How Long Does It Take to Reach the Peak
— 114K Measured
"Should I have held longer?" isn't a question a single trade can answer. We pooled a matured sample of 114,454 rows to publish the average peak (MFE) and trough (MAE) within a 20-trading-day window after entry, plus how often a single window isn't enough.
The Population Baseline — an Average of 114K Rows
As of 2026-08-26 · matured sample 114,454 rows (with mfe_20d/mae_20d) · source: universe_snapshot
| Metric | Average | Definition |
|---|---|---|
| MFE (peak) | +14.58% | Highest return reached within 20 trading days after entry |
| MAE (trough) | −9.85% | Deepest decline experienced within the same window |
These two numbers aren't from a single signal or a single stock — they're the average across the entire scanner universe. Any trading-journal tool can already tell you "that stock ran up to 19%." What you can't get without a population baseline is whether that 19% is above or below the average of 114,000 measured rows — that's what this number is for.
How Long It Takes — the Limits of a 20-Day Window
Measured 2026-08-13 · 249 matured picks · source: scan_pick + price_history recompute
We haven't separately computed the exact day-by-day distribution of "how many days until the peak" (e.g. what share peaks within 5 days, within 10 days) as of writing this — we don't invent numbers before the sample is in place.
What we do have: when we recomputed the 249 picks our scanner actually selected over a 20-trading-day window, 42.6% didn't reach their peak until after day 20. Put differently, judging "did it reach its peak" with a single 20-day window leaves close to half the cases without an answer yet when the window closes.
Why We Split the Sample
This article draws on two different samples. The population average (MFE/MAE) uses universe_snapshot, where the matured sample is 114,454 rows — a sample this large is what lets us answer "how much does it move on average" with any stability. The picks the scanner actually selected (scan_pick), on the other hand, are still only 540 rows — too few, in our judgment, for pick-level conclusions (e.g. "picks with this signal peak in an average of N days"), so we don't draw those conclusions in this article.
How to Actually Use This
- Don't call "no peak yet" a failure too quickly — a large share of the sample peaks after day 20. Before concluding "it didn't work this time," suspect the observation window itself.
- Check the baseline whenever you see "peak return +N%" — whether that N% is above or below the 114K-row average (+14.58%) is the number that actually matters.
- MFE doesn't tell you when to exit — reaching a peak and being able to sell at that price are two different facts. The next article, Was the Exit Early, or the Entry Bad?, covers how to tell the two apart.