Most market commentary is a mood. Ours is a number with a name: the Triaxial Regime Hazard (TRH) and its P* — the desk's hazard read of a 5% or bigger SPY drop within 10 sessions, on closes; a hazard read, not a forecast — adjusted for how fragile leadership has become and read against whether informed money is already acting on it. The Council, the 9:15 desk read, both bulletins, the AB Portfolio model book and the home page all read from the same market context. The model book itself follows the market as it is — Maximum Adaptive Market Adherence (MAMA): where leadership stands today, how much risk the regime allows today, when the leading industry is in its own trend today — never a forecast of any of them.
Liquid names. Market adherence.
Large and mega caps plus liquid sector ETFs — the universe where closes and volumes carry information. Leadership is read against the index, never a thin story stock.
We describe the market. You decide.
The publication never tells you to buy or sell. It says what the regime is, what it implies, and what would change the read. You size positions.
Cited, not asserted.
Every AI rationale must cite the TRH verdict and P*. The three methods are published as research papers — TRH, SHC and MAMA — and the desk context is disclosed to Pro members.
Section 01
Hazard is a rate multiplied by fragility; the trigger tells you whether the market is already acting on it. Three axes, one posterior probability, one verdict.
Axis 1 · Calibration
The drawdown-risk ladder gives P₀, the calibrated base rate of a 5% or bigger SPY drop within 10 sessions, on closes, from realised history. It is a rate, not an opinion — and it is scored, so it can be wrong in public.
Axis 3 · Saturation → P*
Crowded leadership with nowhere left to rotate multiplies the base rate (up to 2.2×). TRH v1.1 adds macro fragility from FRED series and the scheduled event window (CPI, FOMC, NFP) as further multipliers. The result is P*, placed on a six-rung ladder from LOW to SEVERE.
Axis 2 · Defender (trigger)
Are hedgers, volatility sellers and the leaders-versus-sinks flows synchronising? Triggered or not, crossed with fragile or resilient, yields the four verdicts — and headline attribution explains what caused the trigger.
Section 02
Direction and danger are different axes. A market can be in an uptrend with an elevated drawdown hazard — the honest read says both, in the same sentence.
The trend read from the component pulses — where momentum, breadth and factor leadership actually sit.
The reconciliation with the TRH hazard: the desk verdict's label, P* and its rung, the event window.
What would change the read — the Level-Map invalidation or the next tier-1 release.
Above the ELEVATED line the word "constructive" is never used without the hazard qualifier in the same sentence.
Live on the home page
The hero terminal plots today's dot on this map and logs one point per session, so the regime path — not just the level — is visible at a glance.
Section 03
Sixteen instruments, one 0–100 scale. The same scale scores 23 sector and industry ETFs; only qualifying and neutral names are shown, each tagged with the session it entered its tier.
Section 04
Before September 2026 our services read different thermometers and could disagree. Now a single Desk Context is built from the TRH read and prepended to every AI prompt on the platform.
Soft by default
Every service receives the same block — verdict, conviction, P*, rung, trigger, fragility, macro, event window, level map, cycle phase and its alignment — and must cite the verdict and P* in its rationale. Sector and rotation signals are secondary to it.
Hard where it matters
Two conditions are enforced by the editorial system rather than by the prompt: a CONFIRMED RISK-OFF read at HIGH conviction, or SPY trading below the Level-Map invalidation. Under either, the illustrative model book adds no new non-defensive positions — the Council gate declines them and the desk read’s audit trail records it.
Section 05
The CAMFS cycle phase is the slow structural layer that sets portfolio composition. The TRH read is the fast tactical layer that gates new risk and sizing — and it now governs how the phase itself may move.
Asymmetric by design
Risk-off is fast — a stress read commits the defensive phase intraday. Relaxing is slow — one step at a time, after a five-session dwell and three confirming closes, and only with the TRH read's consent.
Divergence is published
When the phase and the TRH read differ, the Pro dashboard says so — ALIGNED or DIVERGENT — with the reason: confirmations pending, or TRH stress not yet reflected. The two are never presented as a contradiction.
Composition vs risk
Phase decides the pillar mix, beta budget and veto thresholds of the Trifecta and the AB Portfolio book. TRH gates new risk and sizing in the illustrative model book — including how large and how tight the stops sit.
