The past week's winners keep handing it back. Short-Term Momentum — whether the prior week's multi-day movers keep going or give it back, with yesterday's move stripped out — fell another -0.77% today, a session in the bottom 2% for that factor since 2020, and the path is the point: -5.81% over the past twenty sessions, -8.36% over sixty, four of the past five down. Over the full sample, reversion at this horizon is weak enough to be statistically insignificant; it gets paid in conditions like those of the past month — and especially the past two weeks — which is what makes a 20-day z=-4.0 the most informative number on the page. The selling is ordered: the more a name had run over the prior week, the harder it was hit today. The shape of the week matters as much as the session, though — over the past five days the reversal has been carried mainly by the bid under last week's losers, whose gains run about three times the winners', the dip-buying half of releveraging churn; today the pressure flipped to the winners' side.
| Bucket | Ret 1D Pct | Ret 5D Norm Pct | Ret 20D Norm Pct | Ret 63D Norm Pct |
|---|---|---|---|---|
| 1 | 0.30 | 3.74 | 1.73 | 1.42 |
| 2 | -0.28 | 3.27 | 1.69 | 2.00 |
| 3 | 0.09 | 2.67 | 1.60 | 1.33 |
| 4 | 0.39 | 2.53 | 2.02 | 2.32 |
| 5 | 0.67 | 1.57 | 1.10 | 1.02 |
| 6 | 0.04 | 1.56 | 1.06 | 1.25 |
| 7 | -0.16 | 1.53 | 1.43 | 1.58 |
| 8 | -0.15 | 1.34 | 1.33 | 1.59 |
| 9 | -0.26 | 0.94 | 0.99 | 1.51 |
| 10 | 0.08 | 0.87 | 0.59 | 1.06 |
| 11 | -0.12 | 0.72 | 0.27 | 0.92 |
| 12 | -0.29 | 0.80 | 0.57 | 1.38 |
| 13 | -0.64 | 1.31 | 0.56 | 0.88 |
| 14 | -0.59 | 0.91 | 0.30 | 1.29 |
| 15 | -1.63 | 0.69 | 0.46 | 1.07 |
| 16 | -1.12 | 0.71 | -0.11 | 0.77 |
| 17 | -1.16 | 0.64 | 0.15 | 0.49 |
| 18 | -1.02 | 1.35 | -0.42 | -0.43 |
| 19 | -2.47 | 1.05 | -0.69 | 0.52 |
| 20 | -1.25 | 1.00 | -1.18 | -0.57 |
| Factor | Return | Z-Score | 5d Z | 20d Z | 63d Z | Category | Direction |
|---|---|---|---|---|---|---|---|
| Short-Term Momentum | -0.77% | -2.35 | -4.11 | -4.00 | -3.39 | Style-Momentum | Last week's winners gave back |
| Liquidity | -0.57% | -2.03 | +1.90 | -0.29 | -0.95 | Style | Lowest-turnover names led |
| Gold | +0.64% | +1.97 | -0.19 | +0.49 | -0.95 | Thematic | Gold-exposed names led |
| Residual Volatility | -0.89% | -1.76 | +1.14 | -0.33 | -0.37 | Style-Risk | More volatile names lagged |
| One-Day Momentum | -0.46% | -1.61 | -1.20 | +0.31 | +1.57 | Style-Momentum | Yesterday's losers snapped back |
| Treasury (Duration) | +0.50% | +1.49 | -0.3 | +1.0 | +0.3 | Thematic | Long-duration names led |
| Medium-Term Momentum | +0.35% | +1.19 | -0.3 | +0.1 | +0.9 | Style-Momentum | 20-day winners led |
| Short Interest | -0.16% | -0.94 | -2.07 | -0.94 | +1.06 | Style-Positioning | Most-shorted names lagged |
The published positioning numbers are the fuel here, not the counter-argument. Goldman Sachs Prime Brokerage data has global total leverage at the 93rd percentile of its five-year range even after late July's forced risk-cutting, and JPMorgan, reporting August 4, had quantitative equity funds as the most levered strategy at 450%, with global funds surrendering roughly three points of year-to-date gains in July's tech unwind. Books that full, cut and rebuilt in turns after a stress episode, are what keeps a reversal running — the churn itself keeps handing last week's winners back. Those figures describe books through the prior session at the latest; today's factor prints describe what the trade itself did. The top 50 high-beta momentum names fell -3.68% against -0.40% for the rest of the universe — a session in the weakest tenth since 2020 — while the broader ~190-name version of the same screen fell -0.95%. A run of this length across a documented de-grossing and re-grossing episode is indicative of hedge-fund deleveraging and releveraging churn, a claim about the pattern rather than about anyone's leverage today.
