Skip to main content
GameDev.net gamedev.net
🔒 Locked

Sports Simulation AI

Started by BrandonM Apr 15, 2010 at 2:19 PM 4 replies 2.2k views
Original Post
BrandonM
BrandonM
I'm looking to develop a sports simulation game, for fun. I've chosen the game of baseball for various reasons. One AI issue that is relevant to just about any sport is the handling of AI trades. Here are some questions/dilemmas: 1. When does a team feel the need to initiate a trade? 2. What parameters go into determining who to trade and what the requested return should be? 3. How does the opposing team determine whether or not to accept the trade? Any thoughts or suggestions?
hiigara
hiigara
What do you mean by trade? Buying and Selling Players?
BrandonM
BrandonM
The trading of one or more players from one team to another for one or more players.
Kylotan
Kylotan
At the moment, this isn't an AI problem. It's a domain modelling problem. You need to look at the real world example you're trying to model and formulate a set of rules or guidelines that capture that knowledge. Once you've done that, then you can look into what sort of AI algorithm might implement them.
IADaveMark
IADaveMark
Evaluate the relative utility of each player. This is based not only a relative skills of the players but also of the needs of the team to fill spots. For example, it doesn't do you any good to have 6 of the most skilled first-basemen in the league since you generally only need a starter and someone else that can fill in.

Once you list all the considerations, you can build a mathematical model for the needs of your team and the evaluations of the players - both yours and the trade prospects. (Oh yeah, toss salary in there somewhere, too.) The list will sort itself out by who is expendable and who is needed.

Of course, this is all rather complex... if someone can recommend a good book on the mathematics of utility theory to this fine OP, I'm sure it would help.
Dave Mark - President and Lead Designer of Intrinsic Algorithm LLC
Professional consultant on game AI, mathematical modeling, simulation modeling
Co-founder and 10 year advisor of the GDC AI Summit<
dashurc
dashurc
Treat it like real life. Each player has a perceived value. Each team has a budget. Each team has a goal (if the team is one of the top teams they will look to pick up a good player to improve the team in the short term, if the team is not as good, they might want to trade good veteran players for some young prospects in order to rebuild). By assigning teams budgets, mentalities (which change based on performance), and designing what type of players are good for teams with different mentalities. From this whether the trade can be accepted comes down to budget and fit (see InnocuousFox' advice).

That's what I'd do anyways. You can even extend this to free agents.

Topic Locked

This topic has been locked by a moderator. New replies are not allowed.

Sign in to reply to this topic.