Why borrow instead of sell
The same reasons come up across mandates:- The exposure is meant to be kept: a sale ends it, while a borrow keeps it on the book and releases cash against it.
- A sale has consequences of its own: realizing a position can conflict with a mandate, a reporting period, or a target allocation. Whether those costs exceed the cost of borrowing is your own analysis; borrowing is what makes the comparison available.
- Speed and reversibility: a draw against posted collateral is one transaction, and repaying it restores the starting point. Unwinding and rebuilding a position is neither quick nor free.
The framework every strategy shares
Whatever the borrow earns or funds, the debt side behaves the same way. Before entering any of these:- Set a health-factor floor and treat it as policy. The bands give the vocabulary: open in Healthy, review at Stable, act in Monitor. Decide the floor before the position exists.
- Assume both rates move. Borrow costs float with utilization, and so does anything the proceeds earn; Net APY shows how quickly a spread can invert.
- Know the liquidation math for your collateral class. Ceilings, buffers, and penalties differ by class; the table is in Risk parameters, and the cost of getting it wrong is in How liquidation works.
- Plan the exit before the entry: which asset tops the position up, which wallet holds it, and in what order the position unwinds. The monitoring routine keeps that plan current.
- Size against a bad day rather than the current price. The portfolio health score stress-tests the book against a correlated crash; a strategy that only survives current prices is undersized.
The four strategies
Each page follows the same structure: what it is, mechanics on AGI3, when it fits, what it costs, how it fails, and how it is managed.
