Kanseitō

Factor investing

Sun, 05 Jul 2026 00:00:00 GMT

Nobody needs to pick stocks to hold a structural edge over the plain index. That is the practical promise of factor investing, and it comes with two honest price tags: droughts measured in years, and the permanent risk that a habit stops being a habit.

The problem

Stock-picking, for most participants, is luck wearing analysis as a costume — the evidence on professional managers beating their benchmarks after fees is famously bleak. Yet the pure index is not the only alternative. Decades of academic work, beginning with Fama and French, established that certain measurable traits of stocks have predicted better-than-average returns across markets, across countries, and across a century of data.

The question this chapter answers: which traits are real, why do they pay, and where does the machine break?

The insight

A useful analogy: football teams that dominate possession — a measurable trait — win more on average over a season, though not in every match. A factor is the same structure in markets: a trait that pays statistically, over long horizons, never reliably in any given year. Four have survived the most scrutiny.

Value — cheap beats expensive over time, because investors systematically overpay for exciting stories and underpay for dull ones.
Momentum — the past year’s winners keep winning for three to twelve months, because information diffuses slowly and investors under-react, then herd.
Quality — profitable, low-debt, stable firms outperform on a risk-adjusted basis, because the persistence of high returns on capital is chronically underpriced.
Low volatility — calmer stocks beat wild ones risk-adjusted, directly contradicting the textbook claim that more risk earns more return; the standard explanation is that constrained institutions chase exciting high-volatility names and leave the boring workhorses chronically cheap.

A fifth, size — small beats large — sits in the literature with an asterisk: the premium has faded since publication.

The most instructive relationship in the set is the value–momentum tension. Value buys what has fallen and become cheap; momentum buys what has risen and keeps rising. They are philosophical opposites and, empirically, negatively correlated — when one suffers, the other tends to work. Asness, Moskowitz and Pedersen documented this everywhere they looked: the pair works better together than either does alone, a live demonstration of the Markowitz principle applied not to assets but to styles.

The tension is not a flaw to resolve; it is the diversification.

In plain English

Factors are the market’s recurring habits — patterns of human over- and under-reaction stable enough to harvest by rule. The harvesting instrument is ordinary: broad, cheap, rules-based funds that tilt toward the chosen traits. For a first allocator, a tilt toward quality and low volatility is the most forgiving entry: both lean toward the survivable end of the market, and both align with the compounding arithmetic from the start of this part — the calm compounder beats the exciting average.

Three rules keep the harvest honest.

Hold factors together, not one at a time, because any single factor can disappoint for half a decade.
Never chase the factor that just won — that converts a factor strategy back into momentum-chasing with extra fees.
Treat factor exposure as a tilt on a diversified core, not a replacement for it.

Where this breaks

Every factor endures droughts long enough to shake out most of its holders — value underperformed for roughly the decade after 2010 before reviving. Momentum carries a sharper failure mode: the momentum crash, in which an abrupt regime reversal makes last year’s winners the new losers in weeks, giving back years of gains — the 2009 recovery being the canonical episode. Factor investing pays patience measured in years and punishes any evaluation window shorter than that.

Two deeper limits deserve naming. First, an asset with no fundamentals has no factor structure: a claim with no earnings, no book value, and no balance sheet offers nothing for value or quality to measure, leaving momentum — the most crash-prone factor — as its only habit. In factor terms, such an asset is a pure momentum bet with no ballast, which is a precise, first-principles statement of why it whipsaws; it can be held as a labelled, losable sliver, but it is not “investing” in the factor sense. Second, factors assume the rules of the game are stable, and a regulator can change the rules mid-match: a reclassification, a ban, or a new capital rule is a one-time discontinuity that no habit-based premium survives, and no factor model has a slot for it.

There is also the crowding question, which connects forward to the positioning chapter: a habit known to everyone and harvested by everyone pays less, and unwinds harder when the crowd exits at once. The published factor premiums are almost certainly larger than the future ones.

The action: audit the current portfolio for its accidental factor bets. Most portfolios are already factor portfolios — concentrated in yield, or in glamour, or in one style — without their holders ever having chosen it. Better to hold the habits deliberately, together, and at a size whose worst decade is survivable.

Kelly and fractional sizing

Sun, 05 Jul 2026 00:00:00 GMT

A genuine, positive edge plus an over-sized bet equals eventual ruin. That single sentence is the reason position sizing is a separate discipline from asset selection, and the Kelly criterion is the mathematics that proves it.

The problem

Every allocator eventually faces the question the previous two chapters cannot answer: given an attractive opportunity, how much? Markowitz says how to mix; the compounding arithmetic says why losses hurt disproportionately; neither names a number. Intuition fills the gap badly — conviction inflates size, and humans systematically overestimate their edge.

The stakes are asymmetric in a specific way. Bet too small and the cost is mild: slower growth. Bet too big and the cost is terminal: one ordinary losing streak takes a bite that compounding can never grow back. Sizing is therefore not a detail of a strategy; it is the difference between a strategy surviving and not existing.

