Sample Size Is the Number Everyone Ignores
A 68% win rate means very different things at 12 occurrences and at 200.
Win rate gets all the attention because it's the headline number. Sample size — how many times a setup actually fired and closed — gets glanced at, if it's shown at all. That's backwards. A win rate without its sample size isn't really a statistic yet; it's closer to an anecdote with a percent sign on it.
Why small samples lie convincingly
Flip a fair coin 10 times and getting 7 heads isn't unusual — it happens roughly 1 in 6 times purely by chance. Flip it 200 times and 70% heads would be extraordinary. The same math applies to a trading setup: a handful of favorable outcomes can look like a strong edge even when the underlying win probability is close to a coin flip. The smaller the sample, the more a run of luck looks like a real signal.
What ChartWyz does about it
- Every setup card shows occurrence count (n) directly next to win rate — never one without the other.
- The drill-down view flags a low-sample warning when a setup's closed-trade count is thin for the selected coin, timeframe, and lookback.
- Multi-timeframe grids let you check whether a setup's edge holds up across several timeframes and coins, not just the one that happened to produce a flattering number.
None of this guarantees a setup with a large sample and a strong win rate will keep performing the same way going forward — markets change. What it does is stop you from mistaking noise for an edge in the first place, which is the more common mistake.