Wave-count success rate: six cases and two denominators
An invented outcome ledger shows why 2 out of 4 and 2 out of 6 answer different questions and establish no trading profitability.
Published: (UTC+3)
A question about Elliott wave success rates first needs a defined outcome. Structural eligibility, reaching a level and making a profitable trade are different results. This entirely invented sample teaches record keeping; it measures neither Elliott Wave accuracy nor the performance of our software.
Define the event before inspecting outcomes
Each case starts after its own frozen data cutoff. Observe the next 10 completed bars. The event is reaching 112 before first reaching the invalidation level 96. All six cases are included in advance under one selection rule. Event order is supplied in this exercise; intervening bars are not shown. If the available evidence cannot establish the order, keep the outcome unresolved. These levels are exercise conditions, not trading instructions.
Retain all six observations
Each letter identifies a separate observation selected in advance. Bar numbers run from that case's original record, so bar 3 in two cases need not share a calendar date. D has completed its horizon; F is still being observed. Preserve that distinction even though neither has yet produced the event.
| Case | Known sequence | Status | Accounting |
|---|---|---|---|
| A | 112 at bar 3; 96 at bar 8 | Event first | Numerator and resolved denominator |
| B | 112 at bar 8; no 96 through bar 10 | Event first | Numerator and resolved denominator |
| C | 96 at bar 2; 112 at bar 6 | Invalidation first | Resolved denominator |
| D | Neither level by the end of bar 10 | Horizon complete without event | Resolved denominator |
| E | Bars 4–6 missing; order unknown | Unresolved | Retain in original cohort |
| F | Only 7 bars available, without event | Observation continuing | Retain in original cohort |
Expose the denominator and unresolved cases
Among the four resolved cases A–D, two reached the event first: 2 / 4 = 50%. In the original six-case cohort, two such outcomes are known: 2 / 6 ≈ 33.3%. The latter describes events observed so far in the full cohort; it must not silently turn E and F into failures. Reporting '2 events, 2 other resolved outcomes, 2 unresolved' retains useful information. Excluding unresolved cases also needs justification because it can change the composition of the evaluated sample.
Review what new information changes
A later review must preserve the original record. Add a review date and an updated status beside it. Changing the event, horizon or selection procedure creates another protocol version. Do not choose a definition of success after seeing which percentage looks attractive.
F later reaches 112 at bar 9 without an earlier 96. E remains unresolved. Which fractions can now be reported?
Five cases are now resolved and three reached the event first: 3 / 5 = 60%. Three events are known in the original cohort of six: 3 / 6 = 50%, with one unresolved case. This updates observations under the original rule. It does not make 60% a measured real-world accuracy for Elliott Wave.
Use the protocol in your own study
Save the data, range, selected count and accounting rules first. A trading result additionally needs entries, exits, position sizing and costs: reaching 112 alone does not establish a profitable trade. Six invented observations provide no estimated probability, confidence interval or evidence of effectiveness. Follow the first-calculation guide to work with fixed source data; this article introduces no automatic strategy-testing feature.
First calculation → · Compare Pro and Light → · Request Pro →
Six cases: which observations enter the denominator
The diagram uses the existing fictional cohort A–F. A/B reach 112 before 96; C reaches 96 first; D completes the 10-bar horizon without the event. E has missing bars 4–6 and unknown order; F has only 7 bars so far without the event. Two events among four resolved cases give 2/4=50%. Two events are known in the original cohort: 2/6≈33.3%. E/F do not become losses.
Later F reaches 112 at bar 9 without an earlier 96, while E remains unresolved. The figures become 3/5=60% among resolved cases and 3/6=50% known events in the original cohort. Retain the update date and initial record. These fractions describe invented events under the chosen rule. They measure neither Elliott accuracy, the next outcome’s probability nor profit: trading results additionally require entries, exits, quantity and costs.
| Item | Calculation or value |
|---|---|
| A,B | 2 |
| C,D | 2 |
| E,F | 2 |
| n0 | 2/4=50%; 2/6≈33.3% |
| n1 | 3/5=60%; 3/6=50% |
Why can the four cases outside the numerator of 2/6 not all be colored as losses?
C/D have known but different non-event outcomes, E is unresolved, and F is still being observed. Even “target reached first” does not by itself establish a trading profit.
Sources and limits
Definitions, cases and arithmetic are original teaching material. The existing limitations guide explains requirements for probability claims; this article adds a worked accounting example. The scenario journal preserves the initial hypothesis; the benchmark article addresses a separate comparison of results. No new market observations or product statistics are reported.