How much can we constrain builders without bringing back heavy burdens to proposers?

[Pandapip1]

How does this proposal look? It doesn’t add too much of a burden to proposers (the worst-case computational complexity doesn’t change).

MEV minimization using randomness

I propose the following alternative, which would make MEV very hard for block proposers: Block proposers can pick which transactions to include or not to include in advance. The hash of all these transactions is computed. Let these be the initial orders for those transactions. A block is only valid if the following is true: The transaction with a median order (treated as binary numbers) is the next transaction included in a block. The median can be efficiently found in O(n) time with a select…