Value accrual will move upstream since SUAVE block builders are the point of entry and reduce Ethereum security
Validators are expensive to run and you probably need an expectation of profit (or breakeven) to have security which would add additional costs on the “MEV supply chain”
Depending on how consensus works, it won’t actually resolve censorship concerns past some notion of “plausible deniability” since bids can be encrypted before inclusion (but my guess is this can be circumvented by an off-chain mechanism)
My guess is I’m probably wrong to have most if not all of these concerns, but wasn’t getting any answers on twitter, so wanted to bring the conversation here.
I am also under the impression that SUAVE will serve as the intermediary for every Ethereum block at some point in the future (maybe this was wrong but would be helpful to hear otherwise in case that isn’t actually the plan)
It seems that you deleted your original tweet, lmk if you want to repeat it.
Regarding the questions in this thread (numbers not related):
I don’t see other blockchains ever “enshrining” SUAVE in the way you describe. Maybe in the far future, when it’s proven as a decentralized block builder with a highly credible auctioneer and perfect reliability. But until that point, I expect all chains to adopt a version of PBS through which many block builders can compete. As @ralexstokes points out, that competition forces builders to pay most of their revenue forward to validators.
Yes, running SUAVE infra will incur addtl. costs on the MEV supply chain, but the goal is that this addtl. cost is more than offset by the value that SUAVE creates for users and validators in the supply chain.
I ended up deleting the post because I had some wrong assumptions on what SUAVE was going to be. At some point I saw or heard something on how the goal was to be the only builder on Ethereum and L2s and that Flashbots was working with the Ethereum Foundation on their roadmap to include it - think there was some general confusion about what the plan was since not a lot of info was shared at first. After talking with y’all it was clear that this wasn’t the case so didn’t want to fud anyone.
Re 1. I agree that at some point it might make sense to have an “enshrined” block builder but not now. Imo as long as the enshrined builder’s consensus is leader-based on the MEV Chain, you run the risk the majority of value being captured one layer too early. Also in any scenario where a decentralized builder is enshrined it should probably be an extension of Ethereum’s consensus rather than a separate chain, else it seems inefficient to have a separate validator set.
Re 2. A thought that keeps coming up is why not fork SUAVE, run it on a smaller validator set but on the same orderflow? The fork would have less operating costs and could split the difference with the block leader. Of course there is friction to doing this in practice, but I guess the point is that it seems like the only moat (for any builder) is exclusive orderflow. I think it’s fine as long as the user opts-in to the exclusivity but just food for thought.
Overall, I’ve slowly but steadily come around to the idea of SUAVE thanks to conversations w/ @Quintus . Definitely think that some combination of operating costs and exclusive order flow (maybe preferred is a better word here instead of exclusive) .
Re 1. It’s not a foregone conclusion that an enshrined builder would be part of Ethereum instead of its own chain. The reason is both technical and political. Technical, because that builder would need to support many different domains with different consensus rules, block times, smart contract languages, etc. Political, because you need to bridge different ecosystems, and it may be more tenable for Ethereum, Cosmos, Binance, StarkNet etc. to use a neutral zone (think: the Switzerland of MEV) instead of agreeing on any one’s home chain.
Re 2. Yes, SUAVE needs some form of exclusivity of resources over other builders i.e. some moat. But this is a general rule in the economy, nothing specific to MEV auctions. SUAVE’s moat will likely come from network effect, i.e. it will have more orderflow, more connected domains, a stronger validator set, more searcher and more searcher strategies, and more executors.
At the risk of starting to sound like a broken record: builders have every incentive to find a credible coordination mechanism to avoid paying forward to validators. Clearly the part of value that a validator can find building themselves is hopeless, yet to the extent that the chain value proposition is to provide a permisionless coordination mechanism, it seems like they might eventually use it to credibly coordinate toextract the part of the value that the validators can’t see without the builders.
So I think that part of the argument is a bit more subtle and would need to understand the nature and magnitude of that gap.
Yeah, it’s an interesting point: if you make trustless coordination easy, can builders/searchers use it to increase their own bargaining power over users/validators, which is the actual goal? My intuition is no, but I haven’t thought about it much. Definitely a risk to keep in mind.
Better default algorithm for local block production that could get very close to the MEV in a builder’s block
MEV Oracle which could estimate within some band +/- 1% amount of MEV available in a particular slot
The idea being that if the proposer knows the amount of MEV that is available before their deadline to commit to a builder’s bid they can always end the auction and fall back to the local block. This would keep builders honest with their bids because the proposer could always refuse to accept an artificially low bid. The reputation of low-ball bid builders would provide enough social collateral to keep them honest.