Position Sizing: The Math That Decides Whether You Survive

Most retail traders think trading is about entries. It isn't. Backtested across thousands of trades, the difference between winning systems and losing systems is rarely the signal, it's position sizing. Get the size right and a mediocre signal makes money. Get the size wrong and a great signal still goes to zero.

The setup

Three numbers anchor every trade:

• Account size, how much capital you're trading with.
• Risk per trade, how much of that account you're willing to lose if this trade fully fails. Convention: 0.5-2% of account.
• Stop distance, how far in price (in dollars or percent) your stop sits from your entry.

Position size is whatever number satisfies all three. It is not "how much I want to own" or "how confident I feel." It's a calculation.

The formula

position_size = (account_size × risk_per_trade_pct) / stop_distance_pct

Worked example: $10,000 account, 1% risk per trade, stop is 4% below entry. Position size = ($10,000 × 0.01) / 0.04 = $2,500.

Or in dollar terms:

risk_dollars = account × risk_pct       → $10,000 × 1% = $100
position_size = risk_dollars / stop_distance_pct → $100 / 0.04 = $2,500

Either way, the position is sized so that if your stop fires, you lose exactly the dollar amount you decided was acceptable.

That is the whole concept. Everything else is implementation.

Why this beats every alternative

Sizing by feel: "I'm pretty confident, so I'll go bigger." Confidence is unreliable. It correlates poorly with outcomes and strongly with recent results, you'll be biggest right before your worst losses.

Sizing by leverage: "I'll use 10x." Leverage is a slider, not a size. The position size that 10x produces depends on how much margin you allocate, which is the actual sizing decision in disguise.

Sizing by "I can afford it": "I have $10,000, I'll buy $5,000 of BTC." This ignores stop distance. The same $5,000 position with a 1% stop risks $50; with a 10% stop, it risks $500. Same dollars, 10x different actual risk.

The risk-based formula collapses all those degrees of freedom into a single answer. You decide what loss you can stomach, your strategy defines the stop, and the size falls out automatically.

The 1% / 2% rule

The most common starting point: risk 1% of your account per trade.

Why 1%:

• A 10-trade losing streak (which happens) costs you ~10% of your account. Recoverable.
• A 20-trade losing streak (which has happened to most active traders at some point) costs you ~18%. Painful but not fatal.
• Even a brutal 30% drawdown leaves you with enough capital to trade your way back without needing extreme returns.

If you're new, 0.5% is even safer. If you have proven edge over many hundreds of trades, 2% can be defensible. Above 2%, you're playing with ruin probabilities most people don't survive.

The 1% rule does NOT mean "I'll feel 1% pain on this trade." It means "if the worst case I planned for happens, I lose 1%." Most trades don't go to stop. The 1% is a ceiling, not a target.

Rebalancing as the account grows or shrinks

If your account grows from $10,000 to $15,000, your 1% risk allocation grows from $100 to $150. Position sizes scale up proportionally. This is automatic compounding.

If your account drops from $10,000 to $7,000, your 1% drops to $70. Position sizes scale down. This is automatic deleveraging, your strategy gets less aggressive as your capital base shrinks, which prevents drawdowns from snowballing.

The discipline is to recalculate after every meaningful equity change. Many traders compound up enthusiastically but refuse to deleverage on the way down. That's the asymmetry that turns a 20% drawdown into a 60% drawdown.

Adjusting size for conviction (carefully)

A common refinement: vary risk between 0.5% and 1.5% based on conviction. Highest-conviction setups get the full 1.5%; routine setups get 0.5-1%.

This is defensible if and only if:

1. You've measured that your high-conviction setups actually outperform. (Most people's "conviction" doesn't predict outcomes, they're just biased.)
2. The variation is bounded, never above 2%, never below 0.25%.
3. You define "high conviction" with a written rule, not a feeling.

Without these three guardrails, "varying by conviction" is just sizing by emotion with a justification on top.

A common mistake: setting position size first, then deriving stop

Reverse the formula. "I want a $5,000 position. My stop will be... 1% away to keep risk at $50." Now you're forcing a stop that has nothing to do with the chart's invalidation level. Random noise will trigger that stop constantly. You'll get stopped out of trades that ultimately worked.

The correct order is:

1. Decide invalidation level (where the trade thesis is wrong).
2. That defines stop distance.
3. Apply the formula → position size falls out.

If the resulting position size is "too small" to be interesting, that's the trade telling you something. The trade isn't worth your attention if you can only justify a tiny size at proper stop placement. Find a different setup.

A common mistake: not accounting for leverage in size

You apply the formula and get a $2,500 position size. You open it as a 10x perp with $250 margin. Now your "1% risk" is actually... well, it depends on the leverage and the stop. If 10x means liquidation at ~10% adverse move, and your stop is at 4%, the stop fires first and you lose $100 (your 1%). Good.

But if you used 25x, liquidation is at ~4%, dangerously close to your stop at 4%. A wick at 4.01% liquidates you for $250, 2.5x your planned risk. The position size formula assumed you'd be stopped at the stop, not liquidated at the liquidation.

The fix: always check that your liquidation price is materially further from entry than your stop price. If they're close, lower the leverage until there's comfortable separation. Otherwise the formula is lying to you.

Mental model, position size as how thick your wallet is, not how big your bet

Think of every trade as buying a lottery ticket: you've decided whether the upside is worth the cost, and you've fixed the cost before you check the result. Position sizing is fixing the cost. You're not deciding how much to "win", you're deciding how much you're willing to lose to find out.

Trading is a business of taking lots of bounded losses to occasionally collect unbounded wins. Position sizing is what makes each loss bounded. Without it, "lots of trades" eventually concentrate the losses you didn't bound, and you go broke not from one bad trade but from the cumulative inattention.

Why this matters for trading

Hex37's order form embeds a position sizer (components/position- sizer.tsx, math in lib/risk-math.ts) that takes account size, risk percent, entry, and stop, and back-calculates the right position size and leverage. Use it. Get into the habit of clicking through the sizer on every trade rather than guessing the position size. After a few hundred trades, the math becomes intuitive, but the sizer keeps you honest in the meantime.

Takeaway

Position size is a calculation, not a feeling: account × risk% / stop distance%. Default risk per trade = 1%. Decide stop placement from the chart, then derive position size, never the reverse. Recalculate on equity changes, automatically deleveraging in drawdowns. Verify leverage doesn't put liquidation closer than your stop. Done consistently, this single discipline outperforms almost all the entry-signal optimization most traders obsess over.