ANONYMOUS wrote: We do intend to run more samples when assessing your submissions, but 1000 games is still a lot, so if your agent is not consistently winning with 1000 games, it is probably a close enough thing that it's not "consistent". So submitting this agent would probably be gambling that it is good enough that on the samples we take it outperforms the benchmark. You should definitely be aiming higher, especially since SatisfactoryAgent is not a high benchmark. It is intended that to earn the marks you should be consistently and easily outperforming it (after all, there is another hidden benchmark tier above that you should be trying to beat as well!). The purpose of taking a thousand or more samples is to counteract the effects of "bad luck" and losing "by chance". Your agent should be good enough that a thousand games should be sufficient to prove it is better than another agent. If after a thousand games your win rate is basically the same, you cannot claim to be consistently outperforming the benchmark agent. Gozz