Section 06
TRH reads the drawdown hazard; SHC reports whether one is under way; MAMA says what a rules book holds given both, in which universe, and when. Three reads in a fixed order — never a forecast of any of them — and a guardrail around the book between its scheduled cuts.
WHERE · leadership
Read cross-sectionally against the index — relative rotation and the desk consensus of three reads, taken weekly with volumes. Leadership is a property of industries before it is a property of stocks, so the book is tactical, in the leading industry, not diversified.
HOW MUCH · the regime
The TRH rung, the cycle phase and the SHC state size the sleeves: growth 85 % in CALM, 65 % in GUARDED, 50 % in DEFENSIVE, the defensive sleeve from 5 % to 20 %. The regime never vetoes a name — adaptivity lives in the sizes, where the reads are strongest.
WHEN · the leader’s trend
Inside the growth sleeve the leading industry is weighted by its own channel, not its rank: full in a trend, by third in a range, held at the floor, zero on a break — and a break is read between the cuts. Freed weight rests in bills.
The AB Defensive Guardrail
Between the scheduled cuts the consensus is read weekly, the antagonist sector’s breakout confirms or fails on volume, and when the Divergence Key’s three legs turn on one close, ten consensus names take the growth sleeve at equal weight until the next cut.
+8.4 % vs +8.6 %
Small caps vs large caps, a year
Russell 2000 against the S&P 500, 2000–2026, at 24 % volatility against 19 %. Why the book lives in liquid large and mega caps.
+10.5 % vs +9.1 %
Leading sectors vs all sectors
The three strongest sectors each month against all nine at equal weight, 2000–2026, with a shallower worst drawdown. Why leadership is followed.
−14.5% vs −50.8%
Worst drawdown, monthly closes, Oct 2005 to Oct 2026, hypothetical
compound annual return 9.1% vs SPY 11.2%, about two points a year lower
Hypothetical backtests on index and ETF data, gross of costs, not AB Portfolio results.
Research paper · TRH
Triaxial Regime Hazard
The hazard: P*, a calibrated hazard read scaled by fragility, macro and the event window, gated by an observed trigger. Not a forecast.
Read the paper →
Research paper · SHC
Suppressed Hazard Convergence
Whether one is under way: suppression capacity against price pressure — the touch that says a drawdown is realizing.
Read the paper →
Research paper · MAMA
Maximum Adaptive Market Adherence
The book: where leadership is, how much risk the regime allows, when the leader is in its own trend.
Read the paper →
Section 07
Four modules turn the hazard number into a desk read you can use with discipline.
Level Map
Support, resistance and the invalidation level for SPY and QQQ. The temporary-dips read holds only while the invalidation holds; below it, the read is suspended.
Event Risk Radar
Tier-1 releases (CPI, FOMC, NFP, PCE) define the event window — LIGHT, ELEVATED or HEAVY — which multiplies the hazard and shrinks new position size into the print.
Macro Fragility
Public FRED series — credit spreads, the curve, labour — add a slow-moving fragility multiplier so P* reflects the macro backdrop, not only the tape.
Event Attribution
When the Defender triggers, headline flows are matched to the move so the read names the cause, its confidence, and whether flows confirm it.
Section 08
The same read, in sequence, across the day — drafted by models, reviewed by editors, published with its citation.
08:30
Candidates
The universe is scored and the desk context is refreshed.
09:00
AI Council
Model votes on the Trifecta book, filtered by the TRH hard gate, journaled.
09:15
Desk read of the day
One instrument, regime-aware, with the TRH verdict and P* cited.
11:00
Intraday Council
Re-check against the live read; under the gate the model book takes no non-defensive adds.
Close
Nightly bulletin
Session wrap reconciled with the regime; the trail logs the day's point.
2nd · 16th
AB Portfolio book
12–17 lines cut on the close before each date under the desk regime and the three reads.
Elite gives you the daily and nightly bulletins and the AB Portfolio model book of 12 to 17 lines, cut under MAMA. Pro adds the live desk — the thermometer, the disclosed desk context, the Council trail and the phase alignment. Editorial commentary for self-directed investors, not personal advice.
Cookies & privacy. We use strictly-necessary storage to keep you signed in, and — only with your consent — functional and privacy-respecting analytics storage. We also use Google's tag to measure whether our own ads lead to signups (conversion measurement). We never sell your data.Necessary storage keeps you signed in; analytics and ad-measurement run only with your consent. We never sell your data. Cookie Policy · Privacy Policy