Let Me Explain

By kokoloko. Art by Bummer.
Previous Article Home Next Article


Hello, I'm kokoloko. You may know me as the Underused tiering leader. I was the one behind the decision to use a brand new tiering system for XY UU, something which caused a lot of controversy. This article will explain both why I decided to do this and detail how the system works.

Let's start with the "Why?". First and foremost, I felt as if the traditional system was flawed at its roots. For one, it was too restricted in terms of what we were allowed to ban, due to the old philosophy of banning only Pokémon that were thought to be "broken." A silly restriction which, in my honest opinion, did nothing more than restrict what could potentially be a great metagame (say hello to BW OU for me). Secondly, it was far too slow to get the metagame to an acceptable state, taking literally years to produce decent results. Finally, it created tons of issues due to use of ladders as the vessel for suspect testing (ladder cheating, excess time consumption, etc). This is why I decided to create a new system that avoided all these problems.

How does it work?

First, the fundamentals. The main purpose of this system is to create a new tiering philosophy in which there is never anything overpowered in the metagame, no matter what the cost. We achieve this by banning everything that is potentially "broken" before we begin and then proceed to test the BLs by reintroducing them back into UU one at a time. This process will almost certainly come with collateral damage in the form of bans on "unbroken" Pokémon, but that is a small price to pay in exchange for a balanced metagame 100% of the time. To clarify, we are completely breaking away from the tradition of only banning things that prove to be broken. We are also not afraid to ban moves this time around, because there is literally no reason not to other than tradition. If a move deserves to be banned, it will be.

We only use the ladder to test suspects for a very short period of time; once the council is ready to vote, we vote and the next suspect is tested. This can take anywhere between two days and two weeks, which eliminates the time consumption problem I mentioned earlier. The most important thing, however, is that instead of banning stuff as the metagame develops, we ban everything that is deemed potentially broken by the council before the metagame even begins, and then develop the metagame ourselves by reintroducing suspects into the metagame, one at a time, to check if they were truly broken. Essentially, we tier backwards.

Now, let's get into the specifics. I'll explain this step-by-step.

We begun with "UU Beta," which was nothing more than a ladder that included everything that didn't make the OU cutoff. At this point, I picked an 11-man-council to make decisions on what was potentially broken alongside me. We used this "UU Beta" period as a playtest in order to come up with an initial banlist, banning Pokémon in batches. If six out of the 12 of us (11-man-council plus myself) voted something potentially broken, it left the tier to be re-tested at a later date. We banned Pokémon, moves, abilities, and items (Mega Evolution stones, but still) until we ended up with what we thought was a balanced metagame with nothing potentially broken in it. We reserved the right to amend our initial banlist if we later came to the realization that we missed something, a right which we ended up executing later on Tornadus-T and Zygarde. Including those two, but not the things we banned that later rose to OU via usage, we started with a total of 23 bans.

The next step was to make the tier official and begin testing every BL individually to see if they were truly overpowering. This is where we are right now. Tests will occur in the order of least likely to be broken to most likely to be broken, as to arrive at an ideal* metagame more quickly. Exceptions will be made in such cases that we believe reintroducing a certain suspect before a different suspect will give the latter a better chance to stay in UU (in other words, we're also working to keep as many of the suspects as possible in the tier).

Every suspect will be tested at least once, and everything will be tested once before anything is tested twice. Once we have tested every suspect once, we will probably begin testing them in groups, to check if they balance each other out without over-centralizing** the metagame around them. There is also a possibility that suspect tests will be open for public participation, using the standard ladder system, after the council goes through every suspect once.


Okay, those were the basics of my tiering system. There really isn't much else to say here other than you have to remember that this is the first time this system has ever been used, so I've been ironing out the kinks as we go along. On top of that, there's no guarantee this will produce "better" results than the traditional system, and, if at any point we realize that my system is producing worse results, I'm totally open to reverting back. That said, if you still have any questions or doubts that you want me to address, check out the XY UU Tiering System Thread, which contains an FAQ section, or send me a PM and I'll do my best to get back to you with a satisfactory answer.

*This tiering system accepts the fact that there actually isn't an ideal or perfect metagame that we're supposed to have, so when I say ideal, think of it as the closest thing we can manage to come up with.

**In this context, over-centralization means that the metagame is "balanced" and can deal with said suspects, but at a greater cost than benefit. Cost and benefit being measured in terms of metagame and teambuilding diversity.

Previous Article Home Next Article