First game confused on dice

[Lamon]

I figured out the red die..but sometimes there is 1 or 2 dice not sure why. Also my choices are usually both people fall down or i fall down. I have never seen the AI have to make that choice it is always push or my guy knocks down. Are there different dice based on the player type? I litterally have 4 people on the field now due to KO's.

[Dode]

Have you played the campaign, which is actually an extended tutorial? I assume you've not read the rulebook, either.

I suggest taking a look here: http://forum.bloodbowl-game.com/viewtopic.php?f=61&t=5500

[Lamon]

Meh should have waited for a free weekend. Appreciate the help.

[Procrastinating Gamer]

I assume you're talking about the Block Dice.

If the Attacker has the same Strength as the Defender, you get 1 dice.
If the Attacker has a higher Strength than the Defender, you get 2 dice and can pick which one to use.
If the Attacker has more than twice the Strength of the Defender, you get 3 dice and can pick which one to use.
If the Attacker has a lower Strength than the Defender, you get 2 dice and your opponent gets to pick which one to use.
If the Attacker has less than half the Strength of the Defender, you get 3 dice and your opponent gets to pick which one to use.

A player can only assist a block if they aren't in any other enemy tackle zones (so the only enemy player they're adjacent to is the one they're helping to beat into a pulp). Players with Guard ignore this restriction. Each player assisting in a Block adds +1 to the Attacker's Strength. Assists work defensively as well, so every enemy who isn't marked by one of your players and is adjacent to the attacker gives the defender +1 Strength.

The Block Dice are always the same, all you can influence is how many are rolled and who gets to pick which dice applies. There's been some accusations that the AI cheats the dice roller, but it's honestly hard to tell if that's the case or if it's just a case of cognitive bias (ie; you remember the failures more than the successes) and we'd need a sample size of thousands of rolls to check.