"There are three ideas in this post:
1) change of distribution algorithm for starting distribution to make it more even

2) increase territory maximum to 16 to reduce the luck driven and sometimes tedious 8x8 end game

3) rating change based on performance measured by attack wins - losses/2 making it unprofitable to sit with one territory. "

1) Great. You can fail before the game even starts because all of your dice are next to big stacks. If the distribution is more even, its more about how you manage your attacks. Though yes, keep it the same for once the game starts. It makes the game more exciting because when you hit "End Turn" you cross your fingers for a great distribution.

2) It needs to be tested to see if the 16x16 end game actually happens or not. Sandbox server! I would think it wouldn't...but you never know.

3) Great - as long as this fact is VERY visible to players. I think this makes the game more interesting too - people will take greater risk for the sake of getting points. However if people don't know about it, the current "turtling" will continue.

Overall I think this solves most of my qualms with the current system. Youve got my vote.

Well about the turtling stuff. What I really see for a problem is that people always try to kick the weakest player. "I will die happy, if you die first". It sucks. They should go for the strongest player, should try to get 1st. However the current scoring system supports kicking weak players, while you leave free field for the dominator, you will finish 2nd never the less. Thats whats wrong, and thats what makes me unhappy in current kdice matches.

i'd like to help you test-play this new method :)
Also, if a player dies before his first turn, he should get 0, instead of [0-5/2=-2.5]. It's only fair.. Losing a game without playing is a b*tch :p

*shrug* losing before you get a turn is a risk you take just by playing. It's plenty fair.

so here youa are plotting, while Iactually take part in real life. you guys got a strange kind of timing nevertheless:
1st dice random distribution.
Im fine with doint the slot thing for 1st round but keep it as it is right now in rest of the game. for the reasons already given in this thread.
2nd 16 dice max.
Personally i simply dont like the idea.
cause attacks get really uncalcuable. People will complain why they lost 16-8 attacks while theres actually a serious probability for doing so. It will turn the game into one big moaning i fear.
3rd scoring ideas sound interesting. But If you want my opinion i say no. call me conservative. Because i am.

"Personally i simply dont like the idea.
cause attacks get really uncalcuable. People will complain why they lost 16-8 attacks while theres actually a serious probability for doing so. It will turn the game into one big moaning i fear. "

It's -already- one big moaning. There are currently many topics about reasonably likely things happening.

1. even dice distribution
I don't much like this idea because it seems unnecessary. Seems to me it might make a lot less fighting, if people have very even stacks theres a lot less opportunties for attacking. But maybe it would work out.

2. 16 dice max
Great idea, I think. Less endgame, more fighting, more craziness, more fun. What could go wrong? Increasing the number of dice you get in the beginning in conjunction with this would be cool (and why not give all players same number of dice?). I'd also like to see slightly bigger maps, if this were at all possible.

3. Scoring based on attacks vs losses
I really don't see why this is a good idea at all, it would totally change the nature of the game. It seems like a clumsy way to go about penalizing players with 1/2 territories who don't attack, with many unwanted side-effects like rewarding mediocre players who go on killing sprees. It would fundamentally alter the strategy in a hard to grasp and opaque manner, for me ruining the beautiful simplicity of kdice (i.e. play to get the highest placing). There must be lots of other ways to fix this problem that dont change everything so drastically. Though I do think the scoring system could improve, I don't think this would be a good improvement.

I would suggest testing this after increasing the max dice, as the max dice thing would change the dynamic so much that you may find the problem of non-playing players changes or disappears. Same with the change in distribution, it may become unnecessary with an increase in max dice. So my approach would be to first try out 16 max, then worry about the other ideas, since increasing the max number is a very simple, easy to comprehend change in keeping with the current rules and style.

Great game Ryan, and thanks for continuing to improve it!

Here's the test server for 16 max games:

http://kdice.com:8080/

I like the way xax thinks.

16 dice are great fun...