Hansen's February 2026 paper Physics-Based Forecasts of Investment Sentiment treats US equity sentiment as only partly a reaction to news. Its clock is a roughly 90° Sun–vertex angle between the Mercury–Saturn center, weighted by tidal force (
T ∝ m/r³), and the Jupiter–Neptune center, whose weights were fitted in December 2000 to Hansen's Market Resilience Indexes. Clusters of these configurations define Anxiety-Free Periods (AFPs), which typically fade when the final 90° column expires.
Geometry of the Inner Orbital Center (Mercury, Venus, Earth, Mars, Jupiter, Saturn) and the Outer Orbital Center (Jupiter, Saturn, Uranus, Neptune; Jupiter and Saturn sit in both centers): two mass-weighted planetary-group centers with the Sun at the vertex. When the angle is near 90° (Hansen uses 86°–94°), an AFP column is active, plotted as orange bars/columns in the US Stock Market Price Level chart below.
Boom, Bust, and the Working Rule
Across thirteen AFPs from 1933–2024, the average cumulative gain is about 28%, followed by a 13% post-peak decline—near the 98th and 95th percentiles of duration-matched history. Physics is proposed to time the turn; economics to size it.
Thirteen AFPs, named by the year of the price peak (1933-2024, exact dates in the chart above): geometry window = first 90° event through last 90° event; prior/post lows and returns are his market annotations (visually dated orange on the DJIA chart below). Average prior low is about 2½ months before the first event. The fourteenth cluster is the 2026 AFP (not in the 1933–2024 sample). Appendix U dates its 90° Fridays as Apr 24–May 22, 2026; Aug 07–Sep 18, 2026; and Nov 20–Dec 04, 2026. There is no completed peak/post-low for that episode yet.
A century of prices with orange AFP columns; column height represents intensity.
Each column is one 90° configuration event. A cluster of those bars is one Anxiety-Free Period.
2026 has three columns (late April–May, August–September, November–December).
Pre-peak ("boom") and post-peak ("bust") cumulative returns for the thirteen Anxiety-Free Periods since 1933, with empirical percentile ranks based on duration-matched periods in the full weekly Dow Jones Industrial Average history. The ranks measure the share of same-duration historical periods with less extreme returns. High percentiles show that gains consistently fall in the extreme tail, while declines do so in most cases.
From Orbital Angle to Standing Waves
Market impact is attributed not to the angle itself but to a four-step causal chain ending in Schumann-type modes (the Earth's ionosphere waveguide ringing at ~8, 14, 21 Hz). A 90° configuration is hypothesized to laminarize solar emissions, stabilize ionospheric height, and lock global standing-wave frequencies near 14–21 Hz; this coherence is then linked to analytical optimism, with overlap in the 13–30 Hz EEG band.
Hypothesized causal mechanism for Anxiety-Free Periods. The diagram proposes a causal chain from solar-system geometry through solar emissions, ionospheric stability, global electromagnetic standing waves, and human physiological entrainment. It is repeated at the start of each section, highlighting the relevant step; here, Element 1 is highlighted.
Which Links Are Physics, Which Are Claims
Stabilized solar output and ionospheric height extend accepted space-weather physics; the solar-geometry and cognition links remain speculative. Intensity scales with outer-group clustering, measured by Sun-to-outer-center distance, which is offered as the reason 1937 and 1987 were violent while 2006 was mild. Intensity is already represented by the column heights in the US Stock Market Price Level chart above.
Ap Index, F10.7, and Sunspots as Fellow Travelers
The same geometry used to time AFPs is associated with tighter co-movement among Ap Index, F10.7, sunspots, and Oulu neutron-monitor counts when the outer planets cluster. Geomagnetic and radio-flux series are treated as companions of standing-wave stability rather than as separate daily trading signals.
Correlations of Solar Energy Metrics by High Intensity and Low Intensity Weeks.
Ap, F10.7, sunspots, and Oulu co-move more tightly when the outer group is clustered.
Frozen Geometry versus 14-Week RSI
An eight-driver orbital model—six frozen orbital-geometry factors (planet-group angles, distances, and clusters), plus the predicted AFP series and predicted M-Spike series —was calibrated through late 2000 and then frozen. It forecasts 14-week RSI (Wilder’s Relative Strength Index on weekly bars with a 14-week lookback—Hansen’s proxy for multi-month price momentum/investor sentiment, not the usual 14-day RSI) from December 2000 through April 2024 using NASA coordinates alone. The M-Spike is a short orbital-geometry factor—not RSI and not an AFP—computed from planet-group positions and read on DJIA/S&P price as a multi-week M: lift, mid-dip, rebound, fade. About 88% of predicted turns match actual turns within ±1 week, versus a chance baseline near 0.25% (
p < 0.001).
