Notes

There is no algorithm behind the roster. I picked every character by hand, based on how well known I thought they were. That sounds unscientific because it is — but any automated measure I could think of was worse. Screen time favours filler. Wiki page length favours whoever had a fan willing to write it. Popularity polls only cover the top hundred or so and say nothing about the long tail, which is where the interesting decisions are.

So I did it the slow way, and I did it twice, using two different approaches that catch different people.

Pass one: walk the arcs

I went through every arc and pulled out the characters who actually matter to it — the ones who are in the narrative a lot, or who change where the story goes. This is the pass that builds the spine of the roster. If a character carried an arc, they are in, regardless of whether anyone remembers them ten arcs later.

Pass two: walk the groups

The second pass is the one I think makes the game work. I went through memorable groups— the Supernovas, the Straw Hat Grand Fleet, crews and organisations with a strong identity — and took their members, including ones who are individually obscure.

On paper that looks like padding with characters nobody knows. In practice it is the opposite. Almost nobody can name the seventh captain of the Grand Fleet cold. But if you remember that the Grand Fleet exists, you will work your way toward them, or at least get close enough that the name surfaces. The group is the handle. Without it those characters would be impossible; with it they are a satisfying dig.

Manga only

I focused on manga characters and manga images throughout. Partly for consistency, but mostly because I love Oda's art and to me that is what One Piece looks like. It is also why the whole site is built to look like a manga page — hard black borders, screentone, no soft edges anywhere.

Deep cuts, but not cruelty

I wanted a layer of characters that genuinely works a fan's brain: people named only in an SBS answer, or on a Vivre Card, or who appear for a single panel. The line I tried to hold is hard but not impossible. A deep cut should feel like something you could dredge up if you thought about it properly, not something you could only know by having the wiki open.

Partly that is because I wanted the people who really know this series to be rewarded for it. If you have read every SBS, something in here should pay you back for that.

450 is less than it sounds

Most people cannot name 450 One Piece characters. Sat in front of a blank box, that number looks absurd.

But almost nobody actually plays that way, and you should not either. Once you start working arc by arc, or crew by crew, it turns out you know far more than you thought — and it is much more fun than staring at the input trying to summon names from nowhere. Giving your memory somewhere to stand is most of the game.

The hard part was not the picking

I expected choosing the characters to be the difficult bit. It was not. Two other things were much worse.

The first was getting characters into the right groups. A lot of characters belong to several at once, change allegiance partway through, or sit in a group that only really exists in retrospect. Every one of those is a judgement call, and there are hundreds of them.

The second was making sure a typed name finds the character you meant. One Piece has an enormous number of similar names, shared surnames, and characters known by an epithet more than a name. It is genuinely easy to type something reasonable and have it land on some near-namesake you have never heard of. Most of the work behind the input box is spent on that problem rather than on spelling.

If you find a name that resolves to the wrong character — or one you are certain should be accepted and is not — there is a feedback link at the bottom of the quiz. Those reports are the main way the list gets better.