FX Protocol 101 1️⃣0️⃣1️⃣ What if liquidation wasn't the default outcome? If you've used leverage in DeFi long enough, you've probably noticed a pattern that feels almost inevitable. Price moves against you, sometimes slowly, sometimes all at once. Your LTV drifts upward, getting closer to a number you already know is there. And once it crosses that line, the system doesn't negotiate. It doesn't adjust your exposure or try to stabilize your position. It simply closes it. From the protocol's perspective, this behavior is perfectly rational. A fixed threshold is - Easy to enforce - Easy to reason about, and guarantees that the system remains solvent. ℹ️ But the simplicity of that design comes with a cost that most users eventually feel. There is no middle ground. - No gradual reduction of risk - No attempt to absorb volatility - No transition between a position That is healthy and one that is about to disappear. The system waits, and then it reacts "abruptly". Which effectively turns liquidation into the first response to volatility, not the last. What F(X) changes is not the existence of liquidation, but the moment at which the system decides to use it. Instead of relying on a single terminal event, F(X) introduces an intermediate step that most protocols simply don't have: REBALANCING REBALANCING REBALANCING As the position approaches riskier territory, around ~88% LTV, the system intervenes. But instead of closing the position, it begins to reshape it. A portion of the outstanding fxUSD debt is burned, collateral is partially reallocated, and the position's overall leverage is reduced. The important detail here is not the mechanics themselves, but the outcome they create. THE POSITION SURVIVES! Exposure is preserved, even under stress, because the system is able to contract internally rather than forcing an external resolution. This only works because of how F(X) is structured at a deeper level. At its core, the system is governed by a balance constraint: n·s = nf·f + X (Our repetitively-mentioned method) Which means that at any point in time, the total value of the collateral must equal the combined value of stable liabilities (fxUSD) and leveraged exposure (xPOSITION). This relationship isn't a guideline; it's enforced continuously. So when market conditions change and risk increases, the system doesn't need to look outward for liquidity or rely on funding dynamics to restore balance. It adjusts itself. Rebalancing is simply the protocol that brings its own internal state back into line with that constraint. And this is precisely why most other protocols don't attempt something similar. It's not that liquidation is misunderstood; it's that alternative approaches require a fundamentally different architecture. To support rebalancing, you need tight coupling between debt and collateral, deterministic accounting at the system level, and a reliable way to execute partial adjustments without breaking the system's integrity. On top of that, you need an incentive layer that ensures these adjustments actually happen in real time. Without all of this, liquidation remains the only practical option. Of course, this design is not without trade-offs. Rebalancing introduces complexity. It depends on active participation from the system, and during periods of stress, users may still lose value as part of the adjustment process. It's not a free protection layer. But it does replace a binary outcome with a continuous one. Instead of moving directly from "safe" to "gone", the system can adapt along the way. And that seemingly small shift, delaying liquidation, inserting a layer of adjustment before it, changes more than just the mechanics. It changes how risk is experienced. And over time, it changes how users behave inside the system.