Skirmish Game Order/Turns

In terms of a unit's ability, think along the lines of Worms (the computer game). It is played on a square grid, and like in Worms, each unit's "turn" can consist of one attack and one instance of movement, in no particular order. Each unit has a set of three attacks that it can perform, and that set differs for each character. On top of that, units which have "trained" retain the attacks of their previous identity by keeping their old card underneath their new card. Attacks are structured like this:
-------------------
Recruit
[Speed-3/Armor-4/Health-4]

Slash - [Melee/Pierce-3]
0-2-5-6-6-9
If Slash damages, but does not kill its target, the user gains attack +1.

Intimidate - [Blast-3]
0-0-0-0-0-0
Target(s) receive defense -1. User receives attack +1.

Hilt Smack [Melee/Push-2]
1-4-4-4-4-5
This attack can be played immediately following a melee attack which fails to cause damage.
-------------------

To quickly explain some of that, the 6 values for each attack represent the strength as determined by a d6 roll. I think I've posted this before, but I did this because I was always bothered with the HeroClix (amongst others) system of a 2 d6 roll added to a set number. It gives every character the exact same attack curve, differing only by its starting value. I prefer the idea of some attacks being basically sure things, while other really depend on a good roll. Either way, there are several types of attack, the obvious two being melee and ranged. There are also passive attacks, which have no technical "range" requirement, shockwave attacks, which effect all units in a certain radius, and blast attacks, which are basically 1/4 of a shockwave, again with a set radius. There are also standard effects attached to some attacks. Push is pretty obvious, and Pierce, without going into too much detail, allows you to deal additional damage by meeting a certain requirement.

I realize that was a bit long winded, but maybe it will give you an idea of what I'm dealing with here.

First, to clarify the models thing. I started with the idea that each player would begin with three unit "credits." This could be spent on 3 level 1's, 1 level 3, or a 1 and a 2. Then, between each round, each player would earn two credits, to be spent however they wanted, though I would cap the total army size on either side to 6. This way, a player could make a strategic decision to either improve his units or expand his ranks. Having worked a little more on the game since, I am beginning to lean away from that, favoring a system where each player starts with 3 level 1's, and is able to improve them however they wish, but cannot add new models. This solved a very real problem of cost, since more freedom means I need more copies of each model/card to support all possible army configurations.

For the purposes of this discussion, assume 3v3 always. Models will have a health and armor value. When you launch an attack on another unit, your attack value is compared to the target's armor value. If you exceed it (defender wins ties unless otherwise noted) you cause 1 damage. This, I hope, will keep complexity low. Once a unit has been dealt damage equal to its health value, which I assume will happen over several turns, it is destroyed and removed from the field, though it can be used again next round.

Also: Your defense of the 2d6 system is confusing, but probably valid. What I meant to say is that with 2d6 and an attack value of, say, 5, your character has a 1/6 chance of attacking with a value of 12, and a 1/36 chance of attacking with 7 or 17 (the extremes). What I don't like is that the bell curve is always the same: Your characters attack value +7 always has a 1/6 chance, no matter what the initial value, and your ultimate attack will always have a 1/36 chance. I just feel like melee attacks, for instance, should be much more even in spread, while high range archery should have a significant drop-off, requiring a much higher role to even have a chance of success.

So yeah, the game. It's interesting that you would have that reaction to the chess idea. Having gone over this with some people outside of the forum, I've gotten very similar responses to the chess thing, but with several different reasons. What is it about that idea that people think is not good? I've actually be leaning towards it more and more as time goes on.

So Arenacraft has very little story. It's based on an old idea I had for a video game (basically a Worms clone) that seemed too difficult for a first attempt at Actionscript game creation. The story of that game was that two guilds were at war over a series of territories. At their disposal are three training facilities: A soldier's barracks, a temple, and an archer's lair. New soldiers are recruited from each facility, but can train further at any place of their choosing. The main hitch of the game is supposed to be the training system. A novice (archer level 1) that trains at the barracks becomes a rogue. And a recruit (soldier level 1) that trains at the archer's lair also becomes a rogue, but both retain the move-set of their past roles, making their character build very similar, but ultimately unique. This carries forward to the third level, where there are even more possible unit types.

I guess the board game has two "themes" in theory. The first is the straightforward theme of two warring guilds. You are attempting defeat your opponent in a series of strategic battles needed to control the area. First player to win 3 battles wins. Obviously the whole training thing is part of it as well.

The secondary theme is sort of a Super Smash Brothers thing. I know this is a very unrealistic idea, but I initially imagined that this would be a game system where I attach a Creative Commons License and encourage others to use it. I imagined that I would ask around the TGC community and see if some of the designers wanted to release expansions with their own characters, whether invented for Arenacraft or adapted from another game. The only design mandate would be some sort of leveling up system included for all units so that the overall idea of the game stays constant. Designers could certainly use the combo idea of my set, but there's no reason you couldn't use a more straightforward upgrade path where there is no choice involved. So with that in mind, the theme would basically be "A bunch of characters are fighting cause, you know, reasons..."

Also, my last word on the 2d6 is actually a picture. My issue is summed up in this graph, which shows the attack value curve of 10 different attacks using the 2d6 system.
http://oi43.tinypic.com/xcnddg.jpg