Round type vote

[Firebarrage]

I agree 100% on some secrecy being required. That's what the first 1/3 roll does. It ensures that while we do get round types people like more often, you really never know for sure that the weighting was in effect.

[Wisecrack34]

Maybe a 60/40 split would be better, makes sure you don't get too good of an idea of what it may be

[Firebarrage]

Posting at ZeWaka's request. Heres some pseudocode for this thing.

Code:

# player_votes is a list of lists.
# player_votes[i] = list of round types the player voted for
# I assume that round types are treated as integers by the game

def select_round_type(player_votes):
    distribution = [0.0]*NUM_ROUND_TYPES

    if(rand(1,3) == 1):
        return old_select_round_type()
    else:
        for player in player_votes:
            vote_weight = 1 / len(player)
            for vote in player:
                distribution[vote] += vote_weight

    round_selector = rand_float(0,len(player_votes))
    for i in range(0,NUM_ROUND_TYPES):
        if(round_selector <= distribution[i]):
            return i

    return 0

[MrMagolor]

add construction as an option for goon1 2day