Money Management in Trading: How to Manage Your Capital

Accounts don't get blown up by bad entries. They get blown up by size. Money management in trading is about exactly the second one.

A trader who is right 40% of the time makes money for years if every trade risks the same fraction of the account. A trader who is right 65% of the time zeroes out in a week if he doubles his size after three stops. The difference isn't in market analysis, it's in arithmetic.

Let's go through that arithmetic step by step: how much to put on a trade, how to size it against a specific stop, where to set the daily cutoff, and when a position can be scaled up.

What money management is

Money management in trading is the set of rules that define the size of every position and the maximum loss over a period. It answers the question "how much," while the strategy answers "where" and "when."

It often gets confused with risk management. That one is broader: instrument selection, correlation control, working with leverage, protecting API keys. Capital management is its monetary core. Three numbers:

• allowable risk per trade as a percentage of the deposit;
• loss limit per trading day and per week;
• the step by which size changes as the account grows and draws down.

Money management in crypto is stricter than on the stock market: trading runs around the clock, leverage goes up to 100x, and there's no weekend break to cool off. Everything else is built on top of that. We covered the wider account-protection framework separately in the article "Risk Management in Trading".

Why percentages don't work symmetrically

A drawdown and the recovery from it are different numbers. Losing 20% of the deposit is easy, but you have to earn those 20% back from a smaller base, so in percentage terms you need more.

The "50%" row is the key one here: half the deposit is recovered by doubling what's left, and that takes months. Anything deeper than 30% breaks your psyche before it breaks the math. A person starts trading not by the system but by the need to win it back.

A losing streak will happen, guaranteed. At a 50% win rate the probability of five stops in a row is 3.1% for any given group of five trades, but across 200 trades such a streak shows up with a probability above 95%. Not "if," but "when."

At 1%, five stops are barely noticeable. At 20% there's nothing left to trade with afterwards: to get back to a thousand you need +205%.

If futures mechanics are still shaky for you, watch the free lesson from the course "Trading from Scratch" on our YouTube channel: it covers exchanges, contracts, and leverage.

Position size: how to calculate it

Position size in trading is calculated from the stop, not from how much you feel like trading. First the trader decides where the trade is proven wrong, then converts that distance into size. The reverse order (size first, then a stop wherever it fits) is the main reason accounts get wiped.

A fixed % of the deposit

The working range for risk per trade is 0.5% to 2% of the deposit. Scalpers with a high trade count usually keep 0.25–0.5%, medium-term traders 1–2%.

Why that range. At 1% and a run of 10 stops in a row the drawdown comes to 9.6%, and the account is still alive. At 5% the same streak takes 40%, and past that point you enter the zone few traders come back from.

There are two ways to apply the percentage.

Fixed fraction. The percentage is taken from the current balance. The deposit grows, the risk amount grows with it. The deposit falls, size automatically shrinks. A built-in brake during a drawdown.

Fixed amount. The trader works with template sizes (1 000, 2 000, 5 000 USDT) and knows the loss in dollars in advance. Faster to execute, which matters for scalping, but the templates need a manual review once a month.

I keep 0.75% per trade and recalculate the risk amount on Mondays, then work with fixed sizes during the week. That way I don't have to do the math at the moment of entry, when large fills are already coming through the tape and there isn't a spare second for a calculator.

The formula: risk, stop, size

Position size (in units of the asset) = (Deposit × Risk%) / (Entry price − Stop price)

Position notional (USDT) = (Deposit × Risk%) / Stop size in %

Let's put numbers on it. Deposit 5 000 USDT, risk 1%, that is 50 USDT. Entry on BTC/USDT at 67 200, stop under the local low at 66 900.

• Distance to the stop: 67 200 − 66 900 = 300 USDT, or 0.446% of the entry price.
• Size: 50 / 300 = 0.1667 BTC.
• Notional: 0.1667 × 67 200 = 11 200 USDT.
• Required leverage: 11 200 / 5 000 = 2.24x.

Leverage here isn't chosen, it's calculated. It came out of the stop and the allowable risk, not the other way around.

