balance it, safe with the knowledge that any glitch or gameplay upset can be addressed further down the road. Plus, a developer tweaking a game doesn't necessarily preclude players from discovering advanced techniques, and the game evolving in turn. I enjoy playing both games, so I see the merits of each approach. But I think what Smith is suggesting here is that the long-term historical trajectory of a game like Super Smash Bros. Melee or Brawl was almost entirely shaped by the inventiveness and tenacity of its fans. League players don't have the same opportunity, because so many elements in that game's evolution are beyond their control.


Players have been exploring "the nooks and crannies" of the new Smash Bros. since it came out—discovering new ways to unleash ever more powerful attacks and even cut down on character lag. But even as some have been doing that, many others have simply bemoaned the game's deficiencies—pointing to issues like Diddy Kong's overwhelming presence at tournaments as a sign of the game's weakness compared to, say, Melee.

Smash has only been out for the Wii U for a few months. I doubt that the game people play 5 years from now will be identical to the one they're playing today. But if we accept Smith's argument that the perception of balance, rather than balance in and of itself, is what changes over time, then the community's willingness to dive into the new game and search for more diamonds in the rough will determine how great a game the new Smash ends up being.

Read Smith's whole series of Smash articles here: Part 1, Part 2, and Part 3.

To contact the author of this post, write to or find him on Twitter at @YannickLeJacq.