Spatial Arbitrage (BQAE)

Operator & jurisdiction: BASIS is operated by BASIS DIGITAL INFRASTRUCTURE LTD, a Seychelles-incorporated entity (LEI: 254900IX2F2KCWNSSS64).

Currency convention: All portfolio values, performance displays, and internal calculations may be shown in USDT as a USD-equivalent accounting unit. USDT is used for internal display and accounting only. Deposits and withdrawals are supported in native assets only: BTC, ETH, SOL, and PAXG.

1) Objective

Spatial Arbitrage, also known as cross-venue arbitrage, is the act of simultaneously buying an asset on one venue where the price is lower and selling it on another where the price is higher. The BASIS Quantitative Arbitrage Engine (BQAE) is designed to execute this strategy systematically with deterministic execution, execution precision, and state-machine risk controls.

2) Plain-English Overview: Global Price Gaps

Imagine a product selling for $100 in one market and $105 in another. A spatial arbitrage system buys in the lower-priced market and sells in the higher-priced market, capturing the spread after transaction costs.

In digital asset markets, the "markets" are exchanges and liquidity venues, and the assets are instruments such as BTC, ETH, SOL, or PAXG. The BQAE continuously scans cross-venue pricing to detect temporary dislocations and route orders with high execution precision.

BQAE is designed to capture structural alpha from short-lived cross-venue inefficiencies, not directional market exposure.

3) Professional View: Frictions That Define Profitability

Although the idea is straightforward, profitable implementation depends on infrastructure, quantitative modeling, and risk constraints. The BQAE must solve several core frictions.

Latency and routing

Price dislocations can collapse within milliseconds. The system must coordinate both sides of the trade before the spread disappears.

BASIS infrastructure emphasizes:

• Sub-50μs internal latency design

• 100K+ OPS processing capacity

• Proprietary routing infrastructure

• Deterministic execution paths

• Venue-aware order placement logic

This architecture supports execution precision under fragmented market conditions.

Execution quality and slippage control

Both legs must complete at acceptable prices. If one side fills and the other does not, the trade may become unintended directional exposure.

To reduce this risk, BQAE uses:

• Pre-trade slippage modeling

• Liquidity-depth checks

• Conservative fill thresholds

• Atomic execution logic where possible

• Immediate unwind procedures when completion constraints are violated

Settlement and inventory management

Cross-venue arbitrage requires inventory to already exist where it is needed. For example, one venue may require quote inventory while another requires base inventory. Rebalancing this inventory across venues introduces cost, delay, and operational complexity.

The BQAE maintains venue-level inventory models and treasury constraints to support continuous execution without relying on slow reactive transfers.

4) Mathematical Framing: Graph Search for Structural Alpha

Finding profitable multi-venue and multi-asset paths can be modeled as a graph problem. Research Partner frameworks apply concepts related to Bellman-Ford and Floyd-Warshall to represent the trading environment as a directed graph.

Graph representation

Component	Interpretation
Nodes	Assets on specific venues, such as BTC on one exchange or ETH on another
Edges	Tradable conversions or transfer relationships
Edge weights	Exchange rates adjusted for fees, spread, and execution assumptions

A common transformation uses the negative logarithm of effective exchange rates. Under that formulation, profitable cycles correspond to negative-weight cycles in the graph.

Node A -> Node B -> Node C -> Node A

If the product of effective rates across a closed loop exceeds 1 after all costs, the cycle may contain structural alpha.

$$\log(r_1) + \log(r_2) + \log(r_3) + \dots + \log(r_n) > 0$$

Where each ( r_i ) is the effective exchange rate after fees and modeled execution cost.

This approach allows BQAE to evaluate not only simple two-venue spreads but also more complex path-dependent opportunities across fragmented liquidity.

A theoretical cycle is not sufficient for execution. BQAE applies real-time feasibility filters for liquidity, latency, venue health, and inventory availability before any route is eligible.

5) Risk Model and Control Framework

The strategy is governed by deterministic constraints rather than discretionary intervention. Risk controls are implemented as state-machine checks before, during, and after execution.

Risk	Description	BASIS control mechanism
Execution risk	One leg fills while the offsetting leg does not, creating unwanted exposure	Hedge completion window with automatic unwind logic if paired execution constraints fail
Settlement risk	Capital becomes unavailable on a venue due to transfer restrictions or operational issues	Liquidity fragmentation, venue scoring, balance caps, and automatic venue isolation
Slippage inversion	Spread is smaller than total cost after fees and market impact	Expected value gate using conservative slippage and fill assumptions
Latency decay	Opportunity disappears before both legs can complete	Time-validity thresholds and routing only through venues meeting latency standards
Inventory imbalance	Required assets are not positioned on the correct venues	Pre-positioned inventory framework with treasury rebalancing constraints

6) Why Infrastructure Matters

BQAE performance depends on engineering quality as much as on pricing logic. The system is built around:

• Deterministic execution

• Math-constrained opportunity selection

• State-machine risk controls

• Venue health monitoring

• Proprietary routing infrastructure

• Research-driven path optimization frameworks from Base58 Labs

This is essential because the edge in spatial arbitrage is often small, temporary, and highly sensitive to operational precision.

References

[1] Cormen, T. H., Leiserson, C. E., Rivest, R. L., & Stein, C. (2009). Introduction to Algorithms, 3rd Edition. MIT Press. Chapter 24.1: The Bellman-Ford algorithm.