dYdX v4 Social Mitigation Strategy for MEV

[Kam]

Hi everyone,

I’m Kam from Chorus One, and we’ll be one of the genesis validators on dYdX v4. Two members of our research team at Chorus One recently published an extensive paper on MEV in dYdX v4. You can access the article here.

With dYdX transitioning to a sovereign chain secured by a diversified validator set, and given the substantial daily trading volume that is happening on dYdX, there is indeed a risk of malicious validators using their position to extract value through MEV.

Introducing a social slashing mechanism is a great step, as it is currently the most practical way to assess validator behaviors. The introduction of a committee actively addressing this issue is a significant positive development, and I’m looking forward to seeing how this will work over time. The beauty of blockchain is also that anyone can check validator behavior on-chain thanks to tools like what the Skip team developed, but having a committee dedicated to that endeavor is important to take clear actions.

dYdX stakers will be the very first line of protection, as they should stake with trustworthy validators to maintain a healthy validator set, with the Committee serving as a secondary defense when evidence of suspicious behavior emerges. The amount asked seems very fair considering how valuable dYdX is.

In our research article, our team demonstrated that the combination of a slashing penalty and the waiving of the unbonding period for re-delegation to a good actor are effective ways of addressing undesirable validator-driven MEV. The combination of these measures helps mitigate the risks associated with MEV extraction by validators.

The main challenge for the Committee would be to understand if a validator is a false positive or not, because discrepancies can arise due to different factors. For example, some validators may have their own data centers or dedicated servers, leading to better latency performance compared to other validators. Another example is network instability, which can also result in discrepancies between validators. Overall, I’m confident that with such an experienced and reputable Committee, they will be able to clearly assess such situations.

We’re very pleased to see discussions of this nature starting as they are very important for the success of the dYdX chain

[Ivo]

— posting in my individual capacity as a community member –

I am in favor of passing this proposal and introducing the committee. I have this view for a few key reasons:

1. MEV will not be clear-cut. Some posters in the thread seem to be assuming that scanning Skip’s dashboard alone will constitute a complete analysis of possible MEV events. That will not be the case. To present a convincing rationale for slashing, the community will have to do relatively extensive block-by-block analysis, including potentially trying to de-sybil multiple accounts and incorporating high-frequency price data from other venues. Skip’s dashboard is sufficient as an indicator of something to pay attention to but not as a case for slashing in and of itself.
2. Validators/community members have already signalled a lack of comfort with social slashing of any kind. It’s not at all compelling to assume, as some have done, that validators will police each other in the absence of a dedicated body. Without this committee i think the most likely outcome is that validators take a policy of non-interference and the social mitigation strategy fails.
3. If the community is unwilling to pay the measly $84k to fund this project it signals a total lack of credibility for the social strategy. The natural conclusion for someone considering deploying extractive strategies is that the community is incapable of decisive action and is too short-termist to actually stop them. dYdX research has published estimates of possible MEV on the order of basis points of volume. Assuming v3 volume, that would potentially mean 10s of millions in MEV every year. ~$170k to provide a credible deterrent to that seems like amazing value and if the community is unable to make even that tiny spend because of lack of trust or politics it completely undermines the social strategy at its outset.

Finally, there are some concerns about conflict of interest in the choice of committee members. I don’t find them compelling. Reverie is here asking us if they and others should get paid for services rendered, they’re not issuing themselves a grant in some undemocratic way. Reverie is in fact an expert on v4 MEV, they planned and audited the main grants work on MEV so far, and I’d consider them among the most knowledgeable community members on what MEV will look like on v4.

[FelixLts]

Do agree it is needed to further define some of the deliverables of the committee. E.g. specifying how the framework would be delivered and potentially a quarterly (or other frequency) report on the “State of dYdX MEV”. Great points!

[RealVovochka]

Hey @Ivo Thank you for your valuable thoughts as the member of the core team.

Also there can be some skepticism, so far this is only option available so I don’t see the reason not to use it

@Ivo

To present a convincing rationale for slashing, the community will have to do relatively extensive block-by-block analysis, including potentially trying to de-sybil multiple accounts and incorporating high-frequency price data from other venues

If the committee is doing such an activity, personally I don’t have any problems with the size of the compensation.

