Thread by @hasufl

One for my MEV friends: I always said that EIP-1559 has no impact on MEV, but is that true? Assume we are still in PGA world and no bundles

1. in EIP-1559, the goal is for many txns to have the same priority fee (1-2 gwei, wallet default)
2. miners tend to group same-fee txns

3. as a front/backrunner, how do you communicate your target position to miners? Using less gwei would put you behind the entire batch (bad case) or even out of the block entirely (worst case). Using more gwei would put you ahead of the entire batch (bad case)

4. If the above were true (and remember only PGAs no bundles), then EIP-1559 would effectively lead to batch settlement instead of ordering by gas price which makes front/backrunning harder.

However:
a) It depends on how miners group their txns. Standard used to be by gas price, but what breaks the tie between same price txns? Can either be randomized or first-seen.

b) If it‘s first-seen, backrunning changes from a bidding war to a latency war. Think a seeing a 1 gwei txn and then immediately sending the miner your own 1 gwei txn as well. So backrunning MEV only changes but isn‘t removed.

c) We naively assumed that there are no bundles, but this thread shows by both searchers and miners
*want* this txn type to exist. When they can communicate exact ordering preferences, the total value of the block goes up, at the expense of the users.

So bundles will always be invented, only question is they are invented as a private/proprietary sidechannel between some searchers/miners or globally available for all. So IMO futile to long for a pre-bundle world.

Summary: I posited that EIP-1559 effectively batches many txns, breaking the way searchers used to express ordering preferences in a PGA. Do you agree/disagree? Share your thoughts