My idea combines football manager with Ubisoft’s Chessmaster series. A game where you would manage a team of NPCs, teach them chess, choose a strategy and ultimately lead them in battle against opposite chess teams to win the league.
After months of planning, we decided to start building something – a first build, where I could test some numbers and do some simple balancing for the team match between the NPCs.
We had established that each interaction during the teaching phase would lead to either XP gain or loss. We narrowed all NPC skills to just two: Tactical (chess knowledge in general) and Mental (everything related to how the player is doing mentally).
Tactical would act like the ELO rating system, while Mental would multiply the Tactical skill – in order to simulate that players who feel well mentally play better, and vice-versa.
With the first build ready – which could have actually been an Excel document, but we decided to code it because we’d need it anyway – I managed to run a lot of tests and balance the Tactical and Mental states.
Still, I was not satisfied by the dice roll outcome for just the Tactical points. Huge Tactical XP differences did not seem to make much of a difference. That is to say: match results seemed pretty much random.
It was just as if you would play poker and although you’d start with two Aces, you’d just randomly win and lose all the time, even though the odds were in your favor.
We did not want the game to feel too deterministic (high Tactical players to always win against lower Tactical ones), but also not too random (where Tactical points don’t seem to make much of a difference at all).
The fix was easy to implement, but hard to conceive. I’ll explain it next time.
Leave a Reply