Infinite Games need Infinite Markets

🔬🔮 Research
,
1

SuperXGA

Working Draft Proposal 01/23/2025

XGA V2 Builds on the success from V1 and brings a permissionless platform that introduces a Limit Order Book ontop of the MEV Call Market Auction. Additionally by supporting LST’s as the depositing token for rollup access, we can create a flywheel for usage of the CLOB for trading.

  • Lending/Bond Protocol that lets users deposit LST’s for accessing the Rollup, which in turn can be used for margin/leverage. Interest paid on these deposits is also credited as extra. gas on the rollup.
  • There 2 Auctions (Flashbots / XGA) and a CLOB (where users can permissionlessly trade Continuous Futures)
  • You will be able to short (guess who takes that end of the trade).
  • A portion of all fees accrue to Captive Insurance Pool for Manifold Finance as FOLD is the penultimate backstop collateral. XGA token is not as risk share token.
  • Fees are generated from:
    • Protocol Fee as a Percentage of Staked ETH Yield (mevETH)
    • L2 Fee as a Fixed Rate per Transaction (Gas-Like Model, XGA)
    • Lending/Bond Market rates (XGA)
    • Margin Funding Rates (XGA)
    • CLOB trading fees (XGA)

XGA Tokenomics

Quantitative Rewarding, QR mechanism is a two-token model that provides significant advantages for an Ethereum Layer 2 Rollup. By decoupling transaction and staking tokens, the system can achieve greater stability, avoid reward explosions, and maintain a balanced and sustainable tokenomics design. Each step is grounded in the theoretical framework provided, ensuring a comprehensive and feasible implementation plan.

  • Dual Token model
  • Custom Gas Token = LST tokens (e.g. mevETH) == quantitative ETH, i.e. undercollateralized ETH)
  • Reward Token = XGA

Axioms:

  • XGA is a governance token insofar that it controls the treasury, insuring a floor price based off of the treasury holdings
  • Positive token pressure less token selling/vesting

PLF/PBI Gas Token

Protocol for Loanable Funds Gas Token or Protocol for Bond Issuance

Instead of depositing to the Rollup Portal (re: OP Portal) you instead “lend” through the lending protocol your LST. The lending protocol only has one borrower: the rollup itself. You then get credited on the rollup “Mana”. Additional benefits are that you are given up to 7 days worth of additional rewards that would have accrued to the LST on the rollup, meaning you could transact on the rollup with your future LST rewards and still get back 100% of your initial principle. The aggregate amount of lent LST tokens is the margin capacity available to users on the rollup to use for trading in both markets (the call market, and the ‘perps/futures’ market).

This is basically a bond issuance because of the exclusive borrower, however there is no discrete auction taking place.

Cross Margining and Symbiotic Re-staking

Centralized exchanges use USD as the numeraire when calculating margin balances, thus require liquidations of the spot asset to meet margin calls. This liquidation, however, removes the hedge asset used in the cash and carry strategy, and leads to premature liquidation of the whole position. Overall, the absence of cross-margining results in inefficiency in this trading strategy and more generally using margin.

We can additionally use Symbiotic for those wanting to re-stake LST’s to provide insurance against the margin system. This makes other LSTs compatible with earning yield and by virtue increasing the amount of margin available to the system outside of just mevETH deposits.

Tie surcharge liquidation to rate of settlement so that. avg rate of settlement is used to provide baseline for excess rate of settlement difference which can be used to increase interest rate risk calculation

Convenience yield

  1. Stake via mevETH
  2. Re-Stake via Symbiotic
  3. 1+2 = Convenience Yield

Continuous Forwards

We can use the cross margining as a funding rate mechanism for the underlying call forwards as the price is a constant multiple of the spot price. This multiple is determined by the ratio of the margin intensity (κ) and the difference of the net cost of carrying the underlying asset (r-r').

State-contingent token-supply policy

Parameters and Constraints.

  1. The risk free rate r and thus the discount factor \delta=1 /(1+r).

Model Description and Notation:

  • Number of users (m) and stakers (n) remain constant over time.
  • Token holdings (Th), token price (Tp), and strategies for buying (b_i) and selling (s_j) are defined.
  • Service levels (S(t)) and rewards (R(t)) are allocated based on user payments and validator staking.

  1. Single-token Symmetric Equilibrium:

    • Users spend tokens for service, and stakers stake tokens for rewards.
    • Adjust rewards (R(t+1)) based on service level (S(t+1)) to avoid uncontrolled growth.
    • Maintain a balance between demand (m * b_i) and supply (n * s_j).

  2. Two-token Model Symmetric Equilibrium:

    • Token A (staking) and Token B (transactions) are used separately.
    • Ensure stable Token B prices through constant service-to-fees (Ser2Fees) and rewards-to-stake (Rew2Stake) ratios.
    • Set rewards (R(t+1)_B) based on next round’s service level (S(t+1)) to avoid buybacks.

  3. Quantitative Rewarding Mechanism:

    • Increase rewards when service level decreases to adjust supply.
    • Decrease rewards when service level increases to restore payments.
    • Ensure rewards do not grow exponentially by setting initial rewards (R(0)) appropriately.

Buyback/Burn / No Buyback Equilibrium Conditions

Increasing the rewards at the next round helps for the no buy back condition. The reason is that this will make rewards more attractive for the stakers who then stake more tokens (and thus sell less on the market).

Note this does not mean that there never is a burn/buy back action taken by the system, merely that the system targets to not incur such a state.

Service Level

A proxy for the service level S(t) is can be assumed by using the fees as a proxy for deriving S(t)

Burn/No Burn System parameters

This depends on two parameters:

  1. Total amount of service S^{(t)} which is divided among the users at each step t.
  2. Total amount of rewards R^{(t)} which is distributed to the (re)stakers at each step t.

Ecosystem Parameters

Non-exhaustive

Lending/Bond Market Fees

L2 Fee as a Fixed Rate per Transaction (Gas-Like Model)

  1. Service Parameter

    • Name: Per-Transaction Fee
    • Unit: Gwei or USD per transaction
    • Definition: Each transaction on the L2 costs a flat fee.

  2. Growth Parameter

    • Name: Total Number of Transactions
    • Unit: Transactions per day/week/month
    • Definition: The total count of L2 transactions.

  3. Adoption Parameter

    • Name: Average Transactions per User
    • Unit: Tx/user
    • Definition: The typical number of transactions each user makes in a period

Protocol Fee as a Percentage of Staked ETH Yield

  1. Service Parameter

    • Name: Protocol Fee on Yield
    • Unit: “% of staking rewards”
    • Definition: The liquid staking protocol takes a cut of the yield generated by mevETH.

  2. Growth Parameter

    • Name: Total ETH Staked
    • Unit: ETH (or USD equivalent)
    • Definition: The total amount of ETH deposited into the protocol.

  3. Adoption Parameter

    • Name: Average ETH Staked per User
    • Unit: ETH/user
    • Definition: The typical deposit size for each user. The total staked = (number of users) × (average staked per user).

ETH Re-Staking Yield via AVS Boost

Fairness considered possible

How can AVS Boost (and by extension mevETH) provide for higher yields? By providing a conduit for searchers and exclusive orderflow to bypass the monopolistic behaviour of builders. It is the result of the auction strategy difference in builders’ competition that exacerbates revenue loss for proposers: as much as 20% can be attributed to this fact by itself.[^1]

…Furthermore, owing to auction strategy difference, the top 3 builders submit bids 26.87% lower than the other builders, while their total winning rate exceeds 95%. Our research reveals that, in the reality of information difference, the premise that builders bid all profits to proposers does not hold true.[^2]

The Gang learns about Information Asymmetry

PBS cannot achieve robust fairness. For builders with information and auction asymmetry, the emergence of a monopoly state is inevitable. The builder market always tends to a monopoly in the current PBS regime.

  • Can we reduce the information asymmetry about the blockchain state?
    • Yes, with cryptography (ordering decisions made without information about tx contents)
    • Yes, with partial block building and using the relay we can bypass builders and disintermediate searcher access
  • Can we reduce the information asymmetry about the world state?
    • No. Only by reducing latency from event (1) to event (2).
    • We can also craft algorithms to combine transactions within a quantized time period (batching, bundling at the relay level)
  • Light Clients can help, but require Verkle Trie (i.e. wallets would not need RPC backing infrastructure anymore to validate account state thereby letting them propagate transactions)