Recently, Mal was wondering whether the dragon had too few hit points. I replied that I didn't have enough data, and here's the proof:
The blue bars represent the games I've played and the amount of life the winner had left at the end. The knight is on the left of zero, the dragon on the right. The orange line represents a target for the win distribution I'd like the final game to have. So very few ties, and a roughly Gaussian distribution on either side, but with the dragon as likely to win as the knight (assuming equally matched players).
It's going to be really difficult to achieve this considering the asymmetry in the game:
- the knight has fewer hit points
- the dragon has fewer dice, so needs to recuperate more often
- the dragon's attacks and defence are stronger
- the knight has the ability to attack and change range
- the knight has the ability to prevent all damage once
- the knight has the ability to combo attacks
But it gives me a target to aim for. Clearly what I need to begin with is a shed-load more data. That's where you come in - there's PIPs galore available for helping me to fill in those blanks!