Hi YXI friends,
Welcome to YX Insights' Multi-model Signals ("MS"), our daily systematic briefing. This is the Commodities book. It comes in two parts.
The portfolio is the book we actually hold. It is re-optimised every Monday, with exits during the week from assets that turned to Flat in Multi-model Signals. You can see the weights and every change as they happen.
The signal board is the wider read behind it: four independent models score every name under coverage, spanning the commodity ETFs themselves and the producers and processors behind them, across precious metals, industrial metals, energy, agriculture and uranium.
Keep the two apart as you read. A name can be long on the board without being in the book, and every portfolio holding is a 1x position whatever tier the board shows.
Coverage
Commodity ETFs (11): GLD, GDX, SLV, SIL, PALL, PPLT, CPER, USO, BCI, URA, DBA
Commodity Equities (13): XOM, NEM, NTR, BHP, RIO, VALE, ALB, DE, LNG, AEM, FNV, SCCO, MP
We will continue to add new names, expanding across equities, crypto, commodities, rates, and FX.
DISCLAIMER: This newsletter is intended for educational purposes only. Any information or analysis in this note does not constitute an offer to sell or a solicitation of an offer to buy any securities. Nothing in this note is intended to be investment advice, nor should it be relied upon to make investment decisions. Any opinions, analyses, or probabilities expressed in this note are those of the author as of the note's date of publication and are subject to change without notice.
1. The Portfolio Today
CO — Commodities · target book for Tue 4 Aug
7 holdings · 65% invested · 35% cash
BCI 11.8% XOM 10.6% DE 10.4%
LNG 10.3% CPER 8.4% NTR 7.3%
BHP 6.4%
Changes: closed ALB, GDX, SLV
GDX has been sold today. The signal turned flat, and a holding that turns flat is closed the same day rather than held to the next re-optimisation.
ALB, SLV have been sold today. The direction is still long, but fewer than three of the four models now agree with it, and the book only carries names the models broadly agree on.
The proceeds sit in cash until the next re-optimisation — they are not spread across the remaining holdings.
Next re-optimisation: Mon 10 Aug
Every holding is a 1× position — the 2× tier on the signals pages does not apply to portfolio allocation.
Cash is the volatility aim de-risking: calmer markets, more invested. The aim is to run at or below SPY's long-term volatility — a best-effort target, not a guarantee.
2. Today's Signal Board
Today's changes
Multi-model Signals: Today's Changes
- GDX: turned flat
- SIL: turned flat
Book posture: 12 long · 12 flat · 24 names covered.
How to read the board: each row shows the book's live position for that name. The four dots show how each of the four models reads it independently, and where they disagree the commentary below explains. Two things are worth knowing before you start. Flat means out of the market, not short. And a holding tag means we are still in the position under the minimum-hold rule while the consensus has already turned the other way. A long signal here is also not the same as a portfolio holding, which section 1 sets out.
Please help me improve the service with your immediate feedback - thank you.
How Useful Do You Find Today's YXI Signals?
3. Signal Commentary
Each note states the live signal, the factors supporting it, and the counter-case against it. Watch the counter-case: when it builds, the signal is usually the next thing to move.
Commodity ETFs
GLD (Long) — The model is long. Gold has firmed over the past fortnight and the downside tilt in recent returns looks washed out. The recent drawdown, softer credit spreads and easing oil momentum all back the position, and the regime is constructive. That said, the trend has been downwards over the past two months, which leans the other way.
GDX (Flat) — The model is flat. Gold miner strength and gold momentum both look stretched, the trend has been downwards over the past two months, and the regime is defensive. US large-cap and big-cap tech momentum add to the caution. Against that, short-term rate momentum leans mildly bullish, but it is not enough to get us back in.
SLV (Long) — The model is long. Volatility and the way price looks stretched both back the position, and the trend cleanliness read supports staying in. The return in silver looks washed out, which also favours the long, and the regime is constructive. Against that, the trend has been downwards over the past two months and US large-cap momentum leans bearish.
SIL (Flat) — The model is flat. The trend has been downwards for ten months and the regime is defensive, keeping us on the sidelines. Big-cap tech momentum and silver miner volatility add to the caution. Against that, several reads lean bullish: correlation with US large-caps, bitcoin and gold momentum all make the case to get back in, so we would not press the downside here.
PALL (Flat) — The model is flat. The palladium trend looks stretched and the trend has been downwards over the past year, keeping us out. Being flat here is about risk control; this name has a history of sharp rebounds from washed-out levels, so we would not chase weakness. Several reads lean bullish, with oil and silver momentum, risk-adjusted returns and credit spreads all making the case to return, and the regime is constructive.
PPLT (Flat) — The model is flat. Realised volatility argues for caution and the trend has been downwards for ten months, keeping us on the sidelines. Big-cap tech momentum adds to that. Against that, the reads lean bullish: Chinese equity and bitcoin momentum, plus the trend cleanliness, make the case to get back in, and the regime is constructive.
CPER (Long) — The model is long. The trend has been upwards over the past year and the regime is constructive, with gold, inflation-linked bond, emerging-market and Treasury momentum all backing the position. Against that, correlation with US large-caps leans bearish. This name has a history of sharp rebounds, so we would not read too much into any single-session wobble.
