Risk management in crypto is the set of decisions that determine how much you can lose before you find out whether you were right. It is not a personality trait and it is not a single number. It is a chain: a loss budget, a position size that fits it, an invalidation point that defines it, leverage and venue choices that do not silently break it, and custody that keeps the capital there to begin with.
Nothing here guarantees a loss will stay inside a plan. Stops can fill late or not at all, venues can halt, bridges can fail, and correlated positions can lose together. Risk management lowers the frequency and size of avoidable losses; it cannot remove the unavoidable ones.
Published September 3, 2026. Numbers are hypothetical and illustrate the arithmetic, not a recommended allocation.
What crypto risk management actually is
Definition: Crypto risk management is the process of deciding, before a trade, the maximum acceptable loss for the idea and for the whole account, then choosing size, exit, leverage, venue, and custody so that the realised loss stays close to that decision most of the time.
Most losses that end trading accounts are not bad forecasts. They are unbounded positions: trades sized to a feeling, leverage chosen from a dropdown, stops that existed only as intentions, and exchange balances larger than anyone would keep in a wallet they did not control. Each of those is a decision that was never explicitly made. This guide makes them explicit, in the order they matter.
The chain has six links. A weak one anywhere undoes the rest:
- Loss budget — how much of the account one idea may cost, and how much all open ideas may cost together.
- Position size — the order quantity that turns the budget and the exit into a number.
- Invalidation — the price or condition at which the idea is wrong and the position is closed.
- Leverage and liquidation — the product mechanics that can end the trade before the invalidation point.
- Correlation and portfolio heat — how many of your open ideas are actually the same bet.
- Custody and counterparty — where the capital sits and who can lose it for you.
The loss budget: decide the number before the trade
Start with the account value that is genuinely available for speculation. Exclude borrowed money, emergency funds, tax reserves, and anything needed within a year. Then set two budgets:
- Per-trade budget: the fraction of the account one idea may lose. Educational material commonly discusses figures around 0.5–2% for active traders; the right figure is the one you can lose ten times in a row and still trade with a clear head.
- Portfolio budget: the total at risk across all open positions if every stop is hit. This is the number people forget. Five positions at 2% each is 10% of the account exposed to one bad afternoon, and if the five are correlated it is closer to one 10% position.
A percentage budget shrinks as the account shrinks, which slows the compounding of losses, and grows as the account grows without requiring a decision. A fixed dollar budget does neither.
Write both budgets down. The trading journal template has a field for them because a budget that is not recorded is renegotiated in the moment, usually upward.
Position sizing: the budget becomes a quantity
Position sizing is where the loss budget meets the chart. The basic form is:
Quantity = (Account × per-trade risk %) ÷ (Entry price − Exit price + costs per unit)
Worked example: a $10,000 account, a 1% budget ($100), an entry at $2.00, an invalidation exit at $1.80, and roughly $0.01 per unit in fees and expected slippage. The loss per unit is $0.21, so the quantity is about 476 units, or roughly $952 of exposure. Note what the formula did: it made the position smaller because the stop was wider, and it would make the position larger for a tighter stop. Size follows the exit; the exit never follows the size.
The full derivation, the portfolio-level checks, and the cases where the simple formula breaks (leverage, correlation, fast markets) are in the crypto position sizing guide. For the arithmetic itself, the position size calculator accepts an account, a risk percentage, entry, and stop and returns the quantity; the how much crypto should I buy tool walks through the same decision interactively for people who have not sized a trade before.
Invalidation: define "wrong" before you are in
An invalidation point is the price or condition at which the reason for the trade no longer exists. It is not the point at which the loss feels large. For a trade taken because a level held, the invalidation is a close through that level; for a trade taken on a trend, it is the structure break that ends the trend. The support and resistance guide explains why levels should be treated as zones rather than lines, and why a stop placed exactly at a well-known level is often the first thing a fast market takes out.
Three rules keep invalidation honest:
- The stop is the trade. If you would not accept the loss at the invalidation point, the trade is too large, not the stop too tight.
- Structure, then distance. Choose the invalidation from the chart first, then let the formula set the size. Choosing a stop distance to fit a desired size reverses the logic.
- Reward is measured from invalidation. The risk/reward calculator compares the distance to the target with the distance to the stop; FullSwing's own daytrade alerts are only issued when that ratio clears a minimum of 3:1, because a trade that risks a dollar to make fifty cents needs to be right most of the time to break even.
Leverage and liquidation: the exit you did not choose
Leverage does not change how much an asset moves. It changes how much of your account each move represents, and it introduces a second exit, the liquidation price, that the venue enforces regardless of your plan. Two consequences follow.
First, the loss budget must be checked against the liquidation distance, not only the stop distance. If a position can be liquidated before it reaches your invalidation point, the invalidation point is fiction. The liquidation price calculator shows the price at which a position is closed for a given leverage, entry, and margin mode; if that price sits inside the range you expect the market to trade in, the leverage is wrong for the idea.
Second, cross-margin settings mean one position's losses can consume the collateral of another. Several small positions on cross margin behave like one large position. The Hyperliquid guide works through margin, funding, and liquidation mechanics on a perpetual venue in detail; the mechanics differ by venue, and the exchange's own documentation is the authority for the one you use.
Two-times leverage on a position that is already three times too large for the loss budget is six times too large. Leverage multiplies a sizing decision; it does not replace one.
Correlation and portfolio heat: how many bets do you really have?
Crypto assets move together far more than their narratives suggest. In a sharp Bitcoin decline, most altcoins fall harder, and long positions in five different tokens become a single leveraged bet on the market. Portfolio heat is the sum of what every open position would lose at its stop; correlation is the reason that sum understates the real exposure.
Practical controls:
- Add up the stop losses of all open positions before opening another one. If the total exceeds the portfolio budget, something closes first.
- Treat positions in the same sector, or in assets that historically move with Bitcoin, as fractions of one position when sizing. The correlation strategies guide shows how to measure the relationship instead of guessing it.
- Check the market regime. FullSwing's discovery and grammar alerts carry a Bitcoin day-change gate for this reason: a setup that looks clean on one coin is a different trade when the whole market is down 3% on the day.
Custody and counterparty: risk that has nothing to do with price
A perfectly sized, correctly stopped position is still lost if the exchange freezes withdrawals, the bridge is exploited, or the wallet's recovery phrase was photographed. Custody risk is the part of risk management that price charts cannot show, and it is where the largest single losses in crypto's history have happened.
- Keep on the venue only what the current plan needs. Trading capital lives on the exchange; the rest lives in a wallet whose keys you control. The wallet security and recovery guide compares custody models by threat model rather than by brand.
- Treat funded-trading programs as counterparties. A prop firm's rules are risk rules you did not write, and its solvency is a risk you did not choose. The due-diligence checklist in that guide applies before any fee is paid.
- Bridges and new chains are operational risk. Moving funds to a new venue is itself a trade with a loss budget: never bridge more than the plan needs, and test with a small amount first.
The process that keeps the chain intact
Every link above is a decision, and decisions made under pressure drift toward larger size and later exits. Process is the mechanism that moves the decisions out of the moment:
Write the rules once. Budgets, maximum leverage, maximum open positions, and the conditions under which you do not trade at all. The trading psychology and process guide covers why written rules outperform remembered ones.
Rehearse before risking. A paper-trading account tests whether the rules are executable, not whether the strategy is profitable; that distinction is the point.
Record and grade. Every trade gets a journal entry with the planned and realised loss. FullSwing grades its own alerts against what the market did 24 and 48 hours later and publishes the delayed history in the Alert Explorer; the same discipline applied to a personal account is what turns a loss into information.
Review the common failures. The reasons most traders lose money and the beginner mistakes guides are checklists of the links in this chain that most often break.
Drawdowns: the arithmetic that decides survival
The reason budgets are small is not caution; it is arithmetic. A 10% loss needs an 11% gain to recover, a 30% loss needs 43%, and a 50% loss needs 100%. Recovery requirements grow faster than losses, which means the single most important property of a risk plan is that it keeps drawdowns shallow enough to be recoverable by ordinary trading rather than by a miracle.
| Drawdown | Gain needed to recover | Consecutive 1% losses to get there |
|---|---|---|
| 5% | 5.3% | 5 |
| 10% | 11.1% | 11 |
| 20% | 25% | 22 |
| 30% | 42.9% | 36 |
| 50% | 100% | 69 |
The third column is the argument for a 1% per-trade budget in one line: it takes 69 consecutive planned losses to halve the account, and a run of that length is a signal to stop long before it completes. At 5% per trade the same halving takes 14 losses, which any strategy will produce eventually. Set a maximum drawdown at which trading stops for review (many traders use 10–15%), and treat reaching it as information, not as a reason to trade back to even.
A pre-trade risk checklist
Sources
- Investor.gov: Crypto assets — the SEC's investor-education overview of crypto-asset risks, including custody and platform risk.
- CFTC: Digital assets — customer advisories on leveraged and derivative crypto products.
- BIS Working Paper 1049 — evidence on retail crypto trading behaviour and outcomes across 2015–2022.
Join the Discussion