Fees eat part of the risk. The taker fee on futures at most venues is around 0.05% of notional, and entry plus exit gives 0.1%. On an 11 200 USDT notional that's 11.2 USDT, which is 22% of the 50 USDT of planned risk. Plus the funding rate if the position is held through a funding settlement.

Calculations table for a 5 000 USDT deposit and 1% risk (50 USDT):

Look at the last two rows. The wider the stop, the smaller the notional and the lower the leverage. The dollar risk stays the same everywhere, 50 USDT. A trader who slaps a fixed 20x on everything risks pennies on a tight stop and takes a loss several times larger than planned on a wide one.

The position calculator

A position calculator is six input fields and three result lines. It takes ten minutes to build in an ordinary spreadsheet.

Inputs: balance, risk in percent, entry price, stop price, take price, the venue's fee. Outputs: size in units of the asset, notional, required leverage, R:R, and the forced liquidation price.

That last line matters more than it looks. The liquidation price has to sit noticeably further out than the stop, otherwise the exchange closes the position before your protective order does, and the loss isn't 1% but the whole margin.

An example. On isolated margin with 20x leverage and a 0.4% maintenance margin rate, liquidation hits at roughly a 4.6% move against the position. The 0.45% stop from the calculation above doesn't come anywhere near it, but a 4% stop at the same leverage is already dangerously close to the edge. Rates differ across venues and depend on position size; on Binance you can see them in the contract specification.

Liquidation price (isolated long) ≈ Entry price × (1 − 1/Leverage + Maintenance margin rate)

You can work this out in a notebook, but that's awkward at the moment of entry. In Secret Terminal the order lines show size and distance to price, stop and take are placed with the Z key, and "Magnet" snaps the protective order exactly to the candle's extreme. From there you drag the level around the chart with the mouse and the size stays the same.

The daily loss limit

The daily loss limit is a pre-set loss amount after which trading stops for the day. A working value: 3 stops in a row or 3% of the deposit, whichever comes first.

The limit isn't there for the arithmetic, it's there against tilt. After the second or third stop the brain switches from "I execute the system" mode to "get the money back" mode. The trader starts entering without a signal, dragging the stop, adding to a losing position. Those three moves are described in the article "Beginner Mistakes in Trading" and they kill accounts faster than any volatility.

A word on the giveback rule. You make +2.5% over the morning, then the market stalls and the profit starts melting. The moment you've given back half (the account is still +1.25%), the day is closed. It's protection against the classic case where a green day turns red in the last forty minutes.

I hang the limit on the performance calendar in my journal: two red days in a row, and on the third I trade half size. Even if the picture in the order book is perfect. It works better than willpower.

After that all orders are pulled with the spacebar and Secret Terminal gets shut down. Not minimized, shut down: an open order book with prints flashing is an invitation to a fourth trade. How personal rules grow out of journal statistics is covered in the article "Discipline in Trading".

Scaling a position up and down

You can only scale into a position that's in profit. Averaging into a losing one is off-limits unless the ladder was written into the plan before the entry. That rule is worth memorizing word for word.

Pyramiding: how to add correctly

Conditions for adding:

• The position is in profit, the stop is moved at least to breakeven.
• Each subsequent entry is smaller in size than the previous one.
• Total risk after the add-on does not exceed the original 1R.

In numbers. First entry 0.167 BTC at 67 200, stop 66 900, risk 50 USDT. Price reaches 67 600, the stop moves to breakeven at 67 200. You add 0.08 BTC at 67 600 with the stop at the same level: the risk on the second piece is 0.08 × 400 = 32 USDT. Total risk is now 32 USDT against the original 50, size is up 48%, average entry price is 67 330. Exiting at 68 200 gives about 215 USDT of profit across both pieces, that is 4.3R.

Trimming and partial profit-taking

The reverse mechanic kicks in when the move runs out of steam. The signals show up in order flow: the speed of fills on the tape drops, the opposing density level in the order book isn't getting eaten through, and on the footprint the volume shifts to the seller's side. Three indicators in a row, and I unload half.

The working template: take half the size off at 1R, move the rest to breakeven. From there it either runs to 2–3R or closes at the entry price. Expectancy barely suffers, the psychological load drops noticeably.