I agree that $84K is a negligible amount compared to the potential millions lost in MEV. The issue lies in the fact that the creation of a committee may only possibly solve the problem. The initial proposal, as noted by @tane, did not even entail any specific deliverables.

If a detailed description is provided regarding the daily activities that the committee will engage in, concerns regarding the budget will be alleviated.

@Ivo

Reverie is here asking us if they and others should get paid for services rendered, they’re not issuing themselves a grant in some undemocratic way. Reverie is in fact an expert on v4 MEV, they planned and audited the main grants work on MEV so far, and I’d consider them among the most knowledgeable community members on what MEV will look like on v4.

As for Reverie, they are a team of five individuals who are already involved in two major grant programs, in addition to being a VC fund. I doubt that they have the capacity to fully dedicate themselves to all the responsibilities they have taken on. Therefore, I am concerned about the grant program, which has long been criticized by the community. It is worth noting that the proposal for DGP extension was supported two months ago, and DGP is not a cheap endeavor for the community. Hence, Reverie may simply focus on coordinating this grant

[apriori]

Thank you for consideration for the committee.

Committee

Ideally, the community can self-organize and not rely on a committee for recommendations. Perhaps in the future this will be the outcome. In the interim, there is a need for a credible threat that keeps validators honest.

I recommend these two additional resources for more insights into this topic;

• Accountability along the transaction supply chain by @hxrts
• The Edge of MEV; The slow game & switching costs by Christopher Goes

Compensation

I cannot speak for other potential committee members, but this was not a motivating factor in agreeing to participate. If approved, I volunteer to forgo any compensation and re-route the proportional allocation to the community for funding MEV Research.

In general, compensation does serve a purpose.

Imagine validators collude to commit a socially slashable offense. They could, in fact, bribe select committee members to make a case not to slash the offending validators. Having committee members with skin in the game and compensation helps curtail this type of risk.

Reverie has done an outstanding job coordinating this proposal.

Deliverables

The committee should not be seen as labor force.

If governance is looking for more data dashboards or research deliverables, then perhaps there could be alternative ways to achieve this objective.

[magmar-skip]

Hey there - this is Mag from Skip.

I oversaw the development of the dydx.skip.money dashboard, and thought I’d comment on this proposal given it may be a factor in the slashing decisions made by the proposed committee.

The dashboard, as explained in detail in previous articles mentioned above, is an unopinionated display of orderbook discrepancy between block proposers (i.e. validators) on dYdX and a set of honest nodes that Skip runs. This work has been funded by the dYdX grants program, and the exact design was a collaboration between the Skip team and the dYdX research team.

The code will be open source, and any other party is willing to replicate the dashboard and host their own nodes if they deem our nodes to be untrustworthy. Skip does not run a validator on dYdX, or any other MEV-related or trading-related software on dYdX (and does not intend to either) in hopes there are never conflicts of interest in honestly building and hosting the dYdX MEV dashboard.

We believe that orderbook discrepancy is a strong proxy for MEV extraction on dYdX, as any MEV recapture will inherently involve some dYdX subaccounts profiting at the expense of others. Therefore, the dashboard is a tool to identify misbehavior, but it does not have the edge to act on that misbehavior in a way that deters validators from extracting MEV.

In order to have such an edge, we think that the proposed committee is a good idea to make the dashboard useful in preventing MEV. No Skip members sit (or will ever sit) on the proposed committee, due to obvious conflicts-of-interest. The selected participants are trusted members of the Cosmos and other ecosystems and have believed in the power of social consensus - a force that has kept Cosmos mostly MEV-free for almost a decade, even on massively high volume with ample extraction opportunities like Terra pre-implosion. We have always touted reputation-based infrastructure environments as a powerful MEV deterrent, and one of the greatest assets of any application, including dYdX, moving to Cosmos. Therefore, we think this will be an effective (and proven) system for preventing MEV on dYdX - and can do the job until a better mechanism is proposed.

However, we again do not plan to have any influence in the decisions by the committee proposed. Our duty to dYdX and its community is to honestly report data we see on-chain, and no more/no less. We are happy to receive feedback on the dashboard from the committee, or any other members of the community, if any new adjustments will help us better identify MEV extraction on dYdX V4.

