The Edge
Enter one set of assumptions and see the range of outcomes those same odds can produce. Some of them will be uncomfortable.
Define edge by
Win rate — 40%
How sure are you? — ±1%
Nothing in trading is certain — so 1% is the floor. ~50 trades ≈ ±10% · ~150 ≈ ±6% · ~500 ≈ ±3%
Loss size (R)
Average winner (R)
Trades taken
Show me
Expectancy per trade
I can only show you the distribution.
You're the one that has to trade through it.
One edge, every path
The same edge, run again and again. Every path is a valid outcome. None are broken. This is variance doing what variance does.
Drag across to zoom · double-tap or double-click to reset
Press Run again to re-roll the same edge. Notice the gap between the luckiest and unluckiest path — that's what almost nobody shows you.
DrawdownHow deep the losing stretches run, and when one is worth investigating.
How deep underwater
Measured from each new high — the drop, not the altitude. Zones mark when a drawdown is worth investigating.
How rough — 90th-percentile path
75th = fairly rough · 99th = near-worst but still valid
Drag across to zoom · double-tap or double-click to reset
median path
rough path
green = typical · amber = rare · red = investigate
Red does not mean the edge is broken. It means investigate. The threshold is set at this edge's 99th-percentile drawdown, so a working edge can still reach or exceed it in roughly 1 run in 100.
One caveat: this assumes the numbers you entered are true. The most common way an edge "breaks" isn't a drawdown at all — it's that it was never as strong as assumed. Only a real, logged record separates the two.
How strong is the implied expectancy?Break-even is only the floor. The next question is how much evidence an edge of this apparent strength would need.
How strong is the implied expectancy?
Win rate and reward-to-risk determine the expectancy implied by your inputs. The dot shows that calculated expectancy. The shaded bands are my own working interpretation of how much evidence I would want before trusting an edge of that apparent strength. They are not universal limits.
break-even line
your edge (from your inputs above)
Plausible range
0 – 0.3R
needs a large, logged sample
Unusually strong
0.3 – 0.6R
needs a large, logged sample
Exceptional claim
0.6 – 1.0R
short samples may mislead
Extraordinary claim
above 1.0R
requires extraordinary evidence
The expectancy and break-even calculations are mathematical results based on the numbers you enter. The placement and names of the shaded bands are my own working interpretation. They are not statistically derived boundaries, industry standards or judgments about any particular trader, strategy or product. Their purpose is to show that stronger claims should require stronger evidence. Want to test a claim you've seen? Use the Pitch Check tool.