DIKE PROTOCOL · WHITEPAPER v2.1

A Framework for Capital-Efficient Chained Prediction Markets

Multiverse-inspired leverage that recycles collateral down conditional universes, unlocking compounded exposures without fresh capital.

Version · 2.1.0

Runtime · ~7 min read

Updated · Nov 2025

Abstract

Abstract. Protocol Synopsis

DIKE operationalizes Multiverse Finance into a programmable loan primitive that reuses collateral across conditional predictions.

Traditional prediction markets strand capital inside each position. DIKE introduces Prediction Chaining: the collateral from a resolved or mark-to-market winning verse backs a protocol loan that fuels the next bet. Because each child inherits risk from its parent, a single unit of staked value can support multiple sequential exposures.

This paper formalizes the mechanism design, detailing the recursive loan schedule, settlement paths, liquidation thresholds, and worked economics of a three-link chain. The result is a composable primitive for building structured prediction products with transparent risk envelopes.

Quick Stats

0.6x
Reference collateral ratio
HR ≥ 1
Liquidation guardrail
2.91x
Return in worked example
4.98x
Max return in simulations
3 links
Chain depth in case study

Section 1

1. Introduction

Prediction markets aggregate belief but trap capital per position. DIKE reuses that collateral through under-collateralized protocol loans, preserving solvency with deterministic controls.

Capital inefficiency is the dominant drag on prediction market adoption. A user chasing multiple themes must fund each venue separately, idling their bankroll until results finalize. DIKE addresses this deadweight loss by chaining positions. The moment a stake acquires positive equity, a user may borrow against it to seed the next universe, effectively cascading exposure down a deterministic decision tree.

The architecture mirrors Paradigm's Multiverse Finance proposal but adds production-ready logic—tracking loan principal, accruing interest, stress-testing HR, and enforcing DAG constraints so that no child exists without a solvent parent.

Section 2

2. Core Concepts & Terminology

The vocabulary below grounds the rest of the mechanism. Every definition maps to on-chain state the protocol tracks.

Prediction (Verse)

A binary market on a specific event within Paradigm's Multiverse Finance framework. Each verse is conditionally dependent on its parent outcome.

Stake (S)

Capital a participant locks into a prediction. Wins return

S_i

plus profit; losses burn the stake.

Prediction Chain

An ordered sequence

P_1, P_2, \ldots, P_n

where each child position is funded by a protocol loan collateralized by its parent stake, forming a DAG.

Loan (L)

Capital extended against collateral. For verse

P_i

, the protocol issues

L_i = r \cdot S_i

which becomes

S_{i+1}

Collateralization Ratio (r)

Protocol parameter

0 < r < 1

that caps leverage. Example:

r = 0.6

unlocks 60% of posted collateral as a loan.

Health Ratio (HR)

Safety metric

HR = \frac{V_1}{\sum L_i^{\\text{gross}}}

comparing the MTM value of the root collateral to total debt.

Section 3

3. Protocol Mechanism

A chain is governed by deterministic formulas for loan sizing, recursive stake propagation, settlement, and liquidation.

3.1 Chain Initiation

A user deposits $S_1$ into the root prediction $P_1$ . This stake represents the only exogenous capital in the chain and becomes the collateral base for all downstream loans:

S_1 = \text{Initial User Capital}

3.2 Chain Propagation (Leveraging)

Once $P_1$ is live, DIKE extends a loan $L_1$ equal to a fraction $r$ of the staked value. The loan becomes the stake for the child prediction $P_2$ . Recursively:

L_i = r \cdot S_i

S_{i+1} = L_i = r \cdot S_i

S_i = S_1 \cdot r^{\,i-1}

\sum_{i=1}^{n} S_i = S_1 \cdot \frac{1 - r^{n}}{1 - r}

The entire stack therefore compounds exposure while remaining deterministically backed by the root capital.

3.3 Position Resolution & Settlement

Settlement happens verse-by-verse. Define the gross loan as principal plus accrued interest:

L_i^{\text{gross}} = L_i \cdot e^{\rho \Delta t}

Case A · Child Wins

Winnings repay

L_i^{\\text{gross}}

immediately. Residual profit becomes withdrawable equity or fresh collateral for future chains.

Case B · Child Loss

Stake

S_{i+1}

is forfeited. The protocol seizes

L_i^{\\text{gross}}

from parent collateral, reducing user equity in

P_i

Case C · Full Chain Wins

Every prediction resolves favorably; the user repays all outstanding loans and withdraws

S_1 + \sum \\text{profits}

3.4 Liquidation

DIKE monitors the mark-to-market value of the root position $V_1$ relative to the aggregate debt. When the Health Ratio breaches the liquidation threshold, the protocol seizes collateral, repays loans, and collapses the chain.

L_{\text{total}}^{\text{gross}} = \sum_{i=1}^{n-1} L_i^{\text{gross}}

HR = \frac{V_1}{L_{\text{total}}^{\text{gross}}}

Example guardrail: liquidate if $HR < 1$ , ensuring the protocol never carries under-collateralized debt.

Section 4

4. Worked Example

A three-link chain demonstrates leverage recycling, payoff asymmetry, and liquidation sensitivity.

Parameters

Initial stake S₁ = $100
Collateralization ratio r = 0.6
Loan interest ρ = 5% per hop
2x payout for winning predictions

Propagation

Loan L₁ = 60 → Stake S₂ = $60
Loan L₂ = 36 → Stake S₃ = $36
Total deployed capital = $196 backed by $100

Debt Stack

Loan principal = $96
Gross obligation = 60·1.05 + 36·1.05 = $100.8
HR trigger if V₁ < $100.8

All Win Scenario

Payouts: $200 + $120 + $72 = $392
Net after debt = $291.2
Effective return ≈ 2.91x vs 2x siloed

Liquidation Scenario

If V₁ = $95, HR = 0.94
Protocol seizes S₁, repays $95 toward loans
Chain unwinds, children forfeited

Key Takeaway

Recycling collateral via $r = 0.6$ transforms a $100 stake into $196 of total exposure without additional user capital, yielding a 2.91x payoff when all predictions win while retaining deterministic liquidation if the root MTM deteriorates.

Section 5

5. Conclusion

Prediction Chaining turns Multiverse theory into live financial plumbing for capital-efficient speculation.

DIKE delivers a solvency-aware leverage rail for prediction markets. Transparent ratios, deterministic settlement semantics, and chain-level liquidation logic keep the protocol neutral while letting users amplify directional views with a single deposit.

Upcoming iterations extend the primitive with oracle-driven MTM updates, configurable interest curves, and composable vault products layered on top of Prediction Chains.

Section 6

6. References

Primary research underpinning DIKE.

White, Dave. “Multiverse Finance.” Paradigm, 2022.