Position sizing in crypto: the 1% rule, with real numbers

Most people decide how much to buy and think about the stop later. Position sizing turns that around: you decide how much of the account you accept to lose and where the trade is wrong, and the size follows from those two numbers. This guide shows the arithmetic on a $10,000 account, then where it breaks: leverage, fees and funding, gaps, speculative tokens and positions that move together.

One thing first. Sizing does not make a trade good. A correctly sized bad trade still loses. Sizing only decides how much a loss costs, so that a run of them doesn't end your account. This is about limiting losses, not making money.

What position sizing is

Position sizing is choosing how large a position to open so that, if the price reaches your stop, you lose a set amount and no more. The set amount is usually a small fixed share of the account. The best-known version is the 1% rule: risk no more than 1% of the account on any single trade.

The 1% is not a law; 0.5% or 2% work the same way. What matters is that the number is fixed before the trade, small enough to survive a long losing streak, and applied every time.

The formula: risk amount ÷ stop distance

You need two inputs:

Risk amount: the dollars you accept to lose. On a $10,000 account at 1%, that is $10,000 × 1% = $100.
Stop distance: how far, in percent, the price can move against you before the idea is wrong and you exit.

Then:

Position size = risk amount ÷ stop distance

With a stop 5% away: $100 ÷ 0.05 = $2,000.

Check it the other way round: a $2,000 position that falls 5% loses $2,000 × 5% = $100, which is 1% of the account.

A worked example in BTC. Say BTC is trading at $80,000 (a round number for the arithmetic, not a level we expect) and the chart tells you the idea is wrong below $76,000. The stop distance is ($80,000 − $76,000) ÷ $80,000 = 5%. The position is $2,000, which is $2,000 ÷ $80,000 = 0.025 BTC. If the stop is hit, you lose 0.025 × $4,000 = $100.

The stop decides the size: wider stop, smaller position. The stop should come from where the idea fails, never from how big a position you would like to hold.

Position sizes on a $10,000 account

Risk per trade	Risk amount	Stop distance	Position size	Share of account
1%	$100	2%	$5,000	50%
1%	$100	5%	$2,000	20%
1%	$100	10%	$1,000	10%
1%	$100	20%	$500	5%
0.5%	$50	5%	$1,000	10%
2%	$200	5%	$4,000	40%

A $5,000 position and a $1,000 position can carry the same 1% risk: size alone says nothing about risk. And a tight stop produces a large position, which is where fees, slippage and gaps hurt most.

Why small and fixed: losing streaks and the Kelly criterion

Losses come in runs. If you risk a fixed share of what is left after each loss, here is what ten losses in a row do to the account:

Risk per trade	After 10 straight losses	Drawdown	Gain needed to get back
1%	$9,044	9.6%	10.6%
2%	$8,171	18.3%	22.4%
5%	$5,987	40.1%	67.0%

The arithmetic: $10,000 × 0.99¹⁰ = $9,044, × 0.98¹⁰ = $8,171, × 0.95¹⁰ = $5,987; and $10,000 ÷ $5,987 = 1.670, a 67.0% gain to recover.

There is a formula for the "best" bet size, the Kelly criterion, from a 1956 paper by John Kelly, which sizes each bet to maximise long-run growth. A review by MacLean, Thorp and Ziemba explains why it is hard to use as a trading rule: its bets can be very large and very risky in the short term, betting more than Kelly brings more risk and lower growth, and in continuous time betting double Kelly cuts growth to the risk-free rate. It only works "with good data input", and errors in the expected return matter far more than errors in volatility.

Kelly needs your true edge, and almost nobody knows it. Overestimate it and you overbet. A small fixed fraction such as 1% gives up some growth on paper in exchange for surviving the losing runs you will have.

Leverage changes the margin, not the risk

Leverage lets you open the same position with less collateral. It does not change what the trade risks, if the stop is where you planned:

The $2,000 position with no leverage ties up $2,000.
At 10x, it ties up $2,000 ÷ 10 = $200 of margin.
In both cases, a 5% move to the stop loses $100.

The common mistake is to size by margin. Post $1,000 of margin at 10x and you hold a $10,000 position. A 5% stop on that is $10,000 × 5% = $500, which is 5% of the account, not 1%.

Leverage also adds an exit you did not choose: liquidation, when losses eat the margin and the venue closes the position. The CFTC warns that leverage magnifies losses and that traders may lose more than they put in; the SEC notes a broker can sell to cover a margin shortfall without telling you first. In the UK, the FCA has banned the sale of crypto derivatives to retail consumers since 6 January 2021, citing sudden and unexpected losses.

