
Chart patterns, COT, seasonality, ratio trading on metals. A complete 2026 overview of where the market sits and what to watch next.
Contents9 sections
- 01Chart Patterns: The Discipline of Confirmation
- 02Pattern failures and the risk of confirmation bias
- 03COT Positioning: The Most Underused Public Dataset
- 04Seasonality: Real, But Limited
- 05Intermarket Analysis: The Macro Lens
- 06TIPS, breakevens, and the gold price
- 07Ratio Spreads: The Relative Value Lens
- 08Synthesis: Combining the Five Inputs
- 09Read next
Technical Analysis 2026 Investor Overview
Technical analysis in precious metals is most useful when held to a high evidential bar and least useful when treated as a self-contained worldview. This pillar overview covers the five sub-clusters that constitute a defensible technical framework for gold, silver, and PGM markets in 2026: chart patterns, COT positioning, seasonality, intermarket analysis, and ratio spreads. Practitioners who combine these inputs with a clear-eyed view of fundamentals, monetary policy, and physical flows tend to outperform those who treat technical analysis as either gospel or noise.
The value proposition of technical analysis is not prediction; it is risk management. Pattern recognition, positioning data, seasonality, and intermarket relationships provide reference points for sizing, timing, and stop placement. A technical signal that aligns with a fundamental thesis is more actionable than either input alone. A technical signal that contradicts the fundamental thesis is a prompt for review, not a prompt for reversal.

Chart Patterns: The Discipline of Confirmation
The chart pattern literature applied to precious metals draws from Edwards, Magee, and the broader classical technical canon, with adaptations for the volatility profile of gold and silver. The patterns that have demonstrated durable predictive value in metals are limited; most published patterns fail rigorous out-of-sample testing.
The cup and handle, popularised by William O'Neil and observable in gold's 2018-2020 base before the breakout above $1,400, remains a high-probability continuation pattern when accompanied by volume confirmation. The base must be sufficiently long (typically 6 to 18 months for gold), the depth must be reasonable (no more than 30% drawdown from the prior high), and the breakout must occur on volume meaningfully above the base's average.
The inverse head-and-shoulders, observable in silver's 2018-2020 basing pattern before the move from $14 to $30, is a reversal pattern with similar volume requirements. Neckline breaks without volume expansion are routinely false. The ascending triangle, with horizontal resistance and rising support, has appeared at multiple inflection points in the gold tape and tends to resolve in the direction of the longer-term trend.
Pattern failures and the risk of confirmation bias
The honest practitioner accepts that pattern recognition is psychologically vulnerable to confirmation bias. The same chart, viewed by ten experienced technicians, will often yield seven distinct interpretations. Quantitative back-testing of named patterns in metals has produced results that are statistically significant only marginally and often dependent on the lookback window chosen. Pattern analysis is best used as one input among several. See /categories/chart-patterns.
COT Positioning: The Most Underused Public Dataset
The Commitments of Traders (COT) report, published weekly by the CFTC, decomposes COMEX gold and silver futures positioning by trader category. The categories most relevant to precious metals are managed money (speculators, predominantly trend-following hedge funds and CTAs) and producer/merchant/processor/user (commercial hedgers, including miners and refiners).
The canonical COT signal in gold is the managed money net-long position relative to its historical range. Managed money net-longs above the 90th percentile of the trailing 5-year distribution have historically preceded short-term corrections; net-longs below the 10th percentile have preceded rallies. The relationship is not deterministic, and the trader category compositions have evolved (the disaggregated COT format introduced in 2009 changed the comparability of pre- and post-2009 data).
| Category | Typical Behaviour | 2024-2025 Read |
|---|---|---|
| Managed Money | Trend-following | Net-long, near upper decile |
| Commercial Hedgers | Counter-trend | Net-short, expanded |
| Other Reportables | Mixed | Modest net-long |
| Non-Reportables | Retail | Modest net-long |
The 2024-2025 COT readings in gold have shown managed money net-longs persistently in the upper portion of the historical range, which has not prevented continued price appreciation. This is consistent with the regime-shift interpretation, in which official-sector buying outside the COMEX is the dominant flow and futures positioning is no longer the marginal price setter. See /categories/cot-positioning.
Seasonality: Real, But Limited
Seasonal patterns in gold and silver are observable in long-run data but are weaker than retail commentary often implies. The most cited gold seasonal patterns are the August-September strength (Indian wedding season demand, Diwali) and the January effect (new-year portfolio rebalancing, Chinese New Year demand).
Monthly returns analysis of gold since 1975, after the post-Nixon free pricing era began, shows the strongest months are August, September, and January, and the weakest are March and June. The magnitude of the effects, however, is modest, with average monthly returns differing by 1-2 percentage points between strongest and weakest months. The standard deviation of monthly returns substantially exceeds the seasonal differential, meaning that seasonality alone is not a tradeable edge.