Our positioning lenses decline to name the seller. Hedge-Fund Ownership was flat on both the residualized reading (+0.05%, z=+0.4) and the raw one, so the hedge-fund-owned names were neither singled out nor spared. The top 50 are modestly less shorted than the universe and carry a retail-heavy tilt about one standard deviation above it on our habitual early-trading measure, and Short Interest at z=-0.9 shows the crowded short side still working — no squeeze in this. In yesterday's session those same names rose with volume running below the market's pace, which does not amount to evidence of a forceful re-levering.
Three of the day's other style movers are one position wearing different labels. Residual Volatility, Liquidity and One-Day Momentum all closed negative, and all three score the high-beta momentum names on the losing side of the sort. Medium-Term Momentum at z=+1.2 looks like the offset and is not: after July those names sit on the weak side of that window's basket, so a positive reading there is the same selling under another factor's name. Liquidity's internals say who is being trimmed — the lowest-turnover names outperformed the most heavily traded and most heavily optioned ones, with bitcoin-linked equities at the extreme (WGMI -4.77%).
ADP reported private payrolls up 44,000 in July against a 75,000 consensus, as CNBC reported. Iran and Oman reached agreement on a Strait of Hormuz shipping route and are in the final drafting of a joint statement, as Bloomberg reported, with President Trump signaling a deal could be announced as early as Wednesday, as Bloomberg reported. Spot gold rose +3.8% to $4,231.79 an ounce and silver +5.1%, with markets trimming the odds of a September Fed hike to 57%, as the NYT reported. Our Gold factor returned +0.64% at z=+2.0, in the top 1% of Gold sessions since 2020, and the miners did the carrying: AEM +10.32%, NEM +6.76%, GDX +7.48% on 2.5× its typical session volume, with Materials the strongest sector median stock at +2.52%.
The rate and currency legs were tiny by comparison. Every Treasury tenor fell, but the 10-year only 1.4 basis points, to 4.613%, as CNBC reported, and the dollar index eased -0.16% to 99.69, as CNBC reported. Gold moved far more than either, and it is still a bounce inside a three-month hole: GLD -9.78% and GDX -9.03% over the past 63 sessions. Energy took the other side of the Hormuz story, with EOG closing -6.32% after its results.
| ETF | Theme | Today | 1d Ago | 5d Ago | 20d Ago | 63d Ago |
|---|---|---|---|---|---|---|
| GDX | gold miners | +7.48% | +2.46% | +5.00% | +2.85% | -9.03% |
| SLV | silver | +4.15% | +2.63% | +4.14% | -1.14% | -18.35% |
| GLD | gold | +4.11% | +0.66% | +1.30% | -0.88% | -9.78% |
| SPY | large cap | -0.05% | +1.80% | +4.11% | +3.16% | +7.43% |
| MTUM | momentum factor | -1.21% | +3.83% | +7.19% | +0.29% | +10.22% |
| XLE | energy | -2.02% | -0.46% | +1.65% | +7.10% | -1.46% |
| IGV | software | -2.50% | +4.70% | +11.14% | +8.36% | +15.33% |
| DRAM | memory chips & storage | -3.11% | +7.35% | +14.90% | -9.41% | +29.24% |
| XOP | oil & gas E&P | -3.92% | -1.34% | +3.35% | +9.27% | -5.01% |
| Sector | Median Ret Pct |
|---|---|
| Energy | -1.72 |
| Information Technology | -1.69 |
| Communication Services | -1.25 |
| Industrials | -0.54 |
| Utilities | -0.49 |
| Financials | -0.24 |
| Real Estate | -0.24 |
| Consumer Discretionary | 0.05 |
| Consumer Staples | 0.12 |
| Health Care | 0.35 |
| Materials | 1.43 |
This is the volatility regime that pays reversion. Implied single-stock volatility (Cboe VIXEQ) is at 43.5, the 93rd percentile of its own history, while implied index volatility (VIX) is at 15.8, the 37th — stress still priced name by name, not at the index. Realized data agrees on the shape: over the past twenty sessions, style variance is running at 29.8 volatility points, the 97.6th percentile since 2020, against an index leg of 12.4 points at the 26th. Twenty-day realized pairwise correlation of 8.0% remains near the bottom of its since-2020 range, where it has sat for weeks. Elevated single-stock volatility after a stress episode is precisely when short-term reversion runs hottest — and all of it is invisible from the index level alone — SPY finished -0.05%.
| Series | Today Vol Pts | Prev Close Vol Pts |
|---|---|---|
| total | 50.00 | 49.49 |
| market | 12.40 | 12.59 |
| style | 29.80 | 29.92 |
The stock-by-stock tape does not undercut the reversal read — single stocks moving on their own stories at unusual amplitude is the environment short-term reversion feeds on — though it does keep the style block small. Stock-specific moves accounted for 83.8% of cross-sectional variance against a 66.5% one-year average, with the whole style block at 12.6%. Shopify closed +17.13% on its revenue beat and low-thirties third-quarter guidance; Eli Lilly closed +4.86%, adding $43.9B of market value on a 48% revenue jump; AMD closed -7.11% despite a record quarter, on guidance that cleared consensus but not the buy-side whisper; Insulet closed -19.98%; and Materion closed +31.43% on the biggest stock-specific move in the universe, with no driver yet visible in today's coverage. After the bell, AppLovin fell a further -21.95% having closed the regular session at -0.46%, and Western Digital dropped another -9.89% after closing -5.29%, keeping the memory-chip complex unsettled. None of that is the style axes at work. The style signal is the ordering inside the high-beta momentum group, where the week's leaders keep finishing at the bottom — the same pattern four sessions out of five.
The ADP National Employment Report, released at 8:15 AM ET, showed private payrolls up 44,000 in July against a 75,000 Dow Jones consensus and a revised 95,000 in June, with pay growth for job switchers at 7% — and a Dow Jones survey looking for 83,000 hires and 4.2% unemployment in Friday's official BLS report, as CNBC reported. ISM Services, released at 10:00 AM ET, printed 54.1 for July, a third consecutive month of expansion but below consensus, with new orders and employment both holding above 50.
Variance decomposition: live intraday — 20260805 session, bracketed against its trailing-year range. Factor returns are trailing through last close. Total cross-sectional dispersion: 88%ile of the past year.
Variance mix — % of total, today vs 1d ago vs 1-yr avg
market 28%ile style 36%ile thematic 59%ile idiosyncratic 80%ile
Variance explained — today vs. factor's trailing-year range
Marker = the factor's share of today's total variance, placed in its own trailing-year range (box 25–75%ile, ticks 90%ile and max). Amber marker = unusually load-bearing today (≥90%ile of its own year). Factor name green = up today / red = down.
