Random Pokemon Type Generator

What a type actually does

Every creature has one or two of them, and every attack has one. When they meet, the chart decides whether the attack lands hard, lands soft, or does nothing at all.

The full set is Normal, Fire, Water, Electric, Grass, Ice, Fighting, Poison, Ground, Flying, Psychic, Bug, Rock, Ghost, Dragon, Dark, Steel and Fairy.

What that grid describes, mathematically, is a shape with a name, and the name is worth knowing because it explains why the game is any fun at all.

Drawing a type

1. Press Generate for one type. Good for a themed team, or for settling what your next playthrough looks like.
2. Ask for two and build a dual type, which is where the interesting maths lives.
3. Type into Contains to search, though the list is short enough to read.
4. Copy to keep the result.

Where this list comes from

The full set of types as they currently stand, checked by hand. The tool picks from that pool in your browser.

Rock, paper, scissors, and why it never settles

Take three things. Fire burns Grass. Grass drinks Water. Water puts out Fire.

Now try to say which of the three is best. You cannot, and the reason is not that they are equal in power. It is that "beats" here does not behave like "taller than" or "heavier than". If Ali is taller than Ben and Ben is taller than Chandra, then Ali is taller than Chandra. That property has a name: transitivity. Almost every comparison you make in ordinary life has it.

The type chart does not. Fire beats Grass, Grass beats Water, and Water beats Fire. The chain bites its own tail. Mathematicians call a relation like that non-transitive, and games built on it behave in a very particular way.

They never settle. In a transitive game there is a best move, everybody eventually finds it, and the game dies. In a non-transitive game every strategy is beaten by something, so the population of strategies keeps rotating. Whatever everybody is doing this month is exactly what creates the opportunity for the thing that beats it.

That rotation is the reason a competitive game can stay alive for thirty years. It is not balance in the sense of everything being equally strong. It is the total absence of a top.

Nature is playing the same game

This is not a game design trick. It is a thing that exists in the world, and there is a famous piece of fieldwork about it.

In 1996, Barry Sinervo and Curtis Lively published a study in Nature about a small creature called the side-blotched lizard, which lives in the western United States. The males come in three kinds, and you can tell them apart by the colour of their throats.

The orange-throated males are aggressive and hold large territories with many females in them. The blue-throated males hold small territories and guard their single mate closely. The yellow-throated males hold no territory at all; they look like females, and they sneak.

Now watch the circle. Orange beats blue, because an aggressive male with a big territory simply takes what the small territory-holder has. Blue beats yellow, because a male who is watching his one mate closely notices a sneaker. And yellow beats orange, because a male spread thinly across a large territory cannot watch everybody, and the sneakers get in.

Sinervo and Lively went back to the same lizards year after year, and the proportions of the three morphs rose and fell in a cycle. Whichever strategy became common created the conditions for the one that beat it. Nobody ever wins. The frequencies just go round.

That is the type chart, running in a desert, without a designer.

The generation where the circle broke

Here is what happens when the loop fails, and it happened in the very first games.

There were fifteen types then. Psychic was meant to have counters. The in-game hints said so. The strategy guides said so. A trainer standing in a gym in Saffron City tells you outright that Psychic types fear only ghosts and bugs.

Except that Ghost attacks did nothing to Psychic types at all. Not reduced damage. Nothing. It was a programming error, and it was never noticed before release, and the one damaging Ghost move in the game therefore had no effect on the type it was written to answer.

Bug, the other supposed counter, was carried by creatures too weak to matter. And Dragon, the other powerhouse, had exactly one attacking move, which did a flat forty damage no matter what.

So the loop straightened out into a line. Psychic beat almost everything, and almost nothing beat Psychic, and competitive play collapsed into everybody using the same handful of creatures. Transitivity had crept in through a bug, and a transitive game has a best answer, and a game with a best answer is over.

Two types invented to fix it, and one more later

The second generation of games arrived with two entirely new types, and both were repairs.

Dark was created to answer Psychic. It is immune to Psychic attacks, which is the strongest possible statement the chart can make. Steel arrived at the same time, resisting an enormous range of things and blunting the general offensive pressure.

That held for six generations. Then Dragon became the problem, and in the sixth generation a third type appeared: Fairy, which beats Dragon and, again, is completely immune to it.

Notice the pattern. Every type added after the original fifteen carries an immunity, and in each case the immunity points at the thing that had grown too strong. Dark is immune to Psychic. Fairy is immune to Dragon. Steel is immune to Poison. These are not creatures. They are corrections, wearing costumes.

And notice what the designers did not do. They did not weaken Psychic, or nerf Dragon's numbers. They added a new node to the graph and drew a new arrow into the tyrant. That is how you repair a non-transitive system: not by lowering the tallest thing, but by giving it something to fear.

Which is exactly what the lizards do, and nobody had to tell them.

Questions people ask

Is there a best type?

Defensively, Steel has the strongest case, because it resists more than anything else. But the moment a type becomes clearly best, the game has a problem, and history says a correction arrives. The honest answer is that a well designed chart has no best type, and every generation is an argument about whether the current one is well designed.

Why are dual types so much harder to reason about?

Because the multipliers stack. A move that is strong against both of a creature's types does four times damage, and one that is weak against both does a quarter. And a single immunity overrides everything, so a creature with the right pair can be untouchable by a whole category of attack.

Why not add more types?

Because every new type adds a row and a column to the grid, and each new arrow you draw changes matchups you did not intend to change. The designers have said as much: adding types makes balance genuinely difficult. Three have been added in three decades.

Does rock, paper, scissors really appear elsewhere?

Yes, and not only in lizards. The same three-way loop has been demonstrated in bacteria, where one strain poisons another, a second is resistant, and a third grows fastest when left alone. Non-transitivity is one of the ways nature keeps more than one kind of thing alive.

References

1. Sinervo, B., and Lively, C. M. (1996). The rock-paper-scissors game and the evolution of alternative male strategies. Nature, 380, 240–243. https://www.nature.com/articles/380240a0
2. Kerr, B., Riley, M. A., Feldman, M. W., and Bohannan, B. J. M. (2002). Local dispersal promotes biodiversity in a real-life game of rock-paper-scissors. Nature, 418, 171–174. https://www.nature.com/articles/nature00823
3. Pokemon Database, type chart and the Generation 1 and 2 to 5 charts, including the Ghost against Psychic error. https://pokemondb.net/type/old