OPEN-SOURCE SCRIPT
Aktualisiert EWCore
## What this indicator is — and what it isn't
EWCore is an automated Elliott Wave counting and scoring indicator for XAUUSD, built following the Frost/Prechter conventions. Before you use it, two things matter for an honest understanding of what you're looking at:
**1. TradingView imposes a hard technical ceiling on what can be drawn.**
Pine Script allows a single script a maximum of **500 simultaneously visible label and line objects** (default without explicit configuration: only about 50; raisable via `max_labels_count`/`max_lines_count` up to 500 — this is the platform's own hard ceiling, not a limitation of this script). Once that limit is reached, TradingView automatically removes the oldest objects, regardless of how important they are to the analysis.
EWCore already uses this budget in full. Fully labeling multiple Elliott degrees across an entire chart's history — the way a human analyst might on paper or in a static chart — is therefore **technically not possible**. On longer, more branched histories (chains exceeding 600 segments for the historical chain alone have been measured in this project), the available budget falls far short of what would be needed.
**This is why EWCore is not a tool that labels the entire chart continuously in the classic sense.** It is an **aid for recognizing and evaluating individual patterns** — it shows you the currently leading, most recently completed structure in full detail, including score, rule validation, and target zones, plus as much historical context as the label budget allows, prioritized by recency and relevance.
**2. Elliott Wave theory is a theory — not a fixed, standardized interpretation.**
Anyone with hands-on experience in Elliott Wave analysis knows: two experienced analysts labeling the same price stretch from point A to point B will often arrive at different counts. That's not necessarily a mistake — it's a matter of interpretation, and that interpretation depends heavily on context.
Markets behave differently because market-participant psychology differs. Phases of economic and geopolitical confidence produce different structures than phases of uncertainty. Gold, for instance, reacts with above-average volatility, which means certain patterns — complex W-X-Y corrections, for example — appear noticeably more often here than in calmer markets, where they tend to be the exception.
**A theory remains a theory**, however simple that sentence sounds. It can weight and rule-score patterns, but it cannot deliver a final, universally valid truth about a market's future. The score EWCore assigns to each count is a probability weighting based on defined criteria — not a certainty. That weighting is an approximation, not a law of nature.
## On the role of artificial intelligence in building this
This indicator wasn't hand-coded by me — it was built together with Claude, in this final version with Claude Fable 5. That process was instructive, including about the limits of AI models: simpler models failed at the task outright. But even a more capable model like Fable 5 runs into a fundamental limit — not one of computing power, but of the fact that the actual task at hand doesn't have a single, unambiguously correct solution.
Elliott Wave counting depends on pattern recognition, on intuition, and on contextual judgment that doesn't fully translate into parameters and weightings. What a human perceives as "ambiguous" — that feeling of "it could be this, but it could also be that" — an AI cannot grasp in the same way. It decides strictly according to fixed parameters and weightings. That weighting is a stipulation, not a feeling — and it is therefore not automatically the truth, however much one might want to trust it as such.
That doesn't make the AI's work worthless. Quite the opposite: it's a genuine time-saver, a structured second set of eyes that checks consistently against the same rules without ever tiring. But it does not replace the trained, context-sensitive judgment of an experienced analyst. **EWCore is meant as exactly that: an aid, not a verdict.**
## Full user manual
All 260 settings in this indicator are individually documented — with an explanation, default value, and a practical note on how it affects the count:
→ [EWCore User Manual] claude.ai/public/artifacts/6e2fe838-54ab-4736-95b5-b0c169e3c11a
EWCore is an automated Elliott Wave counting and scoring indicator for XAUUSD, built following the Frost/Prechter conventions. Before you use it, two things matter for an honest understanding of what you're looking at:
**1. TradingView imposes a hard technical ceiling on what can be drawn.**
Pine Script allows a single script a maximum of **500 simultaneously visible label and line objects** (default without explicit configuration: only about 50; raisable via `max_labels_count`/`max_lines_count` up to 500 — this is the platform's own hard ceiling, not a limitation of this script). Once that limit is reached, TradingView automatically removes the oldest objects, regardless of how important they are to the analysis.
EWCore already uses this budget in full. Fully labeling multiple Elliott degrees across an entire chart's history — the way a human analyst might on paper or in a static chart — is therefore **technically not possible**. On longer, more branched histories (chains exceeding 600 segments for the historical chain alone have been measured in this project), the available budget falls far short of what would be needed.
**This is why EWCore is not a tool that labels the entire chart continuously in the classic sense.** It is an **aid for recognizing and evaluating individual patterns** — it shows you the currently leading, most recently completed structure in full detail, including score, rule validation, and target zones, plus as much historical context as the label budget allows, prioritized by recency and relevance.
**2. Elliott Wave theory is a theory — not a fixed, standardized interpretation.**
Anyone with hands-on experience in Elliott Wave analysis knows: two experienced analysts labeling the same price stretch from point A to point B will often arrive at different counts. That's not necessarily a mistake — it's a matter of interpretation, and that interpretation depends heavily on context.
