Macro-aware, price-led tactical allocation.

The system is built for one decision: where should capital go for the next four-week tactical window? It uses macro as a current, not a prophecy. Price and breadth lead, macro confirms or de-risks, and the final allocation follows fixed rules. The long-term reasons for this universe live on the Categories page; this page is the operating manual.

Website vs Weekly Report

The website is the evergreen manual: theses, universe design, framework, category methods, audits, and performance. The weekly report is the current evidence packet: this week's macro read, crypto/Defensive state, category choices, ETF representatives, allocation table, trade instructions, and charts. The goal is to keep every weekly report focused on new evidence while the site carries the durable explanation of the process.

The Six-Decision Chain

The edge is not supposed to come from one magic indicator. The system makes six sequential decisions and lets those edges compound over time. First, only ETFs with pure category fit, liquidity, coverage, and low redundancy enter the universe. Second, the macro engine determines the tape and the burden of proof. Third, the crypto state machine chooses Normal, TrendBTC, AltSeason, or cause-matched Defensive allocation. Fourth, the macro regime quality gates the allocation aggressiveness. Fifth, the category engine chooses which themes deserve overweight capital. Sixth, the execution engine chooses the ETF inside the category with the best four-week forward setup. If those decisions are directionally right often enough, the portfolio does not need perfection in any single layer.

Portfolio Construction

In normal mode, the ten categories are sized by rank. The top-2 eligible categories receive 20% each. Ranks 3–8 receive 10% each. Ranks 9–10 receive 0%. Total: 20 + 20 + 10 + 10 + 10 + 10 + 10 + 10 + 0 + 0 = 100%.

When Defensive, Bitcoin, or AltSeason is active, the overlay receives exactly 50% and the category sleeve weights are halved: top-2 get 10% each, ranks 3–8 get 5% each, ranks 9–10 get 0%. Total: 50 + 10 + 10 + 5 + 5 + 5 + 5 + 5 + 5 + 0 + 0 = 100%.

The investable portfolio is now measured as a rolling four-week ladder. Each Friday report creates a 25% tranche, bought at the next Monday open and held for four weeks. The newest tranche replaces the tranche from four weeks earlier. This is why the Portfolio and Performance pages emphasize four-week completed-tranche results instead of one-week noise.

Weekly Decision Sequence

Every run follows the same order. First, the macro engine classifies the backdrop and decides whether a Defensive warning applies. Second, the crypto engine determines NoCrypto, ValueBTC, TrendBTC, or AltSeason using deterministic Bitcoin and macro-risk rules. Third, the overlay priority is resolved: confirmed crypto-cycle exposure comes first, then cause-matched defense if crypto is NoCrypto, then normal category allocation. Fourth, every category is scored through its active macro-condition method. Fifth, one representative ticker is selected inside each category. Sixth, the top two eligible categories receive the overweight sleeves. The AI/reasoning layer explains those decisions, but it does not alter the state machine or override hard risk rules.

Crypto Cycle Overlay

Bitcoin and altcoin exposure are rules-based. ValueBTC and TrendBTC require Bitcoin trend repair and confirmation. AltSeason is stricter: crypto trend, liquidity, sentiment, dollar, credit, and risk appetite must all cooperate. If broad macro risk is elevated while crypto is confirmed, the model can block AltSeason or force the overlay to stay in Bitcoin. The slow Defensive trigger does not replace confirmed Bitcoin-cycle exposure.

Defensive Overlay

Defensive is a 50% overlay, not a permanent asset class and not a discretionary opinion. The edge is that the system asks what benefits from the same macro force that is hurting broad risk assets. A liquidity panic is not the same as an inflation drawdown; a stagflation scare is not the same as a disinflationary slowdown. When active, the remaining 50% still follows the category engine with 10%, 10%, and 5% x 6 across category representatives.

The trigger is broad-market defense: crisis-level macro risk or at least three of five broad-market checks, covering SPY trend damage, QQQ trend damage, HYG/SPY credit breakdown, rising dollar pressure, and weak risk appetite. This is meant to catch persistent bear markets and extended drawdowns. Defensive only takes the overlay when crypto is NoCrypto; confirmed Bitcoin-cycle exposure wins over slow macro deterioration.

The payload is chosen by a cause selector, not by the broad regime label alone. Liquidity stress uses SGOV because cash-like collateral usually wins when markets are forced to de-risk. Inflation/scarcity stress uses XLE because energy cash flows can benefit from the same pressure hurting broad equities. Monetary or disinflation stress uses GLD because falling real-yield pressure and currency hedging matter more than cyclical growth. Stagflation scarcity splits GLD and XLE. Unclear transition stress uses SGOV, GLD, and XLU so the system avoids pretending one cause is obvious when the tape is mixed.

This means Defensive is cause-matched, not label-matched. The same Stagflation Risk headline can come from monetary stress, inflation scarcity, or mixed transition pressure. The payload is selected from the stress cause detected by the macro and market-implied checks, then the remaining 50% of capital still follows the category system.

