- Cross-domain MEV & the crucial conundrum of proposer selection - Christopher Goes
- Tarun Chitra - Towards Theory of Maximal Extractable Value
Talks at the HackerHouse, Paris 2022 WindRanger, EduDAO, BitDAO Chaired by Tina, Steward/Researcher at Flashbots
Cross-domain MEV & the crucial conundrum of proposer selection - Christopher Goes
- tendermint has a weird proposer sharing algorithm
- each validator has a counter
- decremented by current stake each height
- Github link cwgoes/tm-proposer-idris
- MEV attacks abound
- randomised proposer selection
- super linear returns to centralisation
- cross chain MEV is a hard prob to solve, lol
- cross-domain anticorr incentives
- Shared mempools?
-
“Cosmos is going to be a lot of fun next year…”
Question: What if the total capital to realise the arbitrage doesn’t exist. - You’re bridging it - The block proposer gets to choose - Is the value of L1 tokens the MEV they can capture?
Tarun Chitra - Towards Theory of Maximal Extractable Value
Even rotten sandwiches can taste good in a while
Link to paper is here
- Thought: Do you ever find utilitarian MEV?
- Constant function market makers
- e.g. uniswap
- sandwich attack with . have slippage limits
- invariant functions
- price and quantity impact can be the same
- One assumption that is always made: almost all arbitragers are risk neutral, money losing arbs are when someone didn’t compute the backwards trade correctly
- g is continuous and monotonically increasing
-
The amount of quantity: concavity of the tasing function implies price and quantity slippage limits are equivalent
- How do we say this in quantity space (min max price)
- is the slippage limit less than the ratio of the kappa, mu, you can upper bound the sandwich profit
- How do we talk aboutut MEV for aggregations of sandwich attacks
- Routing? how much worse does my route get if a sandwicher gets in there somewhere
- Link to the study of traffic congestion
- Braess’ paradox