> For the complete documentation index, see [llms.txt](https://help.hydravision.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.hydravision.ai/products/asset-profiler/recommended-step.md). # Recommended Step A cyan-bordered panel headed **"Recommended step"**, which appears on some reports and not others. {% hint style="warning" %} **The common misreading.** That its absence means something failed. On most assets there is no single best step, and saying so is the honest answer rather than a gap in the data. {% endhint %} ### Why most reports don't have one Widening your step lowers the share of the target that costs consume: the fixed cost at entry gets spread across a bigger target. That relationship never reverses on its own. So on most assets the cheapest step is always the widest one you could test, purely as arithmetic. Calling that a recommendation would be dishonest. The widest step in your range isn't the best step. It's just the last one on the chart. When this is the situation, the report says so plainly, in a grey note where the recommendation panel would otherwise be: > **The economics do not pick a single step inside this band.** \[...] That is arithmetic, not a measurement, and the widest step is not a recommendation. **Use the band.** Wider costs proportionally less; tighter fills more often. ### What changes on some assets Overnight cost. If holding is expensive enough, widening the step stops being free. A wider step earns a bigger target, but it also takes longer for price to retrace far enough to close. Longer waits mean more nights held, and more swap paid. On assets where swap is heavy, there comes a point where the extra swap outweighs the bigger target, and past that point, wider is worse. That turning point is a genuine measured minimum, and where it exists, the report names it. ### One example

XAUUSD Recommended Step

XAUUSD, 5 July to 28 October 2022: ``` Recommended step · 28.36 ``` Gold's overnight cost on the long side is $58.84 per lot per day. That's heavy enough to bend the curve. Costs bottom out at **3.60%** of the target at a step of 28.36, against 4.69% at the floor and 4.38% at the top of the range. The curve genuinely turns. The recommended step is shown in the same unit as the large step range, so 28.36 here is $28.36 of gold. In the unit `Minimum Step Size` uses on gold (one cent) that is 2836. See [Ticks, pips and price](/products/asset-profiler/ticks-pips-and-price.md). The chart marks the point with a larger dot and a dashed line down to the axis, and the sweep table highlights the row. ### Carry pressure What decides it is the cost curve itself: a step is recommended only when the cheapest point in the sweep sits **inside** the range rather than at its wide end. The number that predicts this appears in the **Cost of holding** panel as **carry pressure**: the wait at the wide end of the range, divided by the number of days of overnight cost that equal the entry cost. * **Below 1** — what you pay at entry dominates. The curve usually falls all the way, and there's no recommendation. * **Above 1** — overnight cost usually takes over by the wide end of the range, and a real minimum appears. Pressure is a predictor, not the test. It has matched the outcome on every asset checked so far, but the recommendation appears only when the measured curve really turns. {% hint style="info" %} **A recommendation is still not an instruction.** Where one appears, it marks the cheapest step *on this asset's cost curve*. It doesn't know how often you want positions to fill, how deep your grids run, or which side of the market you trade. On gold, note that the entire calculation assumes you are on the long side, the side that pays the swap. If you run short, gold's carry is positive and the trade-off doesn't apply to you at all. See [Carry](/products/asset-profiler/carry.md). {% endhint %} ### What to do with this * No recommendation is the normal outcome. The range is the answer, and testing is what places a strategy inside it. * Where one appears, it marks the cheapest point on this asset's cost curve, not a setting to lock. * The recommendation holds only for the side of the market that pays the carry.