Weekly Operating Instructions

1. Read the new Friday report after it is published and treat it as the instruction set for the next Monday open.
2. On Monday, sell the tranche created by the report five Fridays earlier. That tranche has completed its four-week Monday-open-to-Monday-open holding window.
3. Allocate that freed 25% tranche into the new report's allocation table at the Monday open.
4. Leave the three newer tranches unchanged. The live portfolio is always the blend of the newest four report tranches.
5. Repeat every week unless the report is marked unreliable or a data-quality warning says the allocation should not be changed automatically.

Example: a Friday report dated June 12 becomes the Monday June 15 buy. At that same Monday open, the tranche from the Friday report five weeks earlier is sold and replaced by the new June 12 allocation. This is why the public scorecard measures four-week completed tranches rather than one-week noise.

How Categories Are Chosen

Category selection is not supposed to be a single-ticker contest. Each category is scored from exactly three liquid ETFs chosen for pure category fit, strong liquidity, broad internal coverage, and minimal redundancy. The first category evidence point is a 3/2/1 weighted ETF basket: first place receives three parts, second receives two parts, and third receives one part. That is only the starting evidence. The final category score is now produced by a descriptor reasoner that reads technical evidence, the current macro state, and the active true/false macro detail checklist together.

This is the most important control in the system. A category favored by the macro story still needs confirmation from its own ETFs. A category fighting the macro backdrop can still win, but only when the tape is clearly sponsoring it through volume, relative strength, persistence, timing, and risk/reward. Commodity, cyclical, and dollar-sensitive categories such as Traditional Energy, Industrial Metals, Agriculture, Nuclear Energy, and Emerging Markets are not allowed to win just because they look cheap; they need evidence that the relevant macro cause is active and buyers are actually sponsoring the move.

Future Theses

The ten-category universe was not assembled arbitrarily. Each category carries a structural thesis—a reason to believe it will compound capital over a full economic cycle, regardless of which macro regime is active today.

Commodity supercycle. Commodities spent roughly two decades in secular underinvestment following the 2011 peak. Capital expenditure collapsed, mines closed, and supply chains contracted. The result is a structural deficit in industrial metals, energy, and agricultural capacity that monetary policy cannot fix. When demand recovers—from AI power requirements, electrification, and reshoring manufacturing—supply cannot respond quickly. Copper, uranium, silver, agricultural land, and energy infrastructure are the scarcest inputs in the next decade of growth. The universe includes COPX, REMX, URA, PDBA, MOO, GLD, GDX, XLE, and FCG precisely because these categories express the commodity thesis across multiple sub-cycles.

Cheap prices on a ratio basis. The commodity-to-equity ratio and the commodity-to-gold ratio both sit near multi-decade lows. Real assets are historically cheap relative to financial assets. This is not a short-term trade setup; it is a structural rotation that typically takes years to resolve. The system is positioned to capture this mean reversion as it unfolds, week by week, through the category ranking process. Categories that are cheap on a ratio basis score better when macro conditions confirm sponsorship.

Owning the AI stack. Artificial intelligence is not a software story; it is a hardware and infrastructure story. The models require semiconductors, data centers, power, and cooling. The investable expression of this thesis is not a single stock but an entire supply chain: semiconductor capital equipment, specialized chip design, AI-adjacent software, and the physical infrastructure that hosts the compute. SMH, AIQ, BOTZ, and CIBR represent different layers of this stack—AI compute, broad AI exposure, robotics and physical AI, and the cybersecurity layer that becomes more critical as AI attack surfaces expand.

Deglobalization and independence. The post-1990 era of frictionless global trade is ending. Tariffs, friend-shoring, reshoring, and industrial policy are restructuring supply chains on a decade-long timeline. Countries and companies are prioritizing domestic production over efficiency. This benefits defense contractors, domestic industrial firms, and infrastructure builders. ITA, XAR, PAVE, and IGF express this thesis. Defense spending is rising across NATO, Asia-Pacific, and the Middle East simultaneously—a rare alignment that creates durable earnings visibility for aerospace and defense.

Geopolitical strife and scarce-resource competition. Resource scarcity is increasingly a geopolitical problem. Lithium, cobalt, uranium, rare earth elements, and agricultural land are concentrated in politically unstable or adversarial regions. The developed world is actively trying to secure alternative sources. This creates a premium for domestic or allied-nation resource producers that the market is still pricing inefficiently. REMX, URA, COPX, XLE, and GDX all benefit from the geopolitical dimension of resource scarcity—not just from commodity demand, but from the strategic value of reliable supply.

These theses are not used as allocation signals. The system does not buy copper because of the supercycle thesis; it buys copper when the chart and macro score confirm that buyers are actually showing up. The theses exist to define the universe: they explain why these ten categories deserve a permanent seat at the table rather than being rotated in and out based on narrative. A category with a structural thesis behind it is more likely to produce leadership when its conditions are met.

