Position sizing, worked backwards from the stop

Size equals the risk budget divided by the distance from entry to stop. Both sides have to be in the same units. If the risk budget is fifty dollars and the stop is two thousand dollars away in price terms, the size is fifty divided by two thousand, which is nought point zero two five units. That is the entire calculation, and everything else in this guide is about choosing the two inputs honestly.

An account of five thousand dollars risking one percent per trade has a budget of fifty dollars. BTC is at one hundred thousand and the level that invalidates the idea is ninety eight thousand, so the stop distance is two thousand, or two percent. Size is nought point zero two five BTC, which is two thousand five hundred dollars of notional. At ten times leverage that needs two hundred and fifty dollars of margin. Notice that leverage never entered the decision: it is a consequence of the size and the margin you chose to post behind it.

Run the same trade with a stop five hundred dollars away instead of two thousand. The budget is unchanged at fifty dollars, so the size becomes nought point one BTC, which is ten thousand dollars of notional against a five thousand dollar account. The risk is identical on paper and materially worse in practice, because a position four times larger pays four times the fees, four times the slippage, and is four times more sensitive to the mark price wobbling through your stop. Tight stops are not free risk reduction.

Between half a percent and two percent of account equity per trade is the band most systematic approaches sit in. The specific number matters less than fixing it in advance and applying it consistently, because the property you want is that a normal losing streak is survivable. Ten consecutive losses at one percent is a drawdown of about nine point six percent, which is uncomfortable and recoverable. Ten at five percent is about forty percent, which needs a sixty seven percent gain to undo.

The clean formula assumes the only cost is the price move, which is never true.

• Fees on entry and exit, plus expected slippage, come out of the same budget
• Correlated positions share one budget: three long alt positions are one trade wearing three hats
• Funding accrues on multi-day holds and is charged on notional, not on margin
• The stop must sit inside the liquidation price, or the venue closes you first and the plan was fiction
• Size against current equity, not against the amount you originally deposited

Some instruments have no stop to place. A binary outcome market pays a fixed amount or nothing, so there is no level at which you exit early with a partial loss. In that case the stake is the risk, and the sizing question collapses to how much of the account you are willing to lose in full on one resolution. The same discipline applies, it is just that the distance to zero is one hundred percent and there is nothing to divide by.