#fakemon types

I’m not especially familiar with the Great Lakes region, so I’m afraid I couldn’t tell you much; doing your own additional research on the climates and landscapes around there should give you some good ideas. A lot of water-types would probably be likely, though, as well as a decent number of ice-types since it’s relatively far north.

As mentioned in another recent ask, I don’t think it matters very much as long as all types have some kind of representative. A commenter on said post mentioned that the first two gens had only three and then one new ghost- and dragon-types as well, so you can technically get away with as few as one or two small lines per type if you like; personally I think that was more Game Freak still figuring out what they wanted to do rather than a good idea to emulate nowadays (especially considering those types... couldn’t even actually be used to their fullest potential given their total lack of meaningful STAB in Gen I...), but you can if you want, and you could probably also do it better than GF did in G1/G2 by, y’know, giving them proper STAB and so on. I’d err on the side of at least three lines (of any size) per type—I usually try to get at least ten individuals per type, which easily works out to three or four lines, but that’s me—and then do the rest to taste.

I think the truth is that it could be anything, at least for most of what you’ve named. Type alone is not always an indicator of habitat. Some water-types live in salt water and need an ocean, others live in fresh water and could be in a river or lake. Grass-types could be most comfortable in a garden or an open plain or a temperate forest or a tropical jungle... or they could be cacti and be more comfortable in a dry and hostile desert. Dark and especially normal don’t say much of anything about where something would live at all; they’re not really associated with any meaningful “elements” or terrains or anything. Factor in that at least some of those will be dual-types, and the other type may or may not indicate where it should live instead. If your grass-type is also a water-type, maybe it should live in an ocean kelp forest instead of a land rainforest... or maybe not. It could work either way depending on what the fakemon itself is. The creatures and objects your fakemon are based on are going to matter more than the actual types if that’s what you’re worried about.

The only thing I think I can meaningfully suggest is that your region obviously won’t have a lot of rivers or lakes and would be otherwise generally landlocked, if you’re short on water-types. Beyond that? I’m afraid I have no idea, at least based on the specific types you’ve named.

It is a tricky balancing act, and I’m not sure I’ve got it down pat myself, especially because I don’t play competitively anymore and was not exactly stellar at it even when I did. I do try to keep reasonable stat ranges and movepool diversity in mind and try not to outright break anything, but honestly I have no idea how well I’m doing that. I think I can identify which of my fakemon would probably be “OU” and which ones are garbage, at least when considering a metagame consisting only of my creations, but I’m basing that on outdated fourth-gen-era competitive knowledge, so... yeah. tl;dr I can certainly try to offer general suggestions, but unfortunately I can’t give you any sort of guarantee that you’ll be able to “perfectly balance” this fakemon or any others you create. Sorry!

The easiest way to handle it would probably be to give it lackluster stat distribution, or at least stat distribution that hamstrings the “brokenness” caused by the type combo. If you’re worried about its defensive strength, just give it bad defensive stats, so even neutral hits are threatening to it. Maybe make all of its stats mediocre instead of outright bad or outright good, a balanced jack-of-all-trades like the actual canon non-legendary water/dragon, kingdra. Or maybe just embrace its defensive strength but make it terrible at attacking. You can also make sure that it has a shallow movepool (pokémon with quadrupedal, ungulate-like designs tend to have poor movepools anyway, actually; there aren’t as many TMs obviously compatible with that body shape as there are compatible with, say, bipeds or things with claws), or give it a lackluster or even partially-detrimental ability if it comes to that.

You can also make an effort to utilize the few “counter” types you have (dragon, fairy, steel) frequently enough to give this starter a run for its money in-game. Around the time the starter evolves, make sure it’s running into wild dragon/steel/fairy-types, and NPC trainers who use them. Use one or more of those types for a late-game gym or even an E4 member, perhaps. If you can stick one or both of steel or fairy onto at least one of your other starters, that would probably also help keep the rival challenging, too.

Out of curiosity, does it have to be a dragon? Obviously it should be whatever type would make you the happiest, but just to play devil’s advocate you could probably also go water/fairy or water/psychic or something and still have a solid kirin design. Both of those are decent type combinations as well, but they have more weaknesses than water/dragon so they aren’t as defensively potent.

Honestly, though, “metagame”-wise, meaning considering all of your fakemon and possibly also canon pokémon and how this starter might do “competitively”... well, so what if it’s a good type combination? Sometimes starters are really good (see blaziken)! Sometimes they straight-up suck (see meganium)! It happens! In terms of metagame balance, there’s actually no reason that starters need to be on par with one another. It’d be nice, sure, but it’s hardly a requirement. Some pokémon are just plain better than others, full stop, so there’s no reason to stress about it too much.

It's not really something I consciously think about often. Usually the concept itself will dictate whether any given fakemon makes sense as X type. If you want your fire starter line to end up fire/steel but there's no metal anywhere on the first stage, don't make the first stage a steel-type, and so on.

If you haven't got the concept yet and are just starting with the type then obviously it's up to you! If you want the line to be steel the whole way through, design it so that it makes sense as steel the whole way through. Do whatever works for what you're doing, basically.

In all honesty I can only think of a few reasons it would even matter whether a line is dual-type from start to finish, and they're largely for reasons of "game balance" if that matters to you (e.g., a steel- or ghost-type starter is probably unfair given early reliance on normal-type moves like tackle and scratch, but by the time it evolves most pokémon should have a wider variety of moves) or because you feel like you have too many/not enough of a certain type in your dex (e.g., if you're overloaded with electric-types but are desperate to keep your pseudolegend electric, try having only the final stage be electric and do something else with the other stages). Or maybe if you feel you have too many/not enough dual-types in general.

To elaborate on starters in particular: my personal rule of thumb is that I avoid any dual-type that would mess up the rock-paper-scissors effect in any way for the basic stage only. Beyond the basic I don't care, so any types like steel or rock can be added from that point on. I would never have a starter begin as a bug-type, for example, because it has interactions with both grass and fire and so would change the SE/NVE relationship somehow no matter which starter you put it on (grass/bug: double-weak to fire, fire/bug: double-resists grass, water/bug: neutral to grass). Bulbasaur as a poison-type is fine because poison doesn't interact with fire or water, but it does interact with grass so a fire/poison or water/poison starter would end up like the bug examples above. I was okay with making Faelan's fire starter, telerond, a fairy-type because fairy's only interaction with the starter triangle is that it's NVE against fire, and no one cares if telerond resists its own secondary STAB for these purposes. A type like psychic or dark has no interactions with any traditional starter type at all, so swimmole beginning life as a psychic-type is also fine with me.

Hm… a lot of grass and water, obviously. Lots of bugs. Poison would also probably be pretty common, with a decent amount of flying-types. Fire and ice would be uncommon, also obviously, and I imagine there wouldn’t be much in the way of steel-types, either.

The remaining types don’t really have anything to do with things that are or are not common in rainforests; fairies and dragons aren’t even real, after all, and fighting has to do with the concept of martial arts and physical combat rather than specific plants, animals or climates. I’d say you could just distribute the last ten types to taste.