Hi YXI friends,
Welcome to YX Insights' Multi-model Signals ("MS"), our daily systematic briefing. Four independent models score every name under coverage, spanning equity indices, bonds, crypto, and megacap equities. The signal board shows where the book stands today. The commentary explains what drives each call.
The methodology and backtest results are in the launch note:
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.
Today's Coverage
Macro Assets: SPY, QQQ, IWM, EFA, EEM, SOXX, IGV, BTC, TLT, IEF, HYG, UUP
Magnificent 7: AAPL, AMZN, GOOGL, META, MSFT, NVDA, TSLA
Megacaps: AMD, AMAT, ANET, ASML, AVGO, CDNS, CRM, CRWD, HUBS, INTC, MDB, MRVL, MU, NET, NOW, OKTA, ORCL, PANW, PLTR, QCOM, SNPS, SNOW, STX, TSM, TXN, VRT, VST, ZS
We will continue to add new names, expanding across equities, crypto, commodities, rates, and FX.
1. Today's Signal Board
Today's changes
IGV: sized to 2× (long)
AMZN: turned long
Book posture: 30 long · 17 flat · 47 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. Where they disagree, the commentary below explains.
Please help me improve the service with your immediate feedback - thank you.
How Useful Do You Find Today's YXI Signals?
2. 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.
Macro Assets
SPY (Long) — The model is long. The change in credit spreads is supportive, and large-cap volatility is behaving. The upward trend over the past seven months and the constructive regime back the position. Against that, momentum across emerging-market equities, big-cap tech and global equities all look stretched and lean bearish.
QQQ (Long) — The model is long. Oil momentum is supportive, and the yield curve is helping alongside firmer long bonds. The two-month uptrend and the constructive regime back the position. Against that, global equity and small-cap momentum both look stretched and lean bearish.
IWM (Long) — The model is long. Gold momentum is supportive, and steadier 7-10y Treasuries add to that. The recent drawdown is working in our favour, and the year-long uptrend and constructive regime back the position. Against that, the trend and returns in small-caps themselves look stretched and lean bearish.
EFA (Long) — The model is long. Momentum in inflation-linked bonds and gold is supportive, and the recent drawdown has washed out. The five-month uptrend and constructive regime back the hold. Against that, price looks stretched here and oil momentum leans bearish, so there is a case for some caution.
EEM (Long) — The model is long. Correlation with US large-caps is supportive, and gold momentum adds to the case. The seven-month uptrend and constructive regime back the hold. That said, emerging-market equities look stretched on both trend and return, and yen momentum leans bearish, arguing for caution.
SOXX (Long) — The model is long. The seven-month uptrend backs the position. That said, the reads here lean bearish: oil momentum is soft, and correlation with US large-caps, big-cap tech and global equity momentum are all working against us. The defensive regime adds to the caution.
IGV (Long) — The model is long. The recent drawdown and firm risk-adjusted returns are supportive. Gold momentum, steady software volatility and correlation with US large-caps all help, and the five-month uptrend and constructive regime back the position. The reads are broadly on-side here.
BTCUSD (Flat) — The model is flat. The trend over the past five months is downwards, which keeps us on the sidelines. Being out here is about risk control; this name has a history of sharp rebounds from washed-out levels, so we would not chase weakness either. The counter-case leans bullish, with short-term rates momentum, credit spreads and the recent return tilt all arguing for getting back in, and the regime is constructive. Global equity and gold momentum lean the other way.
TLT (Flat) — The model is flat. Credit spreads and long-bond trend lean bearish, and the seven-month downtrend and defensive regime support staying out. That said, softer yen momentum and steadier 7-10y Treasuries lean the other way and make the case to get back in. Being out here is about risk control; this name has a history of sharp rebounds from washed-out levels, so we would not chase weakness either.
IEF (Flat) — The model is flat. The reads are thin here, so we are on the sidelines. The two-month uptrend and constructive regime are the case to get back in, but there is little else driving the signal for now.
HYG (Long) — The model is long. Global bond momentum is supportive and the recent drawdown has been shallow. The shape of the yield curve backs the hold, and the ten-month uptrend and constructive regime add weight. Against that, bitcoin momentum and the yield-curve percentile lean bearish, so there is a case for some caution.
UUP (Long) — The model is long. Emerging-market equity and big-cap tech momentum lend modest support, and the five-month uptrend backs the position. Against that, the reads lean bearish: gold momentum, the choppy character of the trend and elevated volatility all argue for caution. The defensive regime adds to that.
Magnificent 7
AAPL (Long) — The model is long. Firmer volatility in the name backs the position, and the two-month uptrend and constructive regime lean the same way. Against that, risk-adjusted returns and the name's own trend look stretched. Softness in global bonds and tighter credit spreads lean bearish too. Those are the case for caution.
AMZN (Long) — The model is long. Oil momentum backs the position, and both bitcoin and its own momentum add support. The five-month trend is upwards and the regime is constructive. Against that, correlation with US large-caps leans bearish, and softer large-cap momentum cuts the same way. This name has a history of sharp rebounds, so we would not lean too hard on the caution.
GOOGL (Flat) — The model is flat. Correlation with US large-caps leans bearish, and risk-adjusted returns cut the same way. The trend over the past two months is downwards and the regime is defensive, both backing the sidelines. Against that, realised volatility leans bullish and is the main case to get back in.
META (Flat) — The model is flat. Correlation with US large-caps leans bearish, and softer bitcoin momentum cuts the same way. The defensive regime backs staying out. Against that, the recent drawdown leans bullish and volatility points the same way, and the five-month trend is upwards, so the counter-case to re-engage is there.
MSFT (Flat) — The model is flat. Volatility leans bearish, and the downside/upside tilt in recent returns cuts the same way. The trend over the past seven months is downwards, backing the sidelines. Against that, correlation with US large-caps and high-yield credit momentum both lean bullish, and the regime is constructive. Being out here is about risk control; this name has a history of sharp rebounds from washed-out levels, so we would not chase weakness either.
NVDA (Long) — The model is long. The upward trend over the past seven months backs the position, and the constructive market regime supports it. That said, the counter-factors lean bearish. Price looks stretched on the month, and momentum in US large-caps, growth stocks and bitcoin all lean the other way. Softer volatility over recent months adds a touch more caution.
TSLA (Flat) — The model is flat. The downward trend over the past two months keeps us out, and the defensive regime supports that. That said, most of the factor reads lean bullish and make the case to get back in. The recent drawdown and softer volatility look washed out, the downside tilt in returns is turning, and momentum in gold leans supportive. Being out here is about risk control; this name has a history of sharp rebounds from washed-out levels, so we would not chase weakness either.
Megacaps
AMD (Long) — The model is long. Investment-grade credit momentum is supportive, and momentum in bonds and high-yield credit adds to the case. The five-month trend is upwards and the regime is constructive. Against that, the three-month return looks stretched and leans bearish, and correlation with US large-caps points the same way. This name has a history of sharp rebounds from washed-out levels, so we would not read the caution too heavily.
AMAT (Long) — The model is long. Correlation with US large-caps over recent months backs the position, and softness in inflation-linked bonds and high-yield credit adds support. The seven-month uptrend and constructive regime lean the same way. Against that, risk-adjusted returns look stretched and recent large-cap strength leans bearish. The bearish reads are the case for caution here.
ANET (Flat) — The model is flat. The prevailing trend versus its average leans bearish, and price looks stretched. Softness in high-yield credit backs the sidelines stance, and the defensive regime pulls the same way. Against that, the year-long uptrend and firmer global equities lean bullish, and those are the case to re-engage. The reads here are genuinely mixed.
ASML (Long) — The model is long. Softer gold and firmer investment-grade credit both back the position. The downside/upside tilt in recent returns adds support, with the seven-month trend upwards and the regime constructive. Against that, risk-adjusted returns and the three-month return both look stretched and lean bearish.
AVGO (Long) — The model is long. Bitcoin momentum backs the position, with the year-long trend upwards and the regime constructive. Against that, softer big-cap tech momentum leans bearish. Gold, volatility and high-yield credit momentum lean the same way. The counter-reads carry some weight but the trend and regime hold the stance.
CDNS (Long) — The model is long. A lighter correlation with US large-caps is supportive on our read, and easing volatility leans the same way. The five-month uptrend and constructive regime back the hold. Against that, softness in oil, high-yield credit and big-cap tech leans bearish and is the case for caution.
CRM (Flat) — The model is flat. The downward trend over the past five months backs staying out, and the defensive regime leans the same way. The downside tilt in recent returns and the uneven quality of the trend add to the caution. The firmer one-month return also reads bearish here. Against that, the recent drawdown leans bullish and is the case to get back in.
CRWD (Long) — The model is long. Softer gold backs the position, and the name's own trend and recent return add support. The two-month trend is upwards and the regime is constructive. Against that, correlation with US large-caps leans bearish, and credit spreads cut the same way.
HUBS (Flat) — The model is flat. The prevailing trend leans bearish and the five-month direction is downwards. However, several factors lean the other way: short Treasuries and short-term rates both look supportive, and narrower credit spreads add to the case to get involved. The constructive regime also argues for a return. For now we stay out.
INTC (Long) — The model is long. Weakness in long bonds backs the position, and firmer realised volatility lends support. The prevailing trend and strength in growth stocks pull the same way. The year-long uptrend and constructive regime add to the case. Against that, softness in oil leans bearish, and that is the note of caution.
MDB (Flat) — The model is flat. Price looks stretched over the past three months and leans bearish, whilst short Treasuries and small-caps add to the caution. The defensive regime supports staying out. Against that, gold momentum leans supportive and the recent skew in returns leans the other way, making the case to re-engage. The reads are somewhat mixed, but a downwards trigger is what we would need. We remain flat.
MRVL (Long) — The model is long. The two-month uptrend and constructive regime back the position, and softness in Treasuries lends further support. Against that, weakness across investment-grade and high-yield credit leans bearish. Softness in oil and global bonds pulls the same way. The counter-factors are the case for caution.
MU (Long) — The model is long. Softer volatility on the month leans supportive, and momentum in gold is helping. The upward trend over the past seven months backs the position too. Against that, risk-adjusted returns and the trend itself look stretched over recent months, and correlation with US large-caps leans bearish. The defensive market regime adds to that caution.
NET (Long) — The model is long. Firmer credit spreads lean supportive here, and the upward trend over the past seven months backs the position. The constructive market regime helps too. Against that, the trend in Cloudflare and elevated volatility on the month lean bearish. The downside tilt in recent returns adds to that caution.
NOW (Flat) — The model is flat. Momentum in longer-dated Treasuries and long bonds is supportive on our read, and strength in global equities leans the same way. Elevated volatility also reads bullish here. Against that, the five-month uptrend and firmer short Treasuries lean bullish and make the case to get back in, whilst the defensive regime argues for patience. For now we stay on the sidelines.
OKTA (Long) — The model is long. Gains over the past month back the position, the trend versus its path leans supportive, and firmer 7-10y Treasuries add to the case. The five-month trend is upwards and the regime is constructive. Against that, soft small-caps and the recent skew in returns lean bearish. The book stays long.
ORCL (Long) — The model is long. Softer volatility on the month leans supportive here. Against that, the downward trend over the past five months and the constructive regime pull in different directions, and most of the reads lean bearish. Momentum in inflation-linked bonds and value stocks leans that way, as does the yield-curve backdrop. The picture is mixed here.
PANW (Long) — The model is long. Strength in small-caps supports the hold, and the ten-month uptrend backs it. Against that, several factors lean bearish. Weakness in bitcoin and the shape of the yield curve argue for caution. Softer volatility, the downside tilt in recent returns and the defensive regime add to that case.
PLTR (Flat) — The model is flat. The downward trend over the past ten months keeps us on the sidelines, and the defensive regime backs staying out. Momentum in oil and global equities leans bearish, and the trend looks messy. Against that, dollar momentum leans bullish and would be part of the case to get back in. Being out is about risk control; this name has a history of sharp rebounds from washed-out levels, so we would not chase weakness either.
QCOM (Flat) — The model is flat. The two-month downtrend and defensive regime back the sidelines stance. Correlation with US large-caps leans the same way, and recent returns look stretched. Against that, strength in global equities and growth stocks leans bullish. Firmer long bonds pull the same way, and those are the case to get back in.
SNPS (Flat) — The model is flat. The five-month uptrend and constructive regime would argue for involvement, and correlation with US large-caps, strength in growth stocks, the softer one-month return and the recent drawdown all lean bullish. That is the case to get back in. Against that, weakness in small-caps leans bearish. The reads here tilt constructive but we remain on the sidelines for now.
SNOW (Long) — The model is long. Softer yen and firmer high-yield credit lean supportive, and gains in US large-caps add to the case. The two-month trend is upwards. Against that, the recent gain looks stretched and lean bearish, and the tight link to US large-caps cuts the same way. The defensive regime is a further note of caution, but we stay long.
STX (Long) — The model is long. Elevated volatility is supportive on our read, and the seven-month uptrend and constructive regime back the hold. Against that, the stretched three-month return leans bearish and looks a touch extended. Momentum in growth stocks, big-cap tech and short-term rates leans the same way and is the case for caution.
TSM (Long) — The model is long. Softness in US bonds backs the position, and firmer short-term rates lean the same way. Recent softness in the name and its lighter correlation with US large-caps add support. The two-month uptrend and constructive regime pull the same way. Against that, the level of the yield curve leans bearish, and that is the case for caution.
TXN (Long) — The model is long. Softness in the dollar backs the position, and strength in growth stocks lends support. Weakness in long bonds leans the same way. The seven-month uptrend and constructive regime pull with the position too. Against that, correlation with US large-caps and softness in the yen lean bearish, and those are the case for caution.
VRT (Long) — The model is long. Softness in big-cap tech and small-caps is supportive on our read, and weakness in bitcoin adds to that. The five-month uptrend and the constructive regime back the hold. Against that, momentum across US bonds and longer Treasuries leans bearish and is the case for caution.
VST (Flat) — The model is flat. The downward trend over the past ten months weighs against getting involved, and the downside tilt in recent returns adds to the caution. Momentum in the yen also leans bearish here. Against that, the constructive market regime leans bullish, and both credit spreads and the softer three-month return argue the case to get back in. For now those are not enough to draw us off the sidelines.
ZS (Flat) — The model is flat. Softness in inflation-linked bonds and short-term rates leans bearish, and the seven-month trend is downwards with the regime defensive. However, the counter-case is substantial: growth-stock momentum, firmer gold and a stronger dollar all lean bullish and make the case to get back in. A trend reversal is what we would need to see. For now we stay out.
3. Position History
The running record of every signal, so you can see each position's history and how long each stance has been held.
Appendix: How the Models Work
Meet the four models
Each model thinks in a genuinely different way. That difference is the point.
1. The Machine-Learning model
This is the evolution of our original engine. We have made upgrades to its inputs and parameters since Q1, but the core model logic remains.
It reads the name's own price behaviour, such as momentum and volatility. It also reads signals from other markets, such as bonds, the dollar, and credit. It spots links across many markets at once, the kind a human would miss.
2. The Neural-Network model
It reads the same market signals as the ML model, but learns them in a completely different way. It finds the recurring patterns across short and long time scales and projects the next move forward. It is a second, independent way of reading the same evidence. The two often disagree, and that disagreement is useful.
3. The Trend model
When a real move is in place, it stays with it. When the price is choppy, it steps aside. No cleverness about tops and bottoms, just staying on the right side of a trend.
4. The Regime model
It reads the market's hidden state, risk-on, risk-off, or neutral, like telling you the season.
None of the four is the whole truth. Together, they get a lot closer.
How they work together
Think of the four models as candidates. For each name, we test it alone and in combination. We judge them on data they were not trained on. That is the honest test of a real edge, not a model that has simply memorised the past.
The Machine-Learning model still does much of the heavy lifting. It is in the winning line-up for most names. The Neural-Network model plays a supporting role, stepping in where it has proven more useful than the others. The principle is simple: the right tool for each name, chosen by evidence.
Whichever line-up drives a name, you always see how all four models are positioned that day, shown as four dots on the page, one per model. They are a transparency layer. You see the full read behind every signal, including where the models disagree.
The pattern of agreement can be telling in its own right. Markets are made of players running different styles. When the models all swing the same way, it often means a name is already stretched and positioning has become one-sided. That is especially true of an all-out risk-off. It is exactly the setup from which sharp reversals come. We have seen an all-out bearish consensus snap back hard more than once. Traders call that a short squeeze. The daily commentary is there to flag those moments.
The real engine: diversification
This is the idea worth taking away. Because the names we cover across a range of asset classes behave differently at different times, their off-days rarely line up. Weakness in one corner is often offset by strength in another. You do not need every name to work. You need the book to work.
We run the system across our full book, meaning we cover all the names at once. We are also expanding our coverage over time, with names spanning equities, crypto, commodities, rates, and FX.
Leverage policy: 2× long-only, no shorting
Shorting is simply too difficult and not worth our time most of the time.
Having examined our live performance using leverage and short strategies and performed extensive testing, I have decided to keep the leverage strategy but remove the shorting strategy.
Our leverage strategy assumes a 5% p.a. borrowing rate.
ETF alternatives (including leveraged ETFs)