I can also say that it’s been a pleasure to work with Reverie as we’ve built this - they’ve served as an effective liason between us and dYdX, and care deeply about preventing the effects of MEV on dYdX. As a highly trusted actor in crypto, I think they’d make a strong team to lead the social punishment strategy as proposed. It is pretty brave to serve on the committee mentioned above given the reputational risks involved, and I think no one ideally wants to see this system be used in practice for slashing. However major community harm should be reciprocated by equal threat to bad actors, and so I think it’s the right system given the damage MEV could cause on dYdX V4.

Thank you to the community and Reverie for funding us to build this dashboard, and here’s the best to a MEV-free and successful V4 upgrade

[RealVovochka]

hey, @apriori

Thank you for your honest position.

I would like to question the payment for the committee’s work as a protection against bribery from malicious actors if MEV is measured in hundreds of thousands of dollars. I believe that a compensation of 2k/month is not sufficient protection against such acts. The main defense lies in the reputation of the committee members.

On the other hand, I understand your position regarding labor force. In any experimental research, it is very difficult to determine deliverables. Therefore, I propose that the committee’s work be compensated retrospectively, and even in a larger volume, if the committee demonstrates its effectiveness. Additionally, I believe that grant program should cover any expenses associated with the committee’s work.

[RealVovochka]

Hello, @magmar-skip , thank you for the dashboard and welcome to the community. I agree that your grant is an effective investment of the community treasury. MEV is indeed a serious problem, and I also agree that your dashboard serves as a strong proxy for MEV extraction.

Regarding conflicts of interest, I apologize for being nitpicky. Is Reverie an investor in Skip or not? You are listed on their website under the Portfolio section.

[carlbergman]

Hi @RealVovochka,

Yes, as listed on our website, Reverie is a small investor in Skip ($10k check). The investment was completed in June of this year – after our first two grants but ahead of our final grant for dashboard maintenance. We disclosed our conflict with the Trustees ahead of their approval, and will continue to follow the conflict-of-interest policy in scenarios that apply.

[Imperator.co]

Hello, first of all Imperator.co is in favor of solutions to limit MEV and bad actors.

The solution, although rudimentary and based on a social component, remains, I think, the most effective possible in the short/medium term. Today, it would be complicated to obtain an infallible and automated solution, so we have to make decisions like this until we have something better.

Concerning the committee, it’s necessary to have a number that allows a certain resilience, but maybe 5 people would be enough. 7 seems to be a totally random number unless there’s another justification that hasn’t been enumerated. It might also be interesting to define the workload to be expected, which would justify the level of compensation requested.

Thank you for your work! We’re staying close to the discussion to see how it is progressing.

[RealVovochka]

Thanks for clarifying. Maybe you could create a separate topic where you describe all the potential conflicts of interest between you and grantees that the trustee knew about but the community didn’t. That way, we can avoid continuing the discussion in this thread.

[Callen_Wintermute]

We support this proposal and agree that having a committee proactively assessing malicious MEV actors is important for the health and UX of the dYdX Chain. Hopefully we can all learn a lot from this and potentially help shape future MEV mitigation strategies.

The committee itself has a great selection of talented and well-versed individuals with great context on both MEV and dYdX.

One thing we’d like to see as a deliverable for the community alongside the framework is a template for the actual slashing proposal itself and the process forward in a governance setting. We believe it’s super important for the community to be able to move quickly with a standardized governance format.

[Huzmond]

Hi all,

I’m Huzmond from Enigma, we are one of the genesis Validators on V4. We support this proposal and the MEV mitigation on the dYdX chain. We find the proposition by Reverie suitable in the short term due to the current lack of experience and the need for everyone to learn and discover better future solutions on this topic.

The committee must have a sufficient number of members and diversity to ensure its resilience. He is composed of talented individual from Cosmos and Ethereum ecosystem who are aligned with dYdX vision.

On the other hand, we also think the mitigation strategy should be inclusive. It may also have a process, which we’ll call the “external process” enabling anyone to identify, make a formal forum proposal, and receive rewards for identifying bad actors before the committee does. This approach promotes community engagement and the recognition of talented individuals.

Thanks for everyone who contribute to this topic and help to shape the future of MEV on dYdX and Cosmos

[CipherLabs]

@Huzmond

On the other hand, we also think the mitigation strategy should be inclusive. It may also have a process, which we’ll call the “external process” enabling anyone to identify, make a formal forum proposal, and receive rewards for identifying bad actors before the committee does. This approach promotes community engagement and the recognition of talented individuals.