USO (Flat) — The model is flat. Softness in long bonds and 7-10y Treasuries weighs on the read, and the trend over the past two months is downwards. Against that, strength in Chinese equities leans bullish. Dollar weakness also points the other way, as does the constructive market regime, so this is the case to get back in. High-yield credit adds a modest drag.
BCI (Long) — The model is long. Bitcoin looks washed out and inflation-linked bonds also lean supportive on a mean-reversion basis. The trend is clean, credit spreads back the position, and the trend over the past two months is upwards alongside a constructive regime. Against that, the trend in broad commodities leans bearish. This name has a history of sharp rebounds from washed-out levels, so we would not read too much into any weakness.
URA (Flat) — The model is flat. High-yield credit softness weighs here, and the trend over the past year is downwards. On the other side, credit spreads lean bullish, as do gains in global equities and gold. The constructive regime also argues for getting back in. Bitcoin weakness adds a modest drag.
DBA (Long) — The model is long. The recent drawdown looks washed out and supports the position, and the trend over the past ten months is upwards alongside a constructive regime. Against that, long bond softness leans bearish. The change in the yield curve and gold momentum also lean the other way, with realised volatility adding to the caution.
Commodity Equities
XOM (Long) — The model is long. Oil momentum backs the position, and the trend over the past ten months is upwards alongside a constructive regime. Against that, gold softness leans bearish. Correlation with US large-caps and softer short-term rates also lean the other way. Volatility adds a touch of caution.
NEM (Flat) — The model is flat. Stretched readings versus trend weigh here, and growth stocks, big-cap tech and US large-caps all lean bearish. The trend over the past year is downwards. Against that, volatility leans bullish, and the market regime is constructive, which is the case to revisit. The reads are otherwise firmly on the cautious side.
NTR (Long) — The model is long. Recent returns and a clean trend back the position, and oil momentum adds support. The trend over the past seven months is upwards and the regime is constructive. Against that, correlation with US large-caps leans bearish. Growth stock momentum also leans the other way.
BHP (Long) — The model is long. The recent drawdown looks washed out and supports the position, with the trend upwards over the past five months and a constructive market regime behind it. Against that, the reads lean bearish: volatility and risk-adjusted returns both look stretched. US large-cap momentum and the broader trend picture add to the caution.
RIO (Flat) — The model is flat. The defensive regime and the two-month trend downwards back staying on the sidelines. Against that, the counter-factors lean bullish: the recent downside/upside tilt, firmer growth-stock momentum, a softer dollar and volatility all make the case to get back in. Softer big-cap tech is the one read on the side of caution.
VALE (Flat) — The model is flat. Strength in global equities and correlation with US large-caps both back staying out, and the two-month trend downwards leans the same way. Against that, the counter-factors lean bullish: the trend versus the broader picture, softer big-cap tech and the recent drawdown make the case to get back in, with a constructive regime behind them.
ALB (Long) — The model is long. Correlation with US large-caps and softer oil momentum back the hold, and three-month volatility leans the same way, with a constructive regime behind it. Against that, the recent downside/upside tilt and strength in growth stocks lean bearish. The seven-month trend is downwards, which also argues for caution.
DE (Long) — The model is long. Lower volatility backs the hold, with softer oil and bitcoin momentum leaning the same way. The two-month uptrend and constructive regime sit on that side too. Against that, the trend versus the broader picture leans bearish. The recent downside tilt is on the supportive side, so the balance keeps us in.
LNG (Long) — The model is long. Softer oil momentum backs the hold, and volatility and the seven-month uptrend lean the same way, with a constructive regime behind it. Against that, weaker global bonds, the change in credit spreads and the risk-adjusted return picture all lean bearish. The reads are two-sided but we hold.
AEM (Long) — The model is long. Risk-adjusted returns back the hold, the year-long trend is upwards and the regime is constructive. Against that, recent strength in US large-caps looks stretched and leans towards caution. On balance we stay with it.
FNV (Flat) — The model is flat. Strength in growth stocks and elevated volatility both back staying out, and the trend versus the broader picture leans the same way. The five-month trend downwards adds to that. Against that, the counter-factors lean bullish: firmer 7-10y Treasuries, the recent drawdown and the constructive regime make the case to get back in.
SCCO (Flat) — The model is flat. Softness in 7-10y Treasuries and in the dollar both back staying out, and the recent drawdown adds to that read. Against that, the counter-factors lean bullish: the three-month volatility and return picture and the constructive regime make the case to get back in. The two-month trend is downwards, which sits on the side of caution.
MP (Flat) — The model is flat. Bitcoin and dollar momentum both lean bearish, and high-yield credit momentum adds to the drag. The year-long downtrend and defensive regime back staying out. Against that, the recent return and change in credit spreads lean bullish and make the case to get back in. This name has a history of sharp rebounds from washed-out levels, so being out is about risk control rather than a call for further weakness.
-----
4. Position History
The running record of every signal, so you can see each position's history and how long each stance has been held.
How the four models work, the leverage policy and the ETF alternatives now live on the YX Insights website.