Copper — Record Close Under Intact Resistance (Event Study)
Re-derived on our data · yfinance HG=F weekly, 2000–present · 13 completed events · LIVE event · 0.8% off the ATH · Data through September 4, 2026 · Updated August 31, 2026 · 6:49 PM ET
Live event now?
YES
last fired 2026-07-31 · 5w ago
Current close
6.694
+0.8% to ATH 6.750 · rec 6.694
Near-term (2w)
-3.2%
23% win · the dip
Worst drawdown
-24%
within 26w · size for it
26-week base
+6.5% / +14.2%
median / mean · 62% win · n=13
Modern regime 26w
+5.1% / +2.3%
median / mean · n=4 (thin)
● LIVE EVENT — copper closed 6.694, 0.8% below the 6.750 ATH; last record-close-under-resistance fired 2026-07-31 (5w ago). The modal post-event drawdown looks underway. Structural-dip entry zone ≈ 5.69–6.23; invalidation = weekly close < ~6.14.
Forward path after the event — completed strands, median, 25–75% band, and the live run
Each thin line is one completed event's 26-week path indexed to 100 (amber = 2003–11 supercycle, purple = 2024–26 modern). Black = median, shaded = interquartile band. The red dashed line is the current run since the last record-close-under-resistance event — it is tracking the modal early drawdown.
Base rate — forward returns after 13 completed events
HorizonnMedianMeanWin%p25 / p75Worst max DD
1w13-0.4%-1.2%46%-2.2 / +0.9-8.7%
2w13-3.2%-4.6%23%-4.8 / -0.2-20.9%
4w13-0.6%-2.7%46%-7.0 / +2.5-20.9%
8w13+0.3%-0.7%54%-10.5 / +9.0-20.9%
13w13+0.8%+5.8%54%-5.2 / +19.9-20.9%
26w13+6.5%+14.2%62%-1.3 / +23.6-24.0%
"Worst max DD" = the deepest close-to-close drawdown any single event saw within that horizon. The 2-8 week window is where the pain concentrates; the edge is a 26-week phenomenon.
Regime split — supercycle vs the modern tariff era
Cohort4w medwin13w medwin26w medwin
2003–11 supercycle n=9+0.9%56%+2.2%67%+15.0%56%
2024–26 modern (tariff) n=4-8.0%25%-5.2%25%+5.1%75%
The supercycle cohort resolved hard-up; the modern cohort takes deeper near-term pain and resolves roughly flat. Same direction, far thinner modern edge — size accordingly.
The verdict — the deck's direction holds on our data

Across 13 completed events since 2000 (record weekly close while the week's high stayed below the prior record intraday high, de-clustered to one per 8 weeks), the modal path is exactly what the briefing claimed: weakness first, resolution later. The 2-week median is -3.2% (23% win — the "sell-the-record" dip), the drawdown runs as deep as -24% inside the first six months, and only by 26 weeks does it clear to +6.5% median / +14.2% mean (62% win). Fading the breakout and buying the structural dip is the empirically-supported expression — not chasing the record close.

The honest correction — the modern regime is thinner than the deck implies

Split by regime, the character changes. The 2003–11 supercycle cohort (n=9) resolved powerfully: +15% median at 26 weeks. The 2024–26 modern cohort (n=4) is a different animal — deeper near-term pain (-8.0% at 4 weeks, 25% win at 13 weeks) and a flat six-month resolution: median +5.1% but mean +2.3%. The upside the deck leans on is real in the supercycle and close to a coin-flip in the tariff era. That argues for the deck's own risk discipline — small size, hedge the tail — even harder, because the modern edge barely clears zero.

Where we actually are — the dip may already be underway

On our data copper is not printing a live event this week. The last record-close-under-resistance fired 2026-07-31 (5 weeks ago) at ~6.44; price has since pulled back to 6.694, now 0.8% below the ATH (6.750) — squarely on the modal post-event drawdown (the red dashed "live run" on the chart). The deck's structural-dip entry zone, 7–15% below the record close, is roughly 5.69–6.23. The watch stays on WAIT: accumulate into that zone on LME confirmation, stand down on a weekly close below the ~6.14 shelf.

Guardrails & caveats
← Backs the copper WATCH on The Board · read the study deck · first principles