If the only money behind the position is the margin you posted, the move that wipes it out is roughly 1 ÷ leverage, and liquidation comes a little earlier because venues require a minimum maintenance margin:

Leverage	Margin on a $2,000 position	Move that wipes the margin	Stop at 5%
2x	$1,000	50%	Stop comes first
10x	$200	10%	Stop comes first
25x	$80	4%	Liquidated before the stop

At 25x, a 4% move costs the full $80, liquidation triggers before that, and the 5% stop is never reached: the venue closes the trade on a move your plan treated as noise. The rule to follow: the liquidation price must sit well beyond the stop, never between your entry and your stop.

Fees and funding add to the loss

The $100 in the formula is the price move only. Real trades cost more.

Trading fees. Fees depend on the venue and the order type. Assume 0.1% per side for the example. Opening $2,000 costs $2.00; closing at the stop, when the position is worth $1,900, costs $1.90. That is $3.90 on top of the $100.

Funding. Perpetual futures have no expiry. To keep their price near spot, longs and shorts exchange a payment, usually every eight hours: the carry cost of a leveraged position. Above spot, longs pay shorts; below spot, shorts pay longs. The payment is per unit held, so it grows with the position. Research on perpetuals notes a common baseline of 0.01% per eight hours when the price gap is small. At that rate, $2,000 pays $0.20 every eight hours, $0.60 a day, $6.00 over ten days. In busy markets it can be far higher.

A long held for ten days and stopped out loses $100 + $3.90 + $6.00 = $109.90, close to 1.1% of the account. If you want the total to stay at 1%, put the costs into the stop distance. A 5% stop plus about 0.2% of fees and 0.3% of funding is 5.5%, and $100 ÷ 0.055 = about $1,818.

Gaps and slippage: a stop is a trigger, not a promise of price

The SEC's investor bulletin explains that a stop price only triggers a market order, and the fill "can deviate significantly from the stop price" depending on the liquidity available. A stop-limit order controls the price, but may not be filled at all if the market moves past the limit.

Crypto trades around the clock, so overnight gaps are rarer, but the same happens in a fast move or on a thin order book. If the $2,000 position is filled 8% below entry instead of 5%, the loss is $2,000 × 8% = $160, 1.6% of the account. If that would be a problem, 1% was too much.

Sizing a speculative token for a zero

For a new or illiquid token, a stop gives little protection: the price can fall 90% between two trades, the pool can be emptied, or nobody may be there to buy. The honest stop distance is 100%.

Same formula: $100 ÷ 1.00 = $100. Accepting a 1% loss on a token that could go to zero means a $100 position, not $2,000. At a lower 0.5% risk, it is $50. Before you buy, run the on-chain checks in our guide on how to size a speculative token so a zero doesn't hurt; they rule out some ways of losing everything, not all of them.

Correlated positions count as one

The 1% rule assumes each trade is a separate bet. In crypto they rarely are: a study of crypto returns by Liu, Tsyvinski and Wu found strong co-movement across coins, with the major coins tracking a crypto market index closely. The IMF found that even bitcoin and the S&P 500 went from a correlation of 0.01 in 2017–19 to 0.36 in 2020–21.

So three longs in three large coins, each at 1%, are closer to one 3% bet on the crypto market: in a broad sell-off all three stops can be hit the same day, for $100 × 3 = $300. Give positions that move together one shared budget. If "long crypto" gets 1.5%, three such trades get 0.5% each, $50 apiece.

A sizing routine before every trade

Turn today's account value (not its peak) into a risk amount: $10,000 × 1% = $100.
Place the stop where the idea is wrong and measure the distance in percent; use 100% for a token that could go to zero.
Add expected fees, and funding for the planned holding time, to that distance.
Divide: risk amount ÷ total distance = position size. Check it backwards.
With leverage, confirm the liquidation price is well beyond the stop.
Add up the risk of open positions that move together, and cut the new one if the group is over budget.

FAQ

What is the 1% rule in trading?

Risk no more than 1% of the account on a single trade: $100 on $10,000 if the stop is hit. It limits the size of a loss; it says nothing about whether the trade is a good one.

How do I calculate position size in crypto?

Divide the dollars you accept to lose by the distance to your stop: $100 ÷ 0.05 = $2,000. Add fees and funding to the distance.

Does leverage increase my risk if I keep the same stop?

Not by itself. A $2,000 position with a 5% stop loses $100 with or without leverage; leverage only reduces the margin you post. Risk rises when leverage is used to open a bigger position than the formula gives, or when liquidation sits before your stop.

Will a stop-loss always close my trade at the stop price?

No. A stop triggers a market order, and in a fast or thin market the fill can be well below the stop. Size so that a worse fill is still a loss you can carry.

How much should I risk per trade?

There is no single right number; the smaller it is, the longer you last. Ten straight losses cost 9.6% of the account at 1% risk and 40.1% at 5%.