Crypto Volatility and Round-the-Clock Markets

The short answer

Traditional risk management leans on a close it never mentions. The bell ends the session, caps the day, and makes "I will look at it in the morning" a coherent plan.

Crypto removes that. The position is live at every hour you are not watching it, the market is reachable at every hour you should not be trading it, and both of those are risk-management problems before they are anything else.

What no close does

A session boundary quietly performs four jobs. Remove it and each one becomes yours.

It ends the trading day. Without a close there is no natural point at which you are finished. Losses can be chased at 3am, and the decision to keep going never has to be made explicitly — it just happens by not stopping. Overtrading and revenge trading is the mechanism; a continuous market removes the last external brake on it.

It creates a reference point. Traditional markets give you an open, a close, a prior settlement. Crypto's "daily close" is a convention someone chose — typically midnight in some time zone — and different venues and charts may not choose the same one. Any rule that depends on a daily candle is depending on that arbitrary boundary, so define which one you mean and keep it fixed.

It concentrates liquidity. A session gathers participants into the same hours. Continuous trading spreads them across the clock, so depth varies enormously by hour and by day of week, and no venue is obliged to tell you when.

It bounds unattended exposure. In a market that closes, overnight risk is a known window. In crypto it is every hour you are asleep, and the position is exposed for all of them.

The practical response is to impose the boundary the venue does not: fixed trading hours, a fixed end to the day, and a rule that positions held outside those hours are sized as unattended positions. Nothing except you will create it.

Gaps without a session break

A common misreading is that continuous trading prevents gaps. It does not, because a gap is not caused by the close. A gap is what happens when there are no trades at the intervening prices.

Two ways that occurs in a market that never shuts:

• Liquidity withdrawal. Resting orders are cancelled faster than price moves. Market makers widen or pull entirely when uncertainty spikes, and the book thins out from the inside. Price then travels across empty space in one move.
• Forced flow. Leveraged positions being liquidated generate market orders that consume what depth remains, which moves price, which triggers further liquidations. The cascade is a crypto-specific amplifier, described in how crypto markets differ, and it operates fastest when the book is thinnest.

Add venue-level failure modes — outages, degraded APIs, withdrawal pauses during exactly the volatility that makes you want to act — and the picture is clear enough. Continuous access is not continuous liquidity, and it is not a guarantee that you can transact at the price you can see.

Why thin hours are expensive

Cost is a function of depth, not of the clock. But depth follows the clock, so the clock predicts the cost.

In thin conditions, three things move together: the spread widens, the depth at each level shrinks, and the same order size therefore walks further up or down the book. This applies on entry, on exit, and — the one people forget — to a stop that triggers while you are asleep.

The numbers below are invented for teaching. They describe no real instrument and are not a trade idea.

Suppose an asset trades with a $0.02 spread during the busiest hours and $0.10 in the quietest, and that an order which fills within one tick in busy conditions walks three ticks in quiet ones. Nothing about your strategy changed. Your round-trip cost did, several times over, and your stop now fills further from its trigger for the same reason.

Two consequences worth acting on:

1. Size for the hour you are trading. The depth you checked during peak hours is not the depth present at the weekend.
2. Expect worse stop fills in quiet hours. A stop is a trigger that becomes a market order — it fills at whatever liquidity exists, which is the least at precisely the hours you are not watching.

Stops in a market that never closes

Three adjustments, none of them exotic.

Place the stop where the idea is wrong, not where the loss feels tolerable. That is the general rule from stop losses and invalidation, and it matters more here because crypto's normal noise is wider. A stop that would be sensibly placed in a quieter market can sit inside routine movement in this one.

Measure normal movement before choosing the distance. Use a range measure over a recent window rather than intuition — volatility and average range covers the method. The purpose is to answer one question: is this stop outside ordinary noise, or inside it?

Let the distance set the size, never the reverse. A wider stop with a smaller position carries identical risk to a tighter stop with a larger one, and gives the idea room to be right. The position size calculator turns that into arithmetic; position sizing explained explains why the sequence runs in that order.

What does not work is the reflex fix: tightening the stop to reduce risk. That does not reduce risk, it reduces the probability of reaching the target while raising the probability of being stopped by noise. Reduce the size instead.

Sleep

Sleep is a risk parameter here, which is an odd sentence and a true one.

Assume you will be asleep during the largest move of the week, because you may be. That assumption produces a simple test for every position you hold overnight: if the adverse move you would call severe happened while you were unreachable, and your stop filled meaningfully worse than its trigger, would the outcome be acceptable?

If not, the position is too large. It is not too risky in some abstract sense — it is too large, and the fix is arithmetic.

Two supporting habits:

• Alerts rather than vigilance. Set price alerts and go to bed. Watching a chart at 4am does not improve a decision; it just makes you present for the impulse.
• Decide before, not during. Write down what you will do at each level while the market is calm. A decision made at 4am in a fast market is made by a different person than the one who planned the trade.

What to do with this

Set your own close. Pick hours, write them down, and treat the venue as shut outside them.

Then re-run your sizing on the assumption that you are asleep, the book is thin, and your stop fills worse than it should. Whatever size survives that set of assumptions is the size the position should have been all along — and the one that lets you stop looking at it.