Scaling to the size of the deposit

The anti-martingale logic: increase risk on a run of winning trades, cut it on a run of losing ones. Martingale works right up until the first long streak of stops, after which there's no account left.

I've tested the scheme on BTC/USDT and on second-tier alts. On bitcoin, half size after a drawdown barely affects the monthly result, there are simply too many trades. On illiquid alts the effect is stronger: there, slippage alone eats part of the risk.

Common mistakes

• Picking the leverage first, then fitting the stop to it. The order is the opposite: the stop comes off the chart, the size is calculated to fit it.
• Calculating risk off free margin instead of the whole deposit. With partial withdrawals the risk amount creeps up without the trader noticing.
• Not building the fee into the calculation. At high frequency that's 20–25% on top of the planned loss on every trade.
• A daily limit that exists on paper but nobody actually checks. Without a journal the limit stays an intention.
• Doubling size after a profitable week. Anti-martingale means a step, not a jump.

A breakdown of real trades and entry mistakes is in the free lesson of our beginner course, where the market is analyzed through the order book and footprint.

When money management doesn't save you

A correctly calculated size doesn't protect against a gap. If a position is held over the weekend or through a macro data release, price can open past your stop and the loss turns out two or three times larger than the calculation.

A live example. A scalper is long an altcoin, stop 1.5%, risk 1% of the deposit. A delisting headline comes out, the order book empties in a second, and the nearest bid on the other side sits 9% lower. The stop fills against it: instead of 50 USDT the loss is 300 USDT, six planned stops at once.

The second scenario: pyramiding in a chop. The scheme works while the market is trending. In a range every add-on catches a stop at breakeven, and the day ends flat minus fees.

The protection here isn't in the calculations, it's in the restrictions: don't hold through known events, cut size on illiquid instruments, don't pyramid inside a range.

FAQ

• What percentage of the deposit should I risk on one trade?

  From 0.5% to 2%, with 1% being the standard for most. Scalpers with dozens of trades a day are better off dropping to 0.25–0.5%, otherwise fees plus a streak of stops in a single session will produce a drawdown of 10% or more.

• How do I calculate position size for a specific stop?

  Divide the dollar risk by the dollar distance to the stop. A 3 000 USDT deposit, 1% risk (30 USDT), entry at 67 200 and a stop at 66 800 gives a size of 0.075 BTC, a notional of about 5 040 USDT and 1.68x leverage. Leverage comes out of the calculation, it isn't picked in advance.

• What leverage is safe for a beginner?

  The leverage cap is secondary, the risk cap comes first. With size calculated properly, 3–5x on bitcoin gives stops of 0.5–1%, which is enough for intraday trading. Leverage of 20x and up demands stops so tight that market noise takes the position out.

• What should I do after three losing trades in a row?

  Stop trading for the day. Three stops in a row mean either the strategy doesn't fit the current state of the market or you've lost concentration. Both are treated with a pause, not with bigger size.

• Can I average into a losing position?

  Only if the ladder was written into the plan before the entry, with the add-on levels calculated in advance and the total stop within 1R. Spontaneous averaging increases risk at the exact moment the market has already shown the idea was wrong.

• How do fees affect the risk calculation?

  On an 11 200 USDT notional with a 0.05% taker fee, entry and exit cost 11.2 USDT. If the risk per trade is 50 USDT, fees take 22% out of the planned loss. At high frequency the difference between maker and taker execution changes the annual result more than most tweaks to the strategy.

• How do I know that money management in trading is working?

  By the equity curve and the depth of the maximum drawdown. If over three months the maximum drawdown hasn't exceeded 10–12% and the curve climbs in steps without cliffs, the system works. Sharp vertical drops mean size got out of control somewhere.

What to do next

Capital management in trading only pays off with daily execution. Work out the risk amount for your deposit, set the daily limit in stone, and start keeping trade statistics.

The second part can be automated: the "Accounts" and "Journal" modules pull trades via API, color every day of the calendar by result, and build the equity curve.

The terminal is free and connects to Binance, Bybit, OKX, MEXC and WhiteBIT.

Download Secret Terminal