Build a Saber Vote Market
Overview
It looks like things are coming full circle. Here, I propose that we build a vote market as we proposed in proposal 0, only for Saber instead of Marinade. I'd recommend you read that proposal for the context, but I'll summarize briefly here:
- I proposed to build a Marinade vote market
- That proposal passed
- We learned that Marinade was developing an internal solution, we pivoted to supporting them
All of that is still in motion. But recently, I connected with c2yptic from Saber, who happens to be really excited about the Meta-DAO's vision. Saber was planning on creating a vote market, but he proposed that the Meta-DAO build it instead. I think that this would be a tremendous opportunity for both parties, which is why I'm proposing this.
Here's the high-level:
- The platform would be funded with $150,000 by various ecosystem teams that would benefit from the platform's existence including UXD, BlazeStake, LP Finance, and Saber.
- veSBR holders would use the market to earn extra yield
- Projects that want liquidity could easily pay for it, saving time and money relative to a bespoke campaign
- The Meta-DAO would own the majority of the platform, with the remaining distributed to the ecosystem teams mentioned above and to users via liquidity mining.
Why a Saber Vote Market would be good for users and teams
Users
Users would be able to earn extra yield on their SBR (or their veSBR, to be precise).
Teams
Teams want liquidity in their tokens. Liquidity is both useful day-to-day - by giving users lower spreads - as well as a backstop against depeg events.
This market would allow teams to more easily and cheaply pay for liquidity. Rather than a bespoke campaign, they would in effect just be placing limit orders in a central market.
Why a Saber Vote Market would be good for the Meta-DAO
Financial projections
The Meta-DAO is governed by futarchy - an algorithm that optimizes for token-holder value. So it's worth looking at how much value this proposal could drive.
Today, Saber has a TVL of $20M. Since votes are only useful insofar as they direct that TVL, trading volume through a vote market should be proportional to it.
We estimate that there will be approximately $1 in yearly vote trade volume for every $50 of Saber TVL. We estimate this using Curve and Aura:
- Today, Curve has a TVL of $2B. This round of gauge votes - which happen every two weeks - had $1.25M in tokens exchanged for votes. This equates to a run rate of $30M, or $1 of vote trade volume for every $67 in TVL.
- Before the Luna depeg, Curve had $20B in TVL and vote trade volume was averaging between $15M and $20M, equivalent to $1 in yearly vote trade volume for every $48 in TVL.
- In May, Aura has $600M in TVL and $900k in vote trade volume, equivalent to $1 in yearly vote trade volume for every $56 of TVL
The other factor in the model will be our take rate. Based on Convex's 7-10% take rate, Votium's ~3% take rate, and Hidden Hand's ~10% take rate, I believe something between 5 and 15% is reasonable. Since we don't expect as much volume as those platforms but we still need to pay people, maybe we start at 15% but could shift down as scale economies kick in.
Here's a model I put together to help analyze some potential scenarios:
The 65% owned by the Meta-DAO would be the case if we distributed an additional 10% of the supply in liquidity incentives / airdrop.
Legitimacy
As I've talked about, assuming futarchy works, the most important thing to the Meta-DAO's success will be acquiring legitimacy. Legitimacy is what leads people to invest their time + money into the Meta-DAO, which we can invest to generate financially-valuable outputs, which then generates more legitimacy.
By partnering with well-known and reputable projects, we increase the Meta-DAO's legitimacy.
How we're going to execute
Who
So far, the following people have committed to working on this project:
- Marie to build the UI/UX
- Matt / fzzyyti to build the smart contracts
- Durden to design the platform & tokenomics
- Joe and r0bre to audit the smart contracts
- me to be the accountable party / program manager
UXD has also committed to review the contracts.
Timeline
December 11th - December 15th
Kickoff, initial discussions around platform design & tokenomics
December 18th - December 22nd
Lower-level platform design, Matt starts on programs, Marie starts on UI design
December 25th - January 5th (2 weeks)
Holiday break
January 8th - January 12th
Continued work on programs, start on UI code
January 15th - January 19th
Continued work on programs & UI
Deliverables on Friday, January 19th:
- Basic version of program deployed to devnet. You should be able to create pools and claim vote rewards. Fine if you can't claim $BRB tokens yet. Fine if tests aren't done, or some features aren't added yet.
- Basic version of UI. It's okay if it's a Potemkin village and doesn't actually interact with the chain, but you should be able to create pools (as a vote buyer) and pick a pool to sell my vote to.
January 22nd - 26th
Continue work on programs & UI, Matt helps marie integrate devnet program into UI
Deliverables on Friday, January 26th:
- MVP of program
- UI works with the program delivered on January 19th
January 29th - Feburary 2nd
Audit time! Joe and r0bre audit the program this week
UI is updated to work for the MVP, where applicable changes are
February 5th - Febuary 9th
Any updates to the program in accordance with the audit findings
UI done
February 12th - February 16th
GTM readiness week!
Proph3t or Durden adds docs, teams make any final decisions, we collectively write copy to announce the platform
February 19th
Launch day!!! 🎉
Budget
Based on their rates, I'm budgeting the following for each person:
- $24,000 to Matt for the smart contracts
- $12,000 to Marie for the UI
- $7,000 to Durden for the platform design
- $7,000 to Proph3t for program management
- $5,000 to r0bre to audit the program
- $5,000 to joe to audit the program
- $1,000 deployment costs
- $1,000 miscellaneous
That's a total of $62k. As mentioned, the consortium has pledged $150k to make this happen. The remaining $90k would be custodied by the Meta-DAO's treasury, partially to fund the management / operation / maintenance of the platform.
Terminology
For those who are more familiar with bribe terminology, which I prefer not to use:
- briber = vote buyer
- bribee = vote seller
- bribe platform = vote market / vote market platform
- bribes = vote payments / vote trade volume
References
Status
$0.0
$0.0
Volume
Pass Threshold
Liquidity Pool Fee
Add a thought or a comment...