Good idea, this is a great suggestion in my opinion.

[Govmos]

CONTEXT:
The absence of MEV mitigation mechanisms within dYdX poses a significant risk. With billions of dollars in trades occurring daily, validators could stand to gain substantially from engaging in order manipulation. Unfortunately, this comes at the cost of users who could suffer losses due to these actions.

However, there is a glimmer of hope. As an appchain, dYdX retains the power to take a more proactive out-of-protocol approach to mitigate MEV. Through on-chain governance, the dYdX community can influence and enforce behaviors, incentivizing positive actions and disincentivizing harmful ones.

ANALYSIS:
The situation at dYdX reveals a classic conundrum in the DeFi space. While MEV is a concern, it is challenging to address it solely through in-protocol solutions. This is because MEV operates within the bounds of protocol rules, even though it can have a detrimental impact on users. Validators may engage in these practices to maximize their profits, which further complicates the issue.

SOLUTION:
At Govmos (the governance branch of the PRO Delegators Validator), we adhere to the idea of creating a committe to assess these challenges on behalf of the community until we get a proper framework and sufficient data to be able to operate with self-defined mechanisms. To achieve this, we recommend the followings to be included in the committee’s framework during this mandate:

• Whistleblower Protection: Establish mechanisms for whistleblowers to report suspicious activities by validators anonymously. Ensure that individuals who raise concerns are protected from retaliation and encourage the reporting of unethical actions.

• Scenario Planning: Develop contingency plans for potential MEV scenarios, allowing for a rapid response in case of an emergency. These plans should outline the steps the committee recommends taking in different situations.

• Regular Reporting: Provide regular reports to the dYdX community detailing the state of MEV mitigation efforts, actions taken, and recommendations made. Transparent reporting can help build trust and accountability.

• Collaboration with Other DeFi Projects: Foster collaboration with other DeFi platforms facing similar challenges. Sharing best practices and coordinating efforts can create a more robust defense against MEV.

• Ethical Guidelines and “Code of Conduct”: Develop and enforce ethical guidelines for validators to adhere to. This code of conduct can specify what behaviors are considered acceptable or not and could be used as a basis for governance proposals.

• Incentive Structures: Evaluate and propose incentive structures that reward validators for constructive contributions to the dYdX ecosystem while penalizing malicious actions. Encouraging ethical behavior through financial incentives can be a powerful tool.

–

Thanks for reading ! If you like our job, you can support by delegating to PRO Delegators Validator

[apriori]

Thank you for consideration for the slashing committee.

Regrettably, I’d like to formally withdraw my candidacy from the slashing committee. The decisions announced by dydx leadership today are not aligned with my values. As such, I am no longer willing to associate my reputation with this brand.

Thank you for the opportunity. Thank you to the community and the Reverie team for their thoughtfulness in this entire process.

[Reverie]

Now that full trading is set to launch, we think the time is right to move ahead with this mitigation strategy and committee appointment. As volume starts to grow, so does the opportunity to profit from MEV. It’s important that the committee is in place to actively review data so as to catch early indications of malicious behavior.

New Member

We’re excited to add Rob Tiong from Rated Network as the seventh member of the committee. Rob joins us with a background in data analysis, as well as experience in assessing the performance of validators through his work at Rated.

Proposal

We are moving ahead with the on-chain proposal to launch a social strategy for mitigating MEV on dYdX v4. The proposal will signal community alignment with the strategy and an intent to take action against bad actors. The goal is to make it clear that the community agrees on the need to prevent MEV.

Here is the DIP outlining the proposal: dYdX v4 social mitigation for MEV proposal · GitHub

If the proposal passes, the Grants program will process the grant and kick off funding for committee members.

[tane]

By the end of term, the committee should bring forward a framework for standardizing certain actions based on discrepancy data.

Great to see a proposal of a framework based on the work that will be done by the committee is a part of the committee’s responsibilities. Looking forward to seeing the reports and actual recommendations if any.

[RealVovochka]

Since the committee approved when Skip dashboard will be accurate?

The trading volume is already significant

[Daniel]

Yes, necessáry strategy of migration and social distribuicion, after descentralized in the chainliks