A discussion arose in another thread over how many archetypes to use and where to draw the lines.

Most customizable games end up with some frequently appearing fleet types. In the broadest sense, you can employ a classic swarm, where you attempt to overwhelm your opponent with a large quantity of relatively cheaper items. Or you can aim for big beefy units that tend to last a long time and dish lots of damage, but you only end up with a few of them. Beyond that, what archetypes you get depend upon what historical or fantasy setting you are in and how the game is designed. With a game's archetypes, you often get a spectrum. Hybrid designs that fit across two archetypes are often possible. I'd guess that most customizable games don't end up with much more than the 9 that Shmitty used in classifying the regional data:

1. Rhymerball: Rhymer and 6+ squadrons

2. DeMSU: Imperial Swarm featuring Demolisher and 4+ activations

3. Rebel Gunline:

4. Rebel Carriers

5. Rebel Swarm

6. Generic Rebels

7. General Imperials

8. Imperial Wedges

9. Imperial Swarm

The three biggest poles are probably: 1. What are you doing with Squadrons? None, few (50 points), a medium amount (70-80 points), or a full set (110+). 2. How many activations? 3. Are you taking big and beefy ships or small ships? There's an extra dimension to this in that upgrades can make a lighter ship more expensive and therefore beefier, or maybe "how upgraded are your ships?" is a fourth question (discuss).

All three of those questions require you to commit points in your fleet. Those points are not available to spent toward answering one of the other two questions. Answering those questions leads to fleets that don't exactly fit the archetypes, or which could fit multiple archetypes. Take PT109's build that has won two Regionals. There are five activations, so its clearly in swarm territory. Its got two ISDs with Motti, so its also clearly in the neighborhood of big beefy units. It does this at the expense of squadrons and upgrades. In short, I can see how it works according to the archetypes, but I'd personally not be quite sure how to call it. And I think that's going to be an ongoing recurring problem: If we're seeing 110+ points of squadrons, we're going to be clear that we've either got a Rhymerball or some kind of Rebel Carrier build. 70-80 points, even with a carrier, especially if they're mostly anti-squadron, and I'm not so sure. At that point, you're a hybrid in some way.

Then there's the question of sub-archetypes. Two MSU lists may work substantially differently from each other.

So opening discussion: What does the community think about archetypes? What is the best way to classify what we see in terms of fleets?

It's difficult, fleets that look the same can actually work in completely different ways, depending on ship marks used, leaders and upgrades. Just as an example even if you stick with just one ship the CR90 it gets complex.....CR90a and b swarms are totally different beasts, with a potential of a mixed CR90 swarm as well.....

It's difficult, fleets that look the same can actually work in completely different ways, depending on ship marks used, leaders and upgrades. Just as an example even if you stick with just one ship the CR90 it gets complex.....CR90a and b swarms are totally different beasts, with a potential of a mixed CR90 swarm as well.....

This. A CR90B swarm with Mothma is going to fly differently than the same exact swarm with Rieekan. The first will probably try to dance around at Medium Range, whereas the second will cram all seven/eight ships down your throat.

It's difficult, fleets that look the same can actually work in completely different ways, depending on ship marks used, leaders and upgrades. Just as an example even if you stick with just one ship the CR90 it gets complex.....CR90a and b swarms are totally different beasts, with a potential of a mixed CR90 swarm as well.....

This. A CR90B swarm with Mothma is going to fly differently than the same exact swarm with Rieekan. The first will probably try to dance around at Medium Range, whereas the second will cram all seven/eight ships down your throat.

And we all know you can't eat more that two corvettes for breakfast.

It's difficult, fleets that look the same can actually work in completely different ways, depending on ship marks used, leaders and upgrades. Just as an example even if you stick with just one ship the CR90 it gets complex.....CR90a and b swarms are totally different beasts, with a potential of a mixed CR90 swarm as well.....

This. A CR90B swarm with Mothma is going to fly differently than the same exact swarm with Rieekan. The first will probably try to dance around at Medium Range, whereas the second will cram all seven/eight ships down your throat.

Sure, but both take advantage of high activation count, using multiple attacks to overwhelm defense tokens, and mobility to improve positioning. It's enough to put them together in broad archetypes.

I will say that defining the archetypes and sorting the fleets into each was one of the more challenging aspects of the regionals data project.

It was difficult to balance granularity with sample size. There was a huge variety of permutations within each archetype, but they did end up sharing enough common threads to be worth sorting together.

It's difficult, fleets that look the same can actually work in completely different ways, depending on ship marks used, leaders and upgrades. Just as an example even if you stick with just one ship the CR90 it gets complex.....CR90a and b swarms are totally different beasts, with a potential of a mixed CR90 swarm as well.....

This. A CR90B swarm with Mothma is going to fly differently than the same exact swarm with Rieekan. The first will probably try to dance around at Medium Range, whereas the second will cram all seven/eight ships down your throat.

And this will be accounted for in the statistics of Admirals used, I think.

My thought about archetypes: I think its a good classification. Going forward it may be useful to track not only the archetype but a subtype (or a secondary type) of the fleet. This way one can get not only a general statistics about archetypes, but a more detailed one about trends within the same archetype as well.

The main archetypes as I see them now are:

1. Imperial bombers (6+ TIE bomber squadrons or damage equivalents)

This may need refinement, but I believe the correct way to track it is by looking at anti-ship avg damage/round)

Rhymerball is a subtype of this.

2. Imperial swarm

4+ activations, not an Imperial bomber fleet.

DeMSU is a subtype (as well as my fleet)

3. Imperial wedges/Imperial gunline.

3 activations or less. At least 2 command-3 ships. Not a bomber fleet.

4. Generic Imerials

Everything else

5. Rebel Bombers

3 Bwing/Scurrges or damage equivalents

6, Rebel gunline

Needs more detailed definition

7. Rebel swarm

4+ activations, Not a bomber fleet

8. Generic rebels.

Everything else

Here's a question: Why is the line "4" activations on a swarm? Having played a bunch of fleets with 4 activations, they simply don't feel that much like a real swarm to me. Now, 5 activations definitely does. You almost certainly have to trade out on squadrons, upgrades, and/or move to small ships, and anything above 5 definitely makes that trade-off. At 4 activations, I feel like I get the best balance between a lot of different worlds and moreover, in the present meta, it feels almost a necessity. You don't want to give your opponent any more of an activation advantage than you have to, and you want to get as much of one as you can. On the whole, most of these fleets feel very hybrid-ish. They tend to have a bit invested everywhere. The 4th activation prevents a swarm fleet from being too lopsided against them. They've decked out a couple of medium ships or one large ship and/or brought a medium squadron contingent without really committing to heavy bombers.

In short, I think the more extreme an archetype is, the clearer it is. Right now, I think the biggest problem is some fleets that really aren't swarms are getting lumped under swarms. They are really hybrid fleets of some variety.

Here's a question: Why is the line "4" activations on a swarm? Having played a bunch of fleets with 4 activations, they simply don't feel that much like a real swarm to me. Now, 5 activations definitely does. You almost certainly have to trade out on squadrons, upgrades, and/or move to small ships, and anything above 5 definitely makes that trade-off. At 4 activations, I feel like I get the best balance between a lot of different worlds and moreover, in the present meta, it feels almost a necessity. You don't want to give your opponent any more of an activation advantage than you have to, and you want to get as much of one as you can. On the whole, most of these fleets feel very hybrid-ish. They tend to have a bit invested everywhere. The 4th activation prevents a swarm fleet from being too lopsided against them. They've decked out a couple of medium ships or one large ship and/or brought a medium squadron contingent without really committing to heavy bombers.