The insight

John Kelly, working at Bell Labs in 1956, showed that for a repeated favourable bet there is a single fraction of capital that maximises long-run compound growth. The formula is friendlier than its reputation: bet the win-probability minus the loss-probability divided by the payout ratio.

A worked example makes it concrete. Take a favourable coin: it wins 60% of the time, and a win pays twice the amount risked. Kelly says bet 0.6 minus 0.4-divided-by-2, which is 0.4 — risk 40% of capital each round. That 40% is “full Kelly.” Half-Kelly is 20%, quarter-Kelly is 10%, and all-in is 100%. Hold those four numbers.

Now run twenty rounds of this genuinely favourable game, starting from any stake. All-in survives only if all twenty flips win: 0.6 to the twentieth power, roughly four chances in a hundred thousand. Near-certain ruin, despite a real edge — because a single loss at 100% sizing is fatal, and the edge only pays those still in the game. Full Kelly grows fastest on paper but with gut-churning swings; a couple of early losses can halve the stake. Half-Kelly compounds into multiples of the start in a representative run, with far shallower drawdowns. Quarter-Kelly is slower and calmest, almost never seriously dented.

The practitioner standard is half or quarter Kelly, and the reason is not timidity but epistemology. Full Kelly is optimal only when the edge is known exactly, and in markets the edge is always an estimate — usually an optimistic one. Betting full Kelly on an overestimated edge is betting over Kelly on the true one, and over-Kelly sizing does not merely add volatility; it turns a winning game into a losing one. The growth-versus-size curve is asymmetric: to the left of the optimum it slopes gently, to the right it falls off a cliff. Fractional sizing is the margin of safety for not knowing the true edge.

In plain English

Direction and size are two separate decisions, and the second one is where survival lives.

An allocator can be right about the idea and still go broke on the sizing — right-and-ruined is a standard failure mode, not an exotic one.

The honest default size is small, precisely because confidence in the edge is the least reliable input in the whole calculation.

Applied without any formula, Kelly thinking becomes three habits.

Survivable loss. Size every position so that its total loss would be survivable without changing any life plans.
Scale to evidence, not excitement. An uncertain edge gets quarter-Kelly treatment, which for most real opportunities means single-digit portfolio percentages.
Zero is an acceptable outcome. Cap speculative positions at a level where zero is survivable, because for the most uncertain bets, zero is a scenario, not a hypothetical.

The same arithmetic operates at every altitude. Institutions and even nations blow up the same way: a real economic engine, levered too hard, destroyed by an ordinary bad stretch it could not survive at that size. Over-betting a genuine edge is one of the oldest failure patterns in finance precisely because the edge is real — the post-mortem always shows the idea was right.

Where this breaks

Kelly’s assumptions deserve to be stated plainly, because markets violate most of them. The formula assumes known, stable probabilities and payouts; markets offer estimates that drift. It assumes sequential, independent bets; portfolios hold simultaneous, correlated positions, which shrinks the safe aggregate size further. It assumes the capital pool is the whole story; real allocators have external liabilities and time horizons that can force exits at the worst moment.

Kelly also optimises growth, not comfort — even half-Kelly drawdowns exceed what most people can psychologically hold, and a plan abandoned mid-drawdown performs worse than a smaller plan followed. The formula’s real gift is not the number; it is the shape of the curve — the proof that over-sizing is always the expensive side of the error.

The discipline layer later in this curriculum turns this chapter into standing policy: evidence builds, size follows, and no single position is ever allowed to be fatal. The immediate action is simpler. For every current holding, ask one question: if this went to zero, would the plan survive? Any position where the answer is no is over Kelly — whatever the conviction behind it.

Liquidity cycles and the Singapore translation

Sun, 05 Jul 2026 00:00:00 GMT

Before judging any asset, ask whether global money is expanding or contracting.

In a drain, cash and quality win and speculation is punished, regardless of how good the individual story is.

That is the working claim of Michael Howell’s liquidity framework, and it comes with one important local translation for anyone whose base currency is the Singapore dollar.

The problem

Most investors analyse companies; far fewer analyse the water the companies float in. Yet in any given year, a large share of a market’s move is the tide, not the boats. Illustratively, in a strong liquidity expansion, if underlying businesses grow earnings around 6% a year while their share prices rise around 16%, roughly ten of those sixteen points came from investors paying more for the same earnings — the tide — and only six from the businesses themselves. Attributing the whole move to stock-picking skill is the most common self-deception in a bull market.

The problem, then, is measurement: what exactly is “the tide,” and can it be read before it turns?

The insight

Howell’s answer is to measure total flow through the financial plumbing and call it global liquidity. Three components feed it.

Central-bank balance sheets. Asset purchases inject reserves; balance-sheet run-off drains them.
Bank credit creation — the under-appreciated one, because most money is not printed by any central bank; it is lent into existence by commercial banks, and when credit tightens, total liquidