1) 8v8 fight

2) 7v8. unless the neutral is in the way for more attack then attack the neutral to gain another attack on him.

you should only attack an 8 neutral for 3 reasons:

A) if you are double the size of your opponent and want to gain more attacks/dice.

B) digging for a cut on him.

C) protecting a future cut on yourself.

other than that don't bother with attacking back neutrals.

8v8, 7v8, 6v8 the neutrals.

No matter how anyone responds, Maza already has the answers determined mathematically which only takes into consideration probability, not a player's relative risk aversion.

FLAG TO ALL ON THE BOARD EXCEPT CHAINBALDE - THEN KILL CHAINBLADE

So Panzer...

You generally play pretty aggressive and are sort of surly at times.

With that said, I also, have been screwed by Chain a few times in the past.

So what happened to cause such angst?

I play just as snow leopard and Marsyas. Let us try to evaluate mathematically these situations. (Of course, we have to make a lot of assumptions to get estimates.) First, I will look at the expected gain in difference of dices between the player and the opponent. In particular, this assumption ignores risk-aversion. Second, I assume that accounts are not cursed such that 8v8 wins with probability 47.1% and 7v8 wins with probability 27.4%. Let me know if you have corrections to these rough estimations.

1. Suppose you have 8 territories and +0 dices and you decide whether to roll 8v1 on a neutral or 8v8 on the opponent. If you roll 8v1 on a neutral, then you win with probability 1 and get +1 dice (because of one more territory). If you roll 8v8 on the opponent instead, with probability 47.1% you gain 1+8 dices (8 dices of the opponent and 1 dice because of one more territory), but with probability 52.9% you lose 7 dices. Therefore, the expected gain in dices is equal to 0.471*9-0.529*7=0.536<1. This analysis suggests that in expectation it is better to roll on small neutrals.

2. Suppose you have +32 and 15 lands, it is a bottleneck and you decide whether to roll 7v8 on the opponent or 8v8 on a neutral. If you roll on the opponent, with probability 27.4% you gain 16 dices (new 8 stack and 8 stack of the opponent is gone) and with probability 72.6% you gain 0 dices such that the expected gain in dices is equal to 0.274*16=4.384. If you roll on the neutral, with probability 47.1% you gain 8 dices and with probability 52.9% you gain 0 dices such that the expected gain in dices is equal to 3.768. This analysis suggests that it is better to roll 7v8 on the opponent.

It's way too early in the morning for mathematics

your first calculation however doesn't take into account the fact that with rolling grey, you don't influence the number of lands your opponent has. With rolling the opponent you have the chance to take a land from him, maybe cutting him (or at least getting one land closer to a cut). If you win, he is maybe one less further from a cut on you. With one land less, maybe he can't make a second attack,...

Taking grey isn't always bad of course.

I mean I appreciate you trying but it's impossible to simply "calculate" the best option in KDice. There are too many variables

case2: with that many reserves and a bottleneck, it's logic you do the 7v8 as well

In most realistic situations with a bottleneck, it's impossible to calculate. People just do the bottleneck 7v8 based on their gut estimation of the risk/reward and based on their playstyle.

In conclusion: I like and appreciate you doing the math, it's good to see some figures. However, for me, it only illustrates the complexity of a KDice decision which is impossible to put in a calculation

jurgen, would you mind possibly having my babies? i admire your wit, humour and intelligence. cheers to you mate!!!!

Gosh mate, now I would even consider it. But we would have to make sure they would have your playstyle and not mine because I don't want my kids playing conservadice all the time.