Chart pattern success rates: audit the full record

Audit pattern success rates, measured targets, re-entry losses, zero-event samples and unresolved outcomes with original calculations and diagrams.

Course outline

1. 1

   Audit the denominator behind a chart-pattern success rate

   One hundred candidates, sixty entries, twenty-four wins. Separate candidates, qualified entries, target touches and profitable trades before quoting a rate.

   7 min

   In short: A pattern success rate needs a stated event and denominator: candidates, qualified entries, target touches or target-before-stop wins. With 100 candidates, 60 entries and 24 wins, the eventual-touch rate among entries is 60% but the win rate is 40%, and the net result is 6R after 0.1R cost per trade.

2. 2

   Calculate a flag target from explicit reference points

   Half the prior move can give a target of 128 or 135. Separate price-difference and percentage projections, then measure reward and risk from the actual entry.

   6 min

   In short: Half the prior move can mean half the price difference or half the percentage change. For an 80 to 120 move projected from 108, the absolute rule gives 108 + 20 = 128 and the percentage rule 108 × 1.25 = 135. Measure reward and risk from the executable entry.

3. 3

   Reconcile a stopped trade and re-entry as one complete episode

   A 29.47 USDT winner can leave an episode profit of only 8.98 USDT. Count the first stopped trade, every fee and the remaining risk budget before re-entering.

   7 min

   In short: A re-entry result includes the first stopped trade and every fee: a 20.49 USDT net loss plus a 29.47 USDT net win leaves 8.98 USDT. Under a fixed episode loss budget, size the next attempt from the remaining budget and the new stop distance.

4. 4

   Zero observed failures do not mean zero risk

   Twenty selected examples without a failure prove little. Calculate what a sample with zero observed failures can establish and check how it was selected.

   7 min

   In short: Zero observed failures in n trials is an observed frequency of 0, not zero risk. Under a binomial model with independent, fully observed trials, the one-sided upper 95% bound is 1 − 0.05^(1/n): about 13.91% for n = 20 and 0.994% for n = 300.

5. 5

   Keep unresolved chart-pattern outcomes in the evaluation

   Separate completed and still-open observations in a pattern evaluation. Calculate transparent bounds without pretending open price paths are wins or losses.

   6 min

   In short: Keep unresolved observations visible instead of dropping them or counting them as wins or losses. With 6 successes, 1 stop and 3 unresolved cases, the completed-case rate is 6 / 7 ≈ 85.71%, while the final rate for all ten can only be bounded between 60% and 90%.

Open full-size diagram

1. Audit the denominator behind a chart-pattern success rate
2. Calculate a flag target from explicit reference points
3. Reconcile a stopped trade and re-entry as one complete episode
4. Zero observed failures do not mean zero risk
5. Keep unresolved chart-pattern outcomes in the evaluation

Educational material. Examples do not establish a profitable strategy. Trading costs, gaps and liquidation can produce losses beyond a planned stop.