Can we exploit differentials in cross-rates for profit? Math says yes. The question is infrastructure.
Status: active — DeFi/Solana path under investigation
Created: 2026-02-12
Vault: C:\Users\Chris.Isaacson\Vault\projects\triangular-arbitrage\README.md
Exploit differentials in exchange rates across currencies or tokens. If the product of rates around a loop ≠ 1.0, there's a profit opportunity. Multiple approaches from full triangular loops to single-leg trades to DeFi flash loans.
Using live ECB daily rates (2026-02-12) across 8 currencies (USD, EUR, GBP, MXN, JPY, CHF, CAD, AUD):
- 264 out of 336 triangular paths show positive profit
- Best full-loop path: USD → AUD → JPY → USD yields $7.13 on $10,000 (0.071%)
- Bellman-Ford found 224 negative cycle indicators
- JPY is consistently the mispriced currency — slower price discovery (confirmed by academic research)
Instead of executing all 3 legs, identify which single pair is mispriced vs implied cross-rates:
- Best single-leg: JPY/GBP at 0.087% deviation — $86.79 profit on $100K
- 2 pairs beat retail spreads (JPY/GBP, JPY/USD)
- 33 pairs beat institutional spreads
- 1/3 the transaction cost of full triangular execution
Conclusion: Differentials are real and detectable. The math works.
We initially concluded "not viable" based on assumptions about infrastructure costs. That was the accountant talking, not the CEO. So we actually researched the numbers.
| Scenario | Upfront | Monthly Cost | Monthly Profit | Capital Needed | Break Even |
|---|---|---|---|---|---|
| Forex Institutional | $135,000 | $1,800 | $1,500 | $100,000 | 90 months |
| Crypto CEX (Cross-Exchange) | $0 | $150 | $300 | $10,000 | Immediate |
| DeFi Solana (Conservative) | $600 | $170 | $430 | $0 | 1.4 months |
| DeFi Solana (Optimistic Niche) | $600 | $170 | $2,830 | $0 | 6 days |
- $135K upfront (Cayman Islands license + FIX API deposit)
- $100K working capital on top of that
- 7.5 years to break even
- Ongoing ROI is decent (83%) but barrier to entry is a wall
- Latency disadvantage vs. HFT firms with dedicated hardware
- CEO Decision: No. The infrastructure bet doesn't pencil out at our scale.
- Zero upfront, $150/month operating
- $10K capital split across exchanges
- $300/month profit — steady but small
- Key risk: Transfer times between exchanges (minutes to hours). Price can move while your assets are in transit.
- CEO Decision: Maybe. Low risk, low reward. Could be a learning exercise.
- $600 upfront, $170/month — radically low barrier to entry
- No trading capital needed — flash loans borrow and repay in a single atomic transaction
- Failed trades cost almost nothing — unprofitable tx simply reverts, you lose only gas (~$0.00025)
- Break even in 1.4 months (conservative) to 6 days (optimistic)
- 240,000+ successful DeFi arbs identified in 1 year ($868M volume across chains)
- Asymmetric risk profile: capped downside (gas on failed txs), scalable upside
Key risks:
- Liquidity risk — the same thin pools that create wide spreads (bigger arb opportunities) are the ones where your own trade moves the price. Flash loans amplify this: borrowing large amounts to capture small differentials means slippage can eat your profit. Thick pools are safe but have tiny spreads; thin pools have big spreads but can't absorb your trade. The sweet spot — liquid enough to execute, inefficient enough to profit — is the real thing you're searching for.
- 86% of crypto trading already automated — highly competitive
- Institutional MEV bots operate in milliseconds
- Smart contract bugs = potential total loss
- Need MEV protection (Jito on Solana) to avoid sandwich attacks
Key advantages:
- Flash loans = zero capital requirement
- Solana gas = near-zero cost per attempt
- Open-source bot templates exist (ExtropyIO, etc.)
- New token launches/pools constantly create fresh niche opportunities
- Failed attempts are essentially free — you can try thousands of times
CEO Decision: Worth investigating further. The risk/reward asymmetry is favorable. Low downside, meaningful upside, and the competitive moat question ("can we find arbs others aren't harvesting?") is an entrepreneurial bet worth making at $600.
- Research Solana DEX ecosystem — Orca, Raydium, Jupiter. Map token pairs, pool sizes, existing bot activity
- Study flash loan mechanics — how they work on Solana, Jito integration, revert behavior
- Build detection prototype — adapt our matrix math to scan DEX pool prices instead of forex rates
- Estimate real opportunity frequency — how many arbs/day exist on Solana DEXes, what's the average profit
- Smart contract development — build or adapt from open-source templates, security audit
- Paper trading — run detection without execution to measure what we would have caught
- Deploy with minimal capital — start small, measure, iterate
This project validated every layer of our theoretical framework:
| Theory | What Happened |
|---|---|
| Utility/Trade | Differentials are real — the math proves value gaps exist |
| Risk | Forex: capability risk killed it. DeFi: asymmetric risk profile favors the bet |
| COO vs CEO | Detection = COO (math, solved). Strategy = CEO (should we invest? investigate first) |
| Cognitive vs Motor | Forex bottleneck is execution speed (motor). DeFi bottleneck is smart contract quality (cognitive) |
| Grounding | Real money, real markets — can't be talked out of a loss |
| Honest Assessment | We initially skipped the CEO step and assumed "too expensive." Actual research changed the picture. |
- detect_arbitrage.py — Full triangular arbitrage detector with Bellman-Ford
- single_leg_analysis.py — Single-leg implied cross-rate deviation analysis
- roi_analysis.py — ROI comparison across all four scenarios
economics, risk, praxis, agents, crypto, defi