I recommend looking at how match ups are done for fighting games...

Scroll down to find the charts..

http://wiki.shoryuken.com/Super_Street_Fighter_II_Turbo_HD_Remix#Consensus_Matchup_Chart

The basic idea is to get high skilled players, and run 10 matches between lists, and use the number won / lost.

For example, let's look at BBXX vs 7tie howl swarm. 5 wins for each would be fairly balanced, 7/3 would suggest BBXX has a good matchup. Etc...

We need to trade accuracy in the matchup for granularity in the list archetype definition.

I'd like to see dyadic data for my own purposes--...

How's your skill at coding?

I think the bigger issue is getting the data in the first place though.

I'd like to see dyadic data for my own purposes--...

How's your skill at coding?

I think the bigger issue is getting the data in the first place though.

Ah, I didn't make myself clear.

If we build an online engine that makes it easy for people to self-report their matches (yes, I know the danger to representative sampling), then the data would come in at a much faster rate. Building that engine requires some ability to code in at least .php, which is what I've learned but not mastered.

Depends on the language, but I did finish a bachelor's in electrical engineering, which required work in Java and C++ as well as RISC assembly. Most of my coding (or, perhaps, "coding") these days is in SAS and SPSS, but I know enough of a couple of different scripting languages to at least get myself in trouble.

Ah, okay. Have you done any work with .php and SQL?

There are people in other game systems that make a lot of money figuring out the paper rock scissors of their meta.

I think an important part of list design is to build lists that can handle Rock, Paper and Scissors. If you push too much into one of the archetypes, you might give yourself an advantage over some lists, but the skill difference will need to be much greater when you find that you've brought Rock and the other person brough paper if you want to win.

So when building a list, I always try to think about some of those examples and think about what tools my list has that I can use to counter them. I don't need to have something that will give me an auto win, just a fighting chance.

If we build an online engine that makes it easy for people to self-report their matches (yes, I know the danger to representative sampling), then the data would come in at a much faster rate. Building that engine requires some ability to code in at least .php, which is what I've learned but not mastered.

I don't think this is good unless you mean for TOs to report. It's not a great data set if it includes casual games or games where the players are unfamiliar with their list (or even the game itself). For finding these kind of balancing statistics the data needs to be as clear of variables as possible.

I think an important part of list design is to build lists that can handle Rock, Paper and Scissors. If you push too much into one of the archetypes, you might give yourself an advantage over some lists, but the skill difference will need to be much greater when you find that you've brought Rock and the other person brough paper if you want to win.

So when building a list, I always try to think about some of those examples and think about what tools my list has that I can use to counter them. I don't need to have something that will give me an auto win, just a fighting chance.

I am not sure you can really classify lists in this game as succinctly as rock/scissor/paper to turret/arc dodger/swarm. Is half a swarm still a swarm? Does a phantom really do better against turreted ships by adding a miniswarm? I would love to see more data on this.

Depends on the language, but I did finish a bachelor's in electrical engineering, which required work in Java and C++ as well as RISC assembly. Most of my coding (or, perhaps, "coding") these days is in SAS and SPSS, but I know enough of a couple of different scripting languages to at least get myself in trouble.

Ah, okay. Have you done any work with .php and SQL?

Nope. )c: I have people for that these days, actually, but I don't think my professional colleagues would want to work for free on my hobby.

I think an important part of list design is to build lists that can handle Rock, Paper and Scissors. If you push too much into one of the archetypes, you might give yourself an advantage over some lists, but the skill difference will need to be much greater when you find that you've brought Rock and the other person brough paper if you want to win.

So when building a list, I always try to think about some of those examples and think about what tools my list has that I can use to counter them. I don't need to have something that will give me an auto win, just a fighting chance.

I am not sure you can really classify lists in this game as succinctly as rock/scissor/paper to turret/arc dodger/swarm. Is half a swarm still a swarm? Does a phantom really do better against turreted ships by adding a miniswarm? I would love to see more data on this.

That's why I think about thing like: How will I deal with lots of ships. How will I deal with 1-2 falcons. How will I deal with Interceptors and Phantoms.

If I have enough tools to give me a reasonable chance of fighting each of those, then it doesn't matter if they bring 6 TIE's instead of 8.

Depends on the language, but I did finish a bachelor's in electrical engineering, which required work in Java and C++ as well as RISC assembly. Most of my coding (or, perhaps, "coding") these days is in SAS and SPSS, but I know enough of a couple of different scripting languages to at least get myself in trouble.

Ah, okay. Have you done any work with .php and SQL?

Nope. )c: I have people for that these days, actually, but I don't think my professional colleagues would want to work for free on my hobby.

The answer to that problem is to make your hobby their hobby too. Well at least that's what I've done so far.

Bazinga!

5ish archetypes then:

Fat Large (defensive, large ship loaded with mini swarm support)

Swarm

Arc dodging

Multi turret tank

2-3big

I know there is also rebel 4 but it doesn't seem as big right now. It could return.