Ah, I was looking for this post.
I got the same barring for the UMich competition held yesterday. Regarding point #1, I believe it was for ranking first in Season 2, as I had not actually won the live event last year. I agree with the decision to promote fairness, however it does seem to be barring only some people going in to the competition with a super strong algo. I’m assuming, but haven’t checked, that the other Michigan friend here stomped the remaining competition.
Considering the UMich comp has passed now, I’ll share that even changing the configurations doesn’t remotely remove competitive advantage. I, too, had wasted a good deal of time preparing for the competition before being barred, and had redeveloped my algo from scratch to be able to simulate any configuration, as well as adaptable metrics for those configurations. I additionally added strategy configuration files integrated closely with the Map Maker which made it a trivial 30-40 minutes to develop a top-tier algo under a new configuration, which any veteran player can do with an intuitive sense of the game settings could use. The configuration problem is only amplified by the forum post that revealed the same set of configs was being used for each live event, which would’ve given me, and any other student that saw that post before it went unlisted…, several additional days/weeks of configuration-specific development, which conveniently you can just upload directly to the playground and begin testing.
Given how large these live events prize pools are compared to any winnings I’ve had so far, I think your best bet is to stay off the leaderboards and try again next year. This probably isn’t the end result C1 is looking for, though, so perhaps we’ll see further clarification about how fairness in the live events would be balanced with keeping the engaged part of the community interested in continuing to be engaged.