Macro Methods

The practical macro framework is intentionally compact. The model does not need dozens of investable macro states. It uses a small set of regimes: risk-on liquidity expansion or Goldilocks, reflation or late-cycle reflation, disinflation or slowdown, stagflation risk, risk-off/high stress, and mixed transition. Beneath that label, the model creates a macro detail checklist: supply shortage, liquidity stress, credit stress, dollar pressure, risk appetite, commodity breadth, monetary hedge demand, defensive rotation, AI sponsorship, emerging-market liquidity support, and similar true/false descriptors. Those descriptors tell the reasoner what kind of opportunity the market should be rewarding. Risk-off regimes can trigger the 50% Defensive overlay. Defensive is intentionally rare: it requires crisis-level macro risk or a bear-defense confirmation where at least three of five checks fire: SPY below its 40-week average or down more than 8% over 13 weeks, QQQ below its 40-week average or down more than 10% over 13 weeks, HYG/SPY credit breakdown, rising dollar pressure, and risk appetite below 45. Once active, the cause selector chooses the defensive payload. Defensive does not override confirmed crypto-cycle exposure.

Macro is factored at two levels. At the category level, it gives the reasoner context for whether the category should matter right now. At the asset level, it helps decide which of the three ETFs is the best expression of the current tape. The same indicator does not mean the same thing in AI, gold, Traditional Energy, Nuclear Energy, Emerging Markets, utilities, and defense. Macro is not allowed to invent strength: the category still needs price, breadth, and volume confirmation.

The macro stance is intentionally strict. Favored means the macro and narrative backdrop are aligned, but the category still needs confirmation. Neutral means the category gets no story credit and must win on price, volume, breadth, and relative strength. Headwind means the macro backdrop is working against the category; in that case the reasoner demands exceptional sponsorship. This is why the method is not a generic momentum screen and not a subjective macro opinion.

The Categories page lists the explicit reasoning map: Category, Macro Condition, Method, and Reasoning. That table is the plain-English rulebook. It keeps the system from feeling discretionary while still acknowledging that a good setup in AI, gold, Traditional Energy, utilities, or defense should not be graded as if they were the same asset.

How Asset Winners Are Chosen

The execution ticker inside a category is chosen with the same descriptor reasoning method used at the category level, but applied to the three ETFs inside that category. The model looks for price behavior that is actually being sponsored by buyers: persistent 13-week and 26-week relative strength, strength versus SPY, strength versus category peers, price above important moving averages, MACD confirmation, constructive stochastic RSI, clean support/resistance, and volume that confirms the price move. Thin-volume rallies, distribution weeks, unsupported bottom-fishing, bearish MACD, and overextended rollovers are penalized.

The active macro state and descriptor checklist change the asset interpretation. In risk-on liquidity tapes, the reasoner wants leadership, momentum, and volume-backed breakouts. In reflation, it wants volume-backed commodity/cyclical sponsorship. In disinflation or slowdown, it wants quality pullbacks, support defense, and risk/reward. In risk-off, it wants defensive relative strength and lower failure risk. The report identifies the active reasoning method in each category so the reader can see why the winner was selected.

No ETF receives a permanent preference inside its category. The three ETFs begin with equal eligibility. Macro then defines the active cause, and technical evidence decides whether the market is actually confirming that expression. A narrow ETF can win, but only when its macro cause is active or when volume, relative strength, and momentum sponsorship are strong enough to justify the risk.

Traditional Energy is the clearest example. XLE is the integrated-major cash-flow expression, XOP is exploration and production beta, and FCG is the natural gas and LNG transition expression. In risk-on or reflation tapes, XOP can win if sponsorship is strong. In gas/scarcity tapes, FCG can win if gas leadership is real. In slowdown, risk-off, or mixed tapes, XLE usually has the cleaner economic profile unless the narrower ETF proves otherwise. Emerging Markets works the same way: IEMG is broad diversified beta, INDA is India quality-growth exposure, and ILF is Latin America commodity/currency reflation beta. None is the default; each has a different burden of proof.

The category itself now changes the asset formula too. AI and Technology reward leadership, sponsorship, and volume-backed momentum. Utilities and Defense reward defensive relative strength, trend persistence, and failure avoidance. Traditional Energy, Industrial Metals, Precious Metals, Nuclear Energy, Agriculture, and Emerging Markets require more respect for cyclicality, dollar pressure, commodity noise, extension, and support/retest behavior. In those sleeves, the model gives more weight to volume in relation to price, MACD/stochastic confirmation, risk/reward, and whether a breakout is actually confirmed rather than merely exciting. This is intentional: the same chart setup should not mean the same thing in a uranium-miner ETF, a gold bullion ETF, a broad emerging-markets ETF, a mega-cap software ETF, and a regulated utility ETF.

Volume in relation to price is a primary audit item. A rally on improving volume, constructive MACD, and rising relative strength is treated very differently from a bounce into resistance on thin participation. A pullback into support can be attractive, but only when the broader basket and the selected ETF show evidence of sponsorship. Support by itself is not a reason to allocate a 20% sleeve.

How The Audits Should Be Read

Category Audit asks whether the two overweighted category representatives beat the other eight representatives. Asset Audit asks whether the selected category winner beat its own three-ETF basket and the other two ETFs inside its category. The main scorecard is the four-week result: whether the same Friday decision worked for an investor holding from the Monday after the report to the Monday open four weeks later.