oh ye, the streaks i am talkin about are generally of the 8v8 dice, before u all explain to me that the odds of 2v2 is much more biased to the def! :)

This....this is actually the Grid's department, I'll call him in.

YOU CAN FLIP A QUARTER 100 TIMES AND GET HEADS EVERY TIME. IT IS STILL RANDOM.

Gotta agree with Kim there. It's still random eventhough it may seem unfair.

Yes, well lets see this, how many attacks does a game last? I have no idea. 50 rounds are quite possible. So 50 rounds, 7 players, say 2 attacks a turn in the middle. That makes a game 700 attacks long.

Well you are right that a strike of 10 has a chance of 1:1024. Well if a game lasts 700 attacks you have roughly a 75% chance at least one will have 10 losses in a row.

Other than that, random numbers are generated in computers by "pseudorandom" that means, that means yes, if you know the parameters the pseudorandom machine works at, you could predict the next dice throws. We don't know how the JavaVM implements the random machine. But there is a major and good corporation (Sun) behind it, so trust it to be done well. Some random machines on linux operating systems even take some data from /dev/random, which overs hardware generated entropy, and is therefore not (much less) perdictable. Typical default Hardware entropy is e.g. seek times on your harddisk, the time the harddisk reading head needs to reaches a specific sector on the disc is a bit random in limits.

Yes, pseudorandom can have effects on reaccuring patterns. But this is very obscure (like a 1:10000 tilts from ideal white noise), and none of us could ever notice this in a game. This is definitely not the source of unfair felt long strikes.

Beside that, there are hardware random devices available on the market. But they are pretty expensive. A normal quality one costs 750,- bucks. There are even available ones which rely on quantuum mechanics and are totally unpredictable, but they cost around 3000,- bucks.

So I can say, if you are willing to pay ~1000,- bucks for kdice to use random mechanics also reliable online lotteries use. Feel yourself invited ;) However be prepaired that even with a quantuum mechanics random generator you will experience 10 losses in a row. At least every 2nd game you play, such a loss-run should occur for 1 player. And even 10th game you play, you should get yourself such a loss-run. If it wouldn't happen, random would be broken ;o)

Of course, there's not seven players throughout the game. :P

Well then correct my numbers by a factor of 2. I don't know, we just would need to know the approx. attacks per game.

So its every 3 games somebody gets a row of 10 losses, and 20 games you play, you get yourself a row of 10 losses.

Note that a row of 8 losses is already considered pretty unfair luck, you well get this every aprox. 4th game then.