size the risk first, then let leverage fall out of the math

position size starts with the amount your account can lose if the stop fills, not with the leverage slider. The core calculation is contracts = risk budget ÷ distance from entry to stop. For a long, that distance is entry minus stop; for a short, it is stop minus entry. Multiplying contracts by entry gives notional, and notional divided by assigned margin gives the leverage that position requires.

take a $10,000 account and allocate 1% to the planned loss: $10,000 × 1% = $100. If the position has a $1,000 isolated-margin sleeve, leverage becomes notional ÷ $1,000. The risk budget remains $100 whether the resulting position needs 2.86x, 6.25x or 16.67x. Leverage controls margin efficiency; the stop distance and contracts control the planned loss.

1. set the dollar risk budget before reading the leverage slider.
2. place the stop from the trade structure, then measure its exact dollar distance from entry.
3. divide the risk budget by that distance to get contracts.
4. multiply contracts by entry to get notional.
5. divide notional by assigned margin, then check the resulting bracket, MMR and estimated liq.

the worked longs use the marks at authoring as entries, not live quotes: BTC at $97,400, ETH at $3,620 and SOL at $214. Each position risks the same $100, but the stop distances differ: BTC uses $584.40 or 0.6%, ETH uses $57.92 or 1.6%, and SOL uses $7.49 or 3.5%. Those distances come from the supplied Binance worked examples and produce different contract counts.

with a $1,000 isolated-margin sleeve, BTC produces the largest notional because its stop is the tightest in percentage terms. The arithmetic is $100 ÷ $584.40 = 0.171116 BTC, then 0.171116 × $97,400 = $16,666.67. ETH produces $6,250 notional, while SOL produces $2,857.14. The risk remains $100 in each row before fees, funding, slippage or a stop gap.

yes, in these isolated-margin examples the stop sits materially before the estimated liq. Using Binance first-bracket MMR, the BTC long at 16.67x has an estimated liq of $91,945.60, while its stop is $96,815.60. That leaves $4,870 between stop and liq. ETH at 6.25x has a $3,055.28 estimated liq and a $3,562.08 stop, leaving $506.80 between them.

SOL at 2.86x has an estimated liq of $140.17 against a $206.51 stop, a $66.34 gap. The approximation is long liq = entry × (1 − 1/leverage + MMR). These figures are an estimate — exchanges compute exact liq from live margin; always double-check on your venue. Cross-margin PnL, fee and funding accruals, stop slippage and bracket changes can move the live result.

MMR is selected from notional, so sizing and liquidation cannot be separated. On Binance, BTC and ETH stay at 0.4% MMR through $300,000 notional, while SOL stays at 0.5% only through $50,000. Bybit starts BTC and ETH at 0.33% through $300,000, while SOL starts at 0.5% through $50,000. Crossing a cap can change both MMR and available maximum leverage.

the three $100-risk examples remain in their first brackets because their notionals are $16,666.67, $6,250 and $2,857.14. Larger risk budgets or tighter stops can push notional into another tier even when the account-level risk percentage is unchanged. Only the first 5 tiers are available in the supplied calculator table; beyond them, its fallback can slightly underestimate MMR, so large notionals need a venue-side check.

fixed percentage risk reduces the next dollar loss as equity contracts. Starting from $10,000 and losing 1% of remaining equity each time gives $10,000 × 0.99 after each stopped position. After 5 losses, the balance is $9,509.90; after 10 it is $9,043.82; after 20 it is $8,179.07. That model excludes fees, funding, slippage and stop gaps.

fixed leverage does not define risk unless the stop distance and capital allocation are also fixed. For contrast, applying 20x to the full remaining account with a 1% adverse stop implies a 20% equity loss per stopped position before costs: 20 × 1% = 20%. After 5 such losses, $10,000 × 0.80⁵ leaves $3,276.80. The comparison isolates sizing arithmetic rather than forecasting any sequence.

a stop-based $100 loss is not the full cash outcome when execution costs are included. Binance’s supplied 0.05% taker rate gives a 0.1% taker round trip, while Bybit’s 0.055% gives 0.11%. On $10,000 notional, that is $10 on Binance or $11 on Bybit before funding. Those amounts equal 10% and 11% of a $100 risk budget, respectively.

at 20x, the supplied round-trip taker drag equals 2% of margin on Binance and 2.2% on Bybit. Both venues use an 8-hour funding interval in the facts pack, but the amount paid or received depends on size, rate and holding time. A sizing sheet therefore needs separate lines for planned price loss, estimated execution fees and funding rather than burying all three inside leverage.

the sizing chain should reconcile risk budget, stop distance, contracts, notional, assigned margin, leverage, bracket and estimated liq in that order. If changing leverage changes your planned dollar loss, the position was sized from margin rather than from the stop. If changing the stop leaves contracts unchanged, the risk budget is no longer fixed. Both breaks are visible before a taker fill reaches the book.

trigger type for stop, take-profit and conditional orders is selectable and varies by product and interface, so it must be checked on the account rather than assumed. The mark also matters to liquidation monitoring, while a stop can face slippage or a gap through its trigger. The venue’s live margin display remains the final check because cross collateral, accrued funding, fees and bracket transitions are outside the isolated approximation.

• confirm that contracts equal dollar risk divided by the exact entry-to-stop distance.
• recalculate notional after any entry change or partial fill.
• check the live MMR bracket after the final notional is known.
• verify the selected conditional-order trigger type on the account.
• compare the venue’s live liq with the isolated estimate before relying on the stop-to-liq gap.