Markets behave differently because market-participant psychology differs. Phases of economic and geopolitical confidence produce different structures than phases of uncertainty. Gold, for instance, reacts with above-average volatility, which means certain patterns — complex W-X-Y corrections, for example — appear noticeably more often here than in calmer markets, where they tend to be the exception.
**A theory remains a theory**, however simple that sentence sounds. It can weight and rule-score patterns, but it cannot deliver a final, universally valid truth about a market's future. The score EWCore assigns to each count is a probability weighting based on defined criteria — not a certainty. That weighting is an approximation, not a law of nature.
## On the role of artificial intelligence in building this
This indicator wasn't hand-coded by me — it was built together with Claude, in this final version with Claude Fable 5. That process was instructive, including about the limits of AI models: simpler models failed at the task outright. But even a more capable model like Fable 5 runs into a fundamental limit — not one of computing power, but of the fact that the actual task at hand doesn't have a single, unambiguously correct solution.
Elliott Wave counting depends on pattern recognition, on intuition, and on contextual judgment that doesn't fully translate into parameters and weightings. What a human perceives as "ambiguous" — that feeling of "it could be this, but it could also be that" — an AI cannot grasp in the same way. It decides strictly according to fixed parameters and weightings. That weighting is a stipulation, not a feeling — and it is therefore not automatically the truth, however much one might want to trust it as such.
That doesn't make the AI's work worthless. Quite the opposite: it's a genuine time-saver, a structured second set of eyes that checks consistently against the same rules without ever tiring. But it does not replace the trained, context-sensitive judgment of an experienced analyst. **EWCore is meant as exactly that: an aid, not a verdict.**
## Full user manual
All 260 settings in this indicator are individually documented — with an explanation, default value, and a practical note on how it affects the count:
→ [EWCore User Manual] claude.ai/public/artifacts/6e2fe838-54ab-4736-95b5-b0c169e3c11a
Versionshinweise
Small bug fixAdded feature list: claude.ai/public/artifacts/778ca0dd-a450-4eee-85d6-5695e67aeeb9
Edited User Manual (Settings Menu Options): https://claude.ai/public/artifacts/c04f7b8f-e01c-4577-8f50-5691f293feb4
Versionshinweise
EWCore v0.2.9.1 — What's new- Extended waves are marked. When a wave is built from 9, 13 or 17 smaller waves instead of 5, a small number now appears on its label.
- Cleaner counts. Zigzag corrections and wedge-shaped moves are recognised more strictly, so fewer legs get the wrong label.
- New Anchor column in the score table shows whether a count starts at a real turning point. That is why a lower-scoring count can rank above a higher one.
- Projections follow price. If price runs past a target, the projection now stretches instead of being left behind.
- One setting moved — the compact status table toggle now sits with the other table settings, and returns to its default once.
On large timeframes: the biggest wave counts need a lot of history. If your data only goes back a few years, there may be just one possible reading rather than the best one. Hover Pivots/lvl to see what the indicator has to work with. Below M15, counting is not reliable.
Edited User Manual v0.2.9.1 (Settings Menu Options):
https://claude.ai/public/artifacts/bc0b689c-d0cf-4339-ba0d-91fa91be9005
Edited feature list v0.2.9.1:
https://claude.ai/public/artifacts/02610669-327b-4e27-b8e4-99f7800a3a0a
Open-source Skript
Ganz im Sinne von TradingView hat dieser Autor sein/ihr Script als Open-Source veröffentlicht. Auf diese Weise können nun auch andere Trader das Script rezensieren und die Funktionalität überprüfen. Vielen Dank an den Autor! Sie können das Script kostenlos verwenden, aber eine Wiederveröffentlichung des Codes unterliegt unseren Hausregeln.
Haftungsausschluss
Die Informationen und Veröffentlichungen sind nicht als Finanz-, Anlage-, Handels- oder andere Arten von Ratschlägen oder Empfehlungen gedacht, die von TradingView bereitgestellt oder gebilligt werden, und stellen diese nicht dar. Lesen Sie mehr in den Nutzungsbedingungen.
Open-source Skript
Ganz im Sinne von TradingView hat dieser Autor sein/ihr Script als Open-Source veröffentlicht. Auf diese Weise können nun auch andere Trader das Script rezensieren und die Funktionalität überprüfen. Vielen Dank an den Autor! Sie können das Script kostenlos verwenden, aber eine Wiederveröffentlichung des Codes unterliegt unseren Hausregeln.
Haftungsausschluss
Die Informationen und Veröffentlichungen sind nicht als Finanz-, Anlage-, Handels- oder andere Arten von Ratschlägen oder Empfehlungen gedacht, die von TradingView bereitgestellt oder gebilligt werden, und stellen diese nicht dar. Lesen Sie mehr in den Nutzungsbedingungen.