| Month | Gold Avg Return (1975-2024) | Silver Avg Return |
|---|---|---|
| January | +1.6% | +1.9% |
| February | +0.4% | +0.2% |
| March | -0.3% | -0.5% |
| April | +0.5% | +0.6% |
| May | -0.1% | -0.2% |
| June | -0.4% | -0.6% |
| July | +0.7% | +0.4% |
| August | +1.4% | +1.1% |
| September | +2.0% | +1.5% |
| October | -0.2% | -0.4% |
| November | +0.6% | +0.5% |
| December | +0.9% | +0.8% |
For sub-cluster analysis see /categories/seasonality.
Intermarket Analysis: The Macro Lens
Intermarket analysis examines the relationships between gold and other macro variables, with the most robust relationships being gold versus the US dollar (DXY), gold versus real yields (10-year TIPS yield), and gold versus broader risk assets.
The gold-DXY inverse correlation has been a reliable feature of the post-2000 era, with rolling 60-day correlations typically in the -0.4 to -0.7 range. The relationship weakens in regime changes, particularly when both the dollar and gold are responding to the same underlying driver (flight to quality, geopolitical stress). Periods of positive gold-DXY correlation have historically been short-lived but informative.
The gold-real yields relationship, with TIPS yields as the standard proxy, has been the dominant model in academic and institutional research since the 2003 introduction of the TIPS market in current form. The model treats gold as a zero-coupon real asset whose price should rise as real yields fall and vice versa. The model worked well from 2008 through 2021, broke down in 2022-2023 as gold rallied despite rising real yields, and has been the subject of substantial revisionist work since.
The breakdown is most plausibly attributed to the official-sector buying channel that operates outside the real-yields framework. Central banks accumulating gold for reserve diversification reasons are not yield-sensitive in the same way that ETF flows are. See /categories/intermarket and /categories/gold-markets.
TIPS, breakevens, and the gold price
For practitioners building macro models, the relevant decomposition is nominal yield = real yield + inflation breakeven. Gold has historically been more sensitive to the real yield component than to the breakeven component, but the 2022-2025 regime has scrambled this relationship. Robust models for 2026 require explicit treatment of central bank flows as an exogenous variable.
Ratio Spreads: The Relative Value Lens
Ratio analysis between precious metals is one of the most useful technical lenses available to investors. The gold-silver ratio (GSR), platinum-palladium ratio, and gold-copper ratio each carry information about both the precious metals complex and broader macro conditions.
The GSR has spent most of the post-Nixon era between 40 and 100, with extremes reached during stress periods. The 2020 COVID spike to 125 was the highest ratio in modern history and was followed by aggressive silver outperformance. The 2024-2025 range of 78-92 has been historically elevated, suggesting either silver upside or gold downside if mean reversion holds.
| Ratio | Long-run Range | 2025 Reading | Interpretation |
|---|---|---|---|
| Gold-Silver (GSR) | 40-100 | 80-90 | Silver historically cheap |
| Platinum-Palladium | 0.5-3.5 | 1.0-1.4 | Palladium normalised |
| Gold-Copper (oz/lb) | 200-600 | 700+ | Gold expensive vs industrial |
| Gold-Oil (oz/bbl) | 10-30 | 40+ | Gold expensive vs energy |
The platinum-palladium ratio has experienced extreme moves, from 0.4 at the 2022 palladium peak (palladium expensive) to over 1.0 currently (platinum normalised). The gold-copper ratio is sometimes used as a recession indicator, with elevated readings suggesting deflationary stress.

For sub-cluster work see /categories/ratio-spreads.
Synthesis: Combining the Five Inputs
The practical synthesis of the five technical inputs runs as follows. Chart patterns provide entry and stop reference points. COT positioning provides a sentiment overlay that flags overcrowded or under-positioned conditions. Seasonality provides a modest tailwind or headwind that should not dominate sizing decisions. Intermarket analysis provides macro context, particularly the gold-DXY and gold-real-yields relationships. Ratio spreads provide relative value reference points across the metals complex.
No single input is sufficient. The discipline is to build a composite read that incorporates all five and to act when multiple inputs align. Acting on a single input, particularly chart patterns alone, is the most common mistake retail technicians make.
For cross-pillar context see /categories/macro-geopolitics and /categories/regulation-tax.


Reader Letters
The mailroom is empty.Be the first to write in.