Collateralization Ratio

There is no single right answer, but most experienced DeFi users target a CR that leaves enough buffer to survive normal market noise without tying up more capital than necessary. For ETH or WBTC collateral, a CR of 180–250% is a common range — high enough to absorb a 30–50% drawdown without liquidation, but low enough to keep capital efficiency reasonable. For stablecoin collateral, the range can be tighter, around 140–180%, since the asset itself moves very little. For long-tail altcoins, most users target 250–400% because both the asset volatility and the lower liquidation threshold demand a larger buffer. The key consideration is not what CR you want today, but how much adverse price action the position can absorb: a CR of 200% on ETH absorbs roughly a 37% drop before liquidation, which may or may not be enough depending on your view of the market over the holding period.

The health factor is a rescaling of the collateralization ratio that puts the liquidation point at exactly 1.0, which makes it easier to reason about at a glance. It is defined as (Collateral × LT) / Debt, or equivalently CR × LT / 10000. When HF equals 1.0, the position is exactly at the liquidation threshold; above 1.0 it is safe, below 1.0 it is liquidatable. The relationship to CR is simple: HF = CR × LT / 10000, so a position at 200% CR with an 80% LT has a health factor of 1.60. Because the health factor is normalized, the same number means the same thing regardless of the asset — an HF of 1.5 is comfortable on any collateral, whereas a CR of 150% means very different things on ETH versus a long-tail altcoin. This is why most lending interfaces display the health factor prominently and bury the raw CR in the details.

Once your CR drops below the liquidation CR — equivalently, once your health factor drops to 1.0 — your position becomes eligible for liquidation by anyone. Liquidators are typically bots that monitor the chain for undercollateralized positions, and they repay part or all of your debt in exchange for a discounted portion of your collateral. The discount is the liquidator's incentive; on Aave it is typically 5–10%, meaning you lose that percentage of the collateral used to repay the debt. You do not lose the entire position at once — most protocols allow partial liquidations that restore the health factor to a safe level — but the loss is real and the process is fast. The way to avoid it is to keep a healthy buffer, watch your health factor, and add collateral or repay debt when the buffer gets thin.

Yes, and it is easy to underestimate. Debt grows continuously through interest accrual, which means your collateralization ratio drifts downward every block even if collateral prices are completely flat. On a loan at 200% CR with a 4% annual borrow rate, CR falls by roughly 8 percentage points over a year — from 200% to about 192%. On a 100% borrow rate (common on long-tail stablecoins), the drift is much faster: 200% to about 167% in a year, and to about 133% in two years. This is why borrowers on high-interest assets need to either monitor their position actively, add collateral periodically, or target a much higher starting CR. The health factor captures the same effect from the other direction, so tracking the health factor and setting alerts at a comfortable level is the standard practice for anyone holding a leveraged position for more than a few days.

Yes — and this is the most common misconception. Liquidation does not require insolvency; it only requires that your position falls below the protocol's liquidation threshold, which is much higher than 100%. On ETH with an 82.5% liquidation threshold, the liquidation CR is 121%, meaning a position with 150% CR is safe but one with 120% CR is already liquidatable. The protocol is not waiting for your collateral to be worth less than your debt; it is enforcing a margin requirement that keeps a buffer between the loan and insolvency, so that even in a liquidation the collateral is worth more than the debt being repaid. The liquidation penalty goes to the liquidator as compensation for stepping in, and the remaining collateral (if any) is returned to you. So a position at 110% CR can be liquidated, even though 110% > 100%, because 110% is below the 121% liquidation line for ETH.

When a position has multiple collateral assets, the effective collateralization ratio uses a blended liquidation threshold weighted by each asset's contribution to total collateral value. If you deposit $60,000 of ETH (LT 82.5%) and $40,000 of WBTC (LT 78%), the weighted liquidation threshold is 0.6 × 82.5 + 0.4 × 78 = 80.7%, and the effective liquidation CR is 10000 ÷ 80.7 = 123.9%. The raw collateral value is still $100,000, so the CR formula is unchanged — only the LT and the resulting liquidation CR change. In practice this means that adding a lower-threshold asset to your collateral mix reduces the blended LT and pushes your liquidation CR higher, giving you less buffer at the same total collateral value. Some protocols handle this differently (Aave V3, for example, uses a "liquidation threshold" per asset and computes the health factor as the sum of each asset's collateral value times its own LT, divided by total debt) but the practical effect is the same.

Higher CR means more safety but lower capital efficiency. At 200% CR you are using $2 of collateral per $1 of debt, which means half your capital is idle from a returns perspective. At 125% CR — the liquidation line for ETH — you are using $1.25 per $1 of debt, which is far more efficient but leaves essentially no buffer. The optimal CR depends on what you are trying to do. If you are taking a leveraged long position on an asset you expect to appreciate, you want the lowest CR that still leaves enough buffer to survive normal volatility without liquidation — which is usually somewhere in the 150–180% range for ETH. If you are using a loan as a source of liquidity without wanting to sell your collateral (the classic "borrow against your ETH" use case), you want a much higher CR in the 300%+ range to minimise the chance of being forced out of your position at a bad moment. The right answer is always the one that matches your strategy and your tolerance for forced liquidation.