Players beat the best score I set for a level, what do I do with the stars?

Star ratings need a target number of moves, and the target you or a program found is sometimes not the real best. Players will find shorter ways and your stars then look silly. Here is a rule for the ratings that also notices when the target itself is out of date.

Why this works

A target set by you is only your best guess, and players will test it. Treating a better result as a sign that the target is wrong, not that the player cheated, rewards the clever and shows you where the level needs fixing, as long as you verify it before accepting.

When not to use it

Skip it when a level has no clear best result, such as an open sandbox or a game of chance. Stars from a move count would be arbitrary there. Reward something the player can see and control, like collected items or time.

In the full entry

• What to build, step by step
• Starting values and the reason for each
• 3 common mistakes and the fix
• 22 lines of example code with 6 checks that pass
• How your AI shows you that it works

Say this to your AI

Use the Spielerando entry "Players beat the best score I set for a level, what do I do with the stars?" (read_entry star-rating-from-moves-and-a-stale-par) in my game. Build it step by step, and show me how you checked that it works.

Or skip this and connect the MCP: your AI then finds all of this by itself.Connect the MCP