In short, I think the more extreme an archetype is, the clearer it is. Right now, I think the biggest problem is some fleets that really aren't swarms are getting lumped under swarms. They are really hybrid fleets of some variety.

I personally would also draw the line at 5. The question is - how to characterize 4 ship DeMSU fleets?

I understand that archetypes for statistical purposes are important, but with such a high degree of variance even amongst members of the same archetype, and no way to account for user skill which plays such a overwhelming factor in this game, I don't think we can ever get to the point where our named groups become anything but the vaguest possible reference, and thus only the most basic help to us.

In fact, I think in some sense being taxonomical about the game is to cheapen it. Sure, it contains a limited pool of permutations by which we can build fleets, which means fleets in competition will gravitate toward to the cards and ships with more point/power/min/max return than others so it is possible to group some of them together. However, I think what is beautiful about this game is that just when we think a list is too strong, someone comes up with a way to beat it, and the archetype list grows. Or at least, they look at an older, broader, archetype in a new way no one has before and uses it in a new way to handle a new threat whereby an older, broader definition grows to include a new sub-archetype, each with an infinite variety of permutations under it as each admiral tweaks an upgrade here or there.

TL/DR - Archetypes are a good statistical tool, but for Armada I can't but feel they hinder creativity, especially among new players, so I think we should use archetypes cautiously. Let's not make them become a box in which players think they have to adhere to in order for their fleets to be "acceptable."

Fair points, Broba. I think the use of "archetypes" is in understanding how different kinds of archetypes are currently performing in the meta. I think it also provides a foundation from which to do the deeper work of building a fleet well. And absolutely if we use them to "box" in, then that's not going to help players be creative. Archetypes are a good way of getting a handle on the present meta and its performance, and upon the community's collected wisdom to date. I'd see a parallel in the study of the opening in chess. Certainly player skill is going to determine the course of the game, but one can hardly argue against getting the best position possible from the opening. It is real easy to simply snowball a good position to victory. The same thing is true in Armada. You want to have a good handle on the possible fleets and their types out there and roughly how they work. That's a bit more complicated here because the game is constantly changing as new units get released.

In the end, all we're aiming at it here is trying to understand the world a bit better. A Tiger and a household cat still belong to the cat family, despite whatever other differences they might have.

Here's a question: Why is the line "4" activations on a swarm? Having played a bunch of fleets with 4 activations, they simply don't feel that much like a real swarm to me. Now, 5 activations definitely does. You almost certainly have to trade out on squadrons, upgrades, and/or move to small ships, and anything above 5 definitely makes that trade-off. At 4 activations, I feel like I get the best balance between a lot of different worlds and moreover, in the present meta, it feels almost a necessity. You don't want to give your opponent any more of an activation advantage than you have to, and you want to get as much of one as you can. On the whole, most of these fleets feel very hybrid-ish. They tend to have a bit invested everywhere. The 4th activation prevents a swarm fleet from being too lopsided against them. They've decked out a couple of medium ships or one large ship and/or brought a medium squadron contingent without really committing to heavy bombers.

In short, I think the more extreme an archetype is, the clearer it is. Right now, I think the biggest problem is some fleets that really aren't swarms are getting lumped under swarms. They are really hybrid fleets of some variety.

I personally would also draw the line at 5. The question is - how to characterize 4 ship DeMSU fleets?

Hybrid build most likely. This is why the subclass is important.

If you have an Imperial Squadron DeMSU, that right there tells you a lot about the list. Mind you, a 4 GSD list with Aggressors and some aces won Vancouver Regionals so I would put in that style of classification

So my take on archetypes.

I like them, I want them becuase it helps people understand what has won. This will lead to people trying their own way out and figuring out what works for them and what does not.

Brodafett brings up a good point though that it may cheapen Armada. Though, I am not sure. There is already overarching Archetypes. There is nothing that can be done about that. However, your own subclass feels important, we'll at least to me. Each list can do that to a degree.