Frozen-model path versus realized RSI and log price; the 88% out-of-sample test begins here.
When Price Rises but Sentiment Does Not
When actual RSI runs below the geometry path, as in 2007–08, Hansen interprets the divergence as economically driven pessimism despite rising prices. Pre- and post-event weekly returns around 26 individual 90° events differ at
p < 0.0001.
Predicted and Actual 14-week RSI, Periods A and B. Arrow A marks the 2007–08
actual-RSI divergence below geometry; B marks an AFP in which both rise together.
The Thin Schumann Test
Usable standing-wave data exist only from June 2015 through December 2021. However, it is not a published Schumann-resonance product. Hansen built the series himself from British Geological Survey magnetometer records at Eskdalemuir, Scotland. Within that window, the predicted 2017 AFP series is associated with frequency stability in modes 2–3 (
p < 0.001), although R² is only about 0.05—the most direct test of standing-wave market impact, but also the thinnest.
US stock-market price and indicators for the 2017 Anxiety-Free Period and M-Spike episodes, June 05, 2015–December 31, 2021. The chart overlays market price, episode indicators, and ellipses A–D marking notable relationships between price patterns and episodic signals. It also shows modes 2 and 3 electromagnetic standing-wave stability, measured on weekends from 8 a.m. to 4 p.m. local time and lagged approximately two weeks.
Sunspots as the Longer Proxy
Solar activity is more stable during 21 of 26 90° events and more often declines by day +21 after an event ends. Shorter geometry-defined M-Spikes (53 cases, 1940–2023) show weaker weekly-return significance (
p ≈ 0.19).Mega Cycle and Phase 1 Valuations
The same inner–outer angle maps onto a four-phase Mega Sentiment Cycle. Phase 1, from 0° to the AFP at 90°, has accounted for about 32% of weeks since 1942 but 53% of returns, with higher S&P 500 P/E ratios but no corresponding growth differential.
Mega Sentiment Cycle (2004–2027), showing its four phases, the timing of Anxiety-Free Periods, and the changing Inner Center–Sun–Outer Center angle. "US stocks" and "US stock market" refer to the DJIA, S&P 500, or related ETFs,
How Panic Is Supposed to Spread
A Geoffrey Moore-style adoption model divides the market into Pioneers (~15%), Early Majority (~35%), Late Majority (~35%), and Laggards (~15%). Micro geometry is said to move the Pioneers, while the Mega cycle moves the slower half. During Phase 1, roughly half the market remains optimistic, so valuation alone is judged unlikely to force a crash larger than 30%.
The 2026 Live Overlap
Phase 1 of Hansen's Mega Sentiment Cycle—the long inner–outer angle from 0° up to the next AFP at 90°—runs from autumn 2022 until the first 90° column of 2026, projected for late April–May, with peak geometry-driven euphoria in late May, a second lift by early August, and a third close in early December. A moderate M-Spike overlaps the first column, making April–June the unstable hinge.
The chart (December 06, 2024–January 01, 2027) shows the US stock-market price level (log) through July 18, 2025—the analysis date—along with the drivers of the three 90° configuration events comprising the 2026 Anxiety-Free Period and the M-Spike as of June 2025. Forward map showing the three 2026 AFP columns, overlapping M-Spike, May peak, and June hinge. "US stocks"and "US stock market" refer to the DJIA, S&P 500, or related ETFs, depending on the period.
2026: Three AFP Windows, Not a Continuous Year-End Rally
Hansen's sketch is: enter several weeks before an AFP, hold through the cluster high, exit on the Micro Market Resilience Index; he prefers equal-weight S&P 500 or the DJIA as cleaner sentiment gauges. That is a multi-month calendar, not a daily trigger and not proof that high valuations cannot break.
2026 is one such labeled window: three separate 90° columns, not a continuous anxiety-free bid into year-end. The first and strongest was late April–May (peak late May). An overlapping M-Spike made April–June the hinge—historically a dip zone, not a melt-up. A weaker second lift was dated August–September. The third column, late November–early December, ends the cluster; Hansen's average script is a boom during the bars and about −13% after the peak. 2006 barely rose, so an AFP is not a guarantee of higher prices. November–December is the close of the episode, not "anxiety-free until Christmas."
Use that map to watch whether price, Ap, sunspots, and standing-wave stability follow the boom-then-fade pattern. For next-week risk keep short-horizon Ap/Kp and real-time L1 solar wind—those still have skill. The Schumann link rests on one AFP and ; 2026 is the live joint test of cavity-mode coherence, solar-metric co-movement, and that price script. Do not size as if the mechanism were already proven. Hansen's paper is research, not investment advice.