Correlation & dispersion — realized vs implied
Pairwise realized correlation (20d): 8.03% 3.1%ile since 2020 · 35%ile 3m
Top-500 pairwise: 10.71% 8.6%ile since 2020 · 54%ile 3m · implied (Cboe COR1M): 7.08 0.9%ile of its own history (gap +3.6pp)
Realized 20d vol (ann. pts): all stocks 50 vs VIXEQ 43 · index 12 vs VIX 16 · style factors 30 97.6%ile
20-day window (19 completed days + today); percentiles vs model history since 2020; '3m' tokens rank vs the trailing 63 trading days; definitions in Terms below. Implied prints (Cboe) are delayed ~15 min. as of 16:41 ET
Top 50 high-beta momentum names (eq-wt, since-2020 rank): -3.7% 8.6%ile · rest of market -0.4%
| Factor | Today | 1d | 5d | 20d | 60d |
|---|---|---|---|---|---|
| Style-Risk | |||||
| Residual Volatility | -0.89% z-1.8 | +0.76% | +1.27% | -0.74% | -1.53% |
| Beta | -0.49% z-0.5 | +2.20% | +7.60% | -1.59% | -3.29% |
| International | +0.14% z+0.4 | -0.48% | -0.81% | -0.15% | -1.62% |
| Style | |||||
| Liquidity | -0.58% z-2.0 | +0.33% | +1.20% | -0.37% | -1.92% |
| Value | -0.29% z-1.2 | -0.11% | -0.19% | +1.17% | -0.89% |
| Leverage | -0.21% z-1.2 | -0.22% | -1.02% | -1.41% | -1.79% |
| Dividend Yield | -0.09% z-0.5 | +0.14% | +0.06% | -0.28% | -0.74% |
| Growth | +0.11% z+0.4 | +0.07% | +0.16% | +0.56% | +2.78% |
| Profitability | -0.07% z-0.4 | +0.13% | -0.38% | +0.49% | -0.64% |
| Size | +0.12% z+0.3 | -0.23% | +0.56% | -0.57% | -0.78% |
| Style-Momentum | |||||
| Short-Term Momentum | -0.76% z-2.3 | -0.46% | -2.99% | -5.81% | -8.36% |
| One-Day Momentum | -0.47% z-1.7 | -0.25% | -0.78% | +0.38% | +4.14% |
| Medium-Term Momentum | +0.35% z+1.2 | -0.13% | -0.19% | +0.19% | +1.54% |
| Long-Term Momentum | -0.51% z-0.8 | +0.02% | +0.08% | -4.40% | -3.90% |
| Style-Positioning | |||||
| Short Interest | -0.16% z-0.9 | +0.35% | -0.79% | -0.71% | +1.09% |
| Hedge-Fund Ownership | +0.06% z+0.5 | -0.12% | +0.01% | -0.14% | -0.22% |
| Style-Flow | |||||
| Morning Activity | +0.20% z+1.5 | +0.00% | +0.54% | +0.56% | +0.04% |
| Short-Sale Activity | -0.05% z-0.4 | +0.28% | +0.11% | +0.13% | -0.23% |
| Thematic | |||||
| Gold | +0.64% z+2.0 | +0.24% | -0.14% | +0.71% | -3.40% |
| Treasury (Duration) | +0.50% z+1.5 | -0.23% | -0.23% | +1.49% | +0.88% |
| Semiconductors | -0.70% z-1.2 | +0.51% | +0.95% | -3.62% | -5.33% |
| Oil | -0.44% z-0.9 | -0.26% | +0.73% | -0.59% | -2.02% |
| Bitcoin / Crypto | -0.18% z-0.6 | -0.05% | -0.22% | -1.16% | -1.10% |
| China | -0.18% z-0.4 | +0.11% | -0.38% | +0.99% | -3.24% |
5/20/60d windows include today's session; 1d is the previous session.
Pairwise realized correlation. Index variance net of the weighted idiosyncratic diagonal, over the cross terms — the estimator used for implied-correlation indices. 20-day window unless noted; history since 2020.
Market variance share. The fraction of cross-sectional variance explained by the market factor — the attribution companion to pairwise correlation, not a substitute for it.
Factor z-score. Today's factor return divided by its daily-return standard deviation over ~1,600 trading days.
Conditional importance. A factor's relative share of what a tree-based fit of today's stock returns explains; sums to 100% across factors.
High-beta momentum cohort. US names above $2B market cap in the top quintile of both raw beta and 12-month momentum; membership frozen at each month's first business day. “Top 50” = the 50 strongest by combined rank.
Stock-specific (idiosyncratic) return. The part of a stock's move not explained by market, style, or thematic factors in our decomposition.
VIXEQ / DSPX / COR1M. Cboe implied indices: single-stock volatility, dispersion, and 1-month implied correlation (top-50 S&P names). Quotes delayed ~15 minutes.
Data compiled by FactorPulse AI; edited and verified by Jeff Klein. For informational purposes only. Does not constitute financial advice, an investment recommendation, or an offer to buy or sell any securities. Always consult a qualified financial professional before making investment decisions.
For more on factor construction methodology, see www.factorpulse.com/glossary.