Borrowing against crypto is a bet on your collateral not falling far enough to be taken. This turns your position into a health factor, and then into the one number that matters: how far the price can drop before somebody else sells it for you.
A health factor is a useful summary and a poor motivator. It is an abstraction, and 1.4 does not feel like anything. The figure that actually changes behaviour is the price your collateral has to reach before the position is taken, because that is a number you can compare against what the market did last month.
So this calculates both. The health factor tells you where you stand. The liquidation price tells you what would have to happen, and whether that is a drop you have already seen this year.
Liquidation does not need a crash, it needs a wick. Prices on thin venues move further than the market as a whole during panics, and the oracle a protocol reads may briefly show a price no orderly market ever printed. A position that would have survived the day can be closed on a spike that lasts one block.
The second thing that eats a buffer is time. Interest accrues on the debt, so the position drifts towards liquidation on its own even when the price does nothing at all. A health factor is a snapshot, and if you opened at 1.3 and stopped looking, the number you remember is not the number you have.
It is rarely a bad view on the market. It is borrowing the maximum on day one. The maximum LTV is where the protocol will let you start, not where it is sensible to sit, and starting there means the first ordinary drawdown does what ordinary drawdowns do.
The other version is borrowing against the same asset you are trying to keep exposure to. If your collateral is the thing you are bullish on, and you borrowed to buy more of it, then a fall does not hit you once. It hits your collateral and your position at the same time, which is the mechanism behind almost every story that ends with somebody losing far more than they thought they had risked.
It is your collateral, weighted by how much of it the protocol is willing to lend against, divided by what you owe. Above one you are fine. At one you are liquidated. It is deliberately built so that a single number tells you whether the position survives, and the further above one it sits, the more room the price has to move against you.
The loan to value ratio is how much you are allowed to borrow when you open the position. The liquidation threshold is how far you are allowed to drift before you are closed. The threshold is always the higher of the two, and the gap between them is the buffer the protocol gives you. Borrowing to the maximum LTV means starting your life inside that buffer, which is why it is such a bad idea.
A liquidator repays part of your debt and takes an equivalent amount of your collateral, plus a bonus of usually five to ten percent for doing it. You keep whatever is left. It is not a total loss, but the bonus is paid out of your position, so it is an expensive way to close a trade and you do not get to choose the moment.
Yes, and this is the honest reason to keep a health factor well above one. Liquidation is automatic, competitive and continuous. Bots watch every position on every major protocol and act within a block of it becoming profitable. There is no notice, no call, and nobody to ask for more time.
There is no safe, only more or less room. Below 1.5 you are exposed to an ordinary bad day. Between 1.5 and 2 you survive most normal volatility. Above 2 you are borrowing conservatively. Match it to the collateral: a stablecoin position and a volatile token position with the same health factor are not carrying the same risk at all.
The arithmetic is the same but it uses the weighted average of all your collateral and the total of all your debt. This calculator handles the single position case, which is where most people are, and the same intuition carries over.
We seal the list every week and keep re-checking every token on it, so you can see what actually happened to them rather than only what is trading today.