Reference:
Hansen's Data Sources:Ap Index (Kp/ap): GFZ Potsdam is the official source. Combined daily file (Kp, ap, Ap, sunspot number, F10.7 since 1932): https://kp.gfz-potsdam.de/ Direct text: https://kp.gfz-potsdam.de/app/files/Kp_ap_Ap_SN_F107_since_1932.txt
Nowcast (last ~30 days): https://kp.gfz-potsdam.de/app/files/Kp_ap_Ap_SN_F107_nowcast.txt Index page: https://www.gfz.de/en/kp-index
DOI: https://doi.org/10.5880/Kp.0001
F10.7 cm Radio Flux: Observed at Penticton (NRC Canada); NOAA/SWPC redistribute it. Phenomenon note: https://www.swpc.noaa.gov/phenomena/f107-cm-radio-emissions 45-day Ap + F10.7 forecast: https://www.spaceweather.gov/products/45-day-forecast 27-day F10.7 + geomagnetic outlook: https://www.spaceweather.gov/products/27-day-outlook-107-cm-radio-flux-and-geomagnetic-indices NCEI Penticton listings (observed): https://www.ngdc.noaa.gov/stp/space-weather/solar-data/solar-features/solar-radio/noontime-flux/penticton/penticton_observed/listings/listing_drao_noontime-flux-observed_daily.txt Adjusted: https://www.ngdc.noaa.gov/stp/space-weather/solar-data/solar-features/solar-radio/noontime-flux/penticton/penticton_adjusted/listings/listing_drao_noontime-flux-adjusted_daily.txt LISIRD Penticton: https://lasp.colorado.edu/lisird/data/penticton_radio_flux/
The GFZ combined file above also carries daily F10.7.
Sunspots: WDC-SILSO, Royal Observatory of Belgium (International Sunspot Number v2). Download hub: https://www.sidc.be/SILSO/datafiles Cite: https://doi.org/10.24414/qnza-ac80 Daily SN is also in the GFZ combined file (column SN).
Oulu Neutron Monitor: University of Oulu / Sodankylä Geophysical Observatory. Record from April 1964; pressure- and efficiency-corrected series is what Hansen uses. Station and query: https://cosmicrays.oulu.fi/ WDC-CR station card: https://cidas.isee.nagoya-u.ac.jp/WDCCR/station_info1.php?sta=OULU
For a single-file start matching Hansen's Study B metrics, pull the GFZ Kp_ap_Ap_SN_F107_since_1932.txt and add Oulu separately from cosmicrays.oulu.fi.
Hansen Anxiety-Free Periods 1933-2051
Two "centers of gravity" among the planets: one is a weighted mix of Mercury, Venus, Earth, Jupiter, and Saturn; the other is a weighted mix of Jupiter, Saturn, Uranus, and Neptune. From the Sun a line to each center (see first diagram above). Hansen's AFP clock is simply how close the opening angle is to a right angle. He counts a "column" whenever it falls between 86° and 94°. Because the inner planets move quickly, the angle wobbles in and out of the band, so one AFP is usually a handful of orange bars a few months apart, not one solid year. To calculate Hansen's Anxiety-Free Periods through 2051, his published weights were used, with NASA/JPL providing the planetary positions.
There are 16 clusters from 1935 to 2051. After the current 2026 AFP, the next full cluster is November 2032–July 2033. Then comes a long empty stretch: the angle falls toward 0° (the two centers lined up) and does not re-enter the band until a weak July 2050 graze and a full 90° column in January 2051. Gaps with no orange bars are 2027–31 and 2034–49.
Hansen does not publish the M-Spike formula and it cannot be fully reengineered. In his account, the M-Spike is the AFP's short-horizon companion: a few weeks of up–down–up that can flip conviction as a 90° column ends. Applied to his 2026 sequence, it is the mechanism he uses to frame the April–June transition—from the first 90° column’s decline into the M-Spike—as the unstable hinge; it is not a second melt-up or a tested trading rule. His own weekly-return test yields only p ≈ 0.19, making the episode central to the story but the weakest published link.
What exists in print is Appendix U of Hansen's paper: From January 11, 2019 through May 29, 2026, every V-Friday—the Friday at the bottom of the M's dip—falls exactly 55 weeks (385 days) after the previous one, a cadence visible in the table but never stated in the prose. That permits projection of candidate V-dates through 2051, but not which ticks Hansen would skip (historically, about one in three) or the ±0.3/−0.5 heights. For 2026, the only signed weeks he prints are May 29 (−0.44), June 5 (−0.43), and June 12–19 (+0.26), all on the first column's down-leg. After June 19, 2026, the dates are projections from the 55-week cadence, not Hansen's forecast.
See also: