Lot size, contracts, actual risk and risk-to-reward from your balance and stop distance. Correct pip and point values for forex, gold, NQ and ES — not a one-size-fits-all guess.
Position size is arithmetic, not a view on the trade. It tells you what the market can take from you if you are wrong. See ICT position sizing for how this fits the models.
Position size is the only part of a trade you control completely. The entry might not fill, the target might not be reached, the setup might fail outright — but the amount at stake is decided before any of that, by arithmetic, and it is the same arithmetic every time.
The intraday ICT models are selective by design. They decline on consolidation days, they decline when the daily bias is unclear, and they decline when the window passes without displacement. The edge, if there is one, comes from a long sequence of qualified setups rather than from any individual trade.
A sequence only survives if no single loss is large enough to change how you trade the next one. That is the real argument for fixed fractional risk, and it is a behavioural argument rather than a mathematical one. At 1% risk a four-loss run costs about 4% and you will take the fifth setup normally. At 5% risk the same run costs about 19%, and almost nobody takes the fifth setup normally after that.
Generic calculators get gold and index futures wrong because they assume forex conventions apply everywhere. They do not:
| Instrument | Measured in | Value per unit |
|---|---|---|
| EUR/USD, GBP/USD | Pips (0.0001) | ≈ $10 per pip per standard lot |
| USD/JPY | Pips (0.01) | ≈ $9.1 per pip per standard lot, varies with rate |
| XAU/USD (gold) | Pips (0.10) | ≈ $10 per pip per standard lot |
| NQ futures | Points | $20 per point per contract |
| ES futures | Points | $50 per point per contract |
The calculator rounds down rather than up, so the figure it gives you always risks slightly less than your stated percentage rather than slightly more. On futures it rounds to whole contracts, which on a small account sometimes returns zero — that is a real answer, not a bug. It means the stop is too wide for the account at that risk level.
The risk-to-reward figure exists to tell you whether a setup is worth taking, and it should be read after the levels are decided rather than used to choose them. Where the stop belongs is a structural question: beyond the swing that invalidates the idea, with a buffer for the wick. Where the target belongs is also structural: the next pool of liquidity in the direction of the trade.
If those two facts produce a poor ratio, the conclusion is that the entry is late or the target is close, and the trade is probably not there. Tightening the stop to improve the number does not make the trade better, it makes it more likely to be stopped out by noise before the idea has a chance to work.