rangeProb-Tracks

Your reward-to-risk is lying to you.

A 2:1 looks like it needs a 33% win rate, because that is what the arithmetic says. Measured on the tape it pays about 29%. The gap is not small, it is not random, and it runs the same way in all six markets: the tighter your stop, the more the maths flatters you.

Below you can price your own geometry against what it actually did, resolved bar by bar on whichever came first. The entry is blind and unconditional on purpose: it is the control, not a strategy. It is the bar your entry has to clear before it is worth taking.

Distances are shown in points for the market you pick, sized off its typical daily range. That is why the same setting reads as 130 points on NQ and under a point on CL, because it keeps “a normal-sized stop” meaning the same thing everywhere, and in every year, however loud the tape has been lately.

Set the trade

—

Target in points: how far you are trying to make

Stop in points: how much you will give it

What the arithmetic gets wrong

Every cell of the grid, priced against stop / (target + stop), the number a fair game would give you. 384 independent settings, six markets, roughly a thousand sessions each.

Reward to risk	Actually	Arithmetic	Gap
Under 0.75	71.5%	69.6%	+1.9	7.8 sd
About even	48.2%	49.1%	-0.9	4.4 sd
1.4 to 2.2	33.3%	36.5%	-3.2	5.7 sd
Over 2.2	18.7%	22.7%	-4.0	8.5 sd

Tight stops, every market: CL-6.4GC-6.0NQ-3.6ES-3.4RTY-2.8YM-2.1

Six markets, same sign, no exceptions, and it grows steadily as the stop tightens. Wide stops with modest targets come in ahead of fair value; tight stops with ambitious targets come in behind it. Noise does not line up like that.

Why it is not an edge. This says nothing about direction, and it will not make a losing entry profitable. It says the price of a tight stop is higher than the arithmetic admits, so a setup that looks break-even on paper is usually not. We tested whether any of our 19 setups shifts this: across 568 combinations, 1% cleared two standard deviations, which is below what a search that size finds by chance. Conditioning does not move it, and we would rather say so than sell you a filter.

What this does and does not tell you

It resolves the race honestly. Every level carries the bar it was first touched on, so when both your target and your stop print on the same session we know which one got there first. That is a real path resolution, not an assumption about the order.

The entry is blind and unconditional: long or short from the session open, every session, no filter. That is deliberate: it isolates the geometry. If you want a reason to be in the trade at all, that is what SPOTS and Tracks AI are for. Combine them.

Days where neither level printed are counted and shown, but they are not in the win rate. In reality you would close those somewhere in between, and we are not going to pretend to know where. Treat a big grey block as a warning that the sample answering your question is smaller than it looks.

Path data covers part of the history, not all of it, so this carries a smaller sample than the rest of the site. The number of sessions is stated on every result.