I don't think a leaderboard based on "time to solve" is a very good criteria. It simply rewards people who see the challenge earlier and have time to complete it. I think throughput and latency would be better criteria. What TPS do you support?
Granted, this might be somewhat "pay to win" but I still think it would be better.
I'm thinking real world markets tend to reward first movers who deliver adequate solutions with the cheese, but surely there's leaderboard room for complementary wine of some objectively substantial performance vintage that may take a bit more time to refine for a balanced pairing---that is, so long as you're not already pre-game drunk at the wheel when the RFP opportunity window closes given misaligned engineering trade-offs tend to have real consequences too.
Granted, this might be somewhat "pay to win" but I still think it would be better.