The Momentum Trade Looks Toppy Going Into July — How to Stress-Test Your Portfolio in One Sheet

The Setup in One Table

The factor that ran the first half is starting to run out of buyers at the margin. Bloomberg’s US equities desk flagged momentum as the best-performing US equity factor through the first half, with the MSCI USA Momentum benchmark up roughly 28% year to date, and called out that the trade is “starting to fade.” The QQQ has already lost 4.5% over the past month while the Russell 2000 — the small-cap laggard of 2025 — has quietly gained 2%. The VIX spike to 22.22 on June 10, followed by an easy back to 16.59 by July 1, is the volatility signature of a market that is no longer one-directional.

Hedge Funds Are Sitting on Their Biggest Quarter in Years

Goldman Sachs’ prime brokerage data shows hedge funds just closed one of their best quarters on record, with the typical fund up double digits across the first two quarters of 2026. That is the kind of return that historically triggers systematic profit-taking into the seasonally weak summer months. The same Bloomberg report notes that retail cash equity trading volume is running about 65% above 2025 levels, with the activity concentrated in AI, hyperscaler, and megacap tech — exactly the names that make up the momentum basket.

The classic setup is well known. Foreign buying of US equities has been a marginal tailwind all year and is starting to fade as the dollar’s recent strength makes US assets more expensive for international buyers. The 2s-10s spread has compressed from 0.74% in early February to 0.35% now, which historically signals late-cycle behavior in the economy. June nonfarm payrolls came in at just 57,000 versus consensus estimates of 100,000 to 110,000, but average hourly earnings still rose to $37.64 from $36.36 a year earlier — a softening labor market without the disinflation that would justify Fed cuts.

Put it together and the picture is uncomfortable. Markets priced to perfection on momentum, hedge funds sitting on outsized gains, retail concentrated in the same handful of AI names, foreign buyers stepping back, and a labor market that is cooling faster than inflation. That is not a forecast. It is the configuration that has preceded every meaningful momentum unwind in the last twenty years.

What an Unwind Would Actually Do to Your Portfolio

A momentum unwind is not a broad market crash. It is a relative-performance event. The leaders — Nvidia, Microsoft, Meta, Alphabet, Amazon, Apple, plus the AI infrastructure supply chain — give back several weeks of relative gains, while small caps, banks, transports, and other cyclical sectors catch a bid. In a typical rotation, the Russell 2000 outperforms the Nasdaq-100 by 5 to 10 percentage points over a four-to-six week window. The S&P 500 as a whole usually closes roughly flat.

The portfolio effect depends entirely on what you actually own, and where. A position in QQQ plus Microsoft plus an AI-themed ETF plus a tech-sector mutual fund is not four positions. It is one bet, repeated four times. Each brokerage app shows you its slice in isolation. None of them tells you that the same underlying names dominate every “diversified” wrapper you bought. That is how a 20% drawdown in Nvidia can become a 6% hit to your net worth — not because you concentrated in Nvidia, but because you unknowingly concentrated in Nvidia four different ways.

The cleanest stress test is to put every account in one sheet and ask a single question: what fraction of my total wealth is exposed to the same handful of names that drove the first-half momentum trade? If the answer surprises you, the portfolio is not as diversified as your brokerage apps have been telling you.

How to Run the Stress Test in One Sheet

The fix is to put every account in one sheet and let formulas do the aggregation. InvestSheet syncs Fidelity, Schwab, Robinhood, and 30+ other brokerages into a single Google Sheet and exposes them through one set of built-in functions. The relevant one for a momentum stress test is IVS_BROKERAGE, which returns the aggregated value of any ticker across every linked account. A typical layout:

From there, the math is yours. Flag any single ticker that crosses 5% of your total portfolio. Flag any single sector that crosses 30%. Flag any single theme (AI infrastructure, AI applications, AI power) that crosses 40%. The numbers will not match what any single brokerage app is showing, and that is precisely the point. The unwind will reward investors who knew what they actually owned before the volatility hit.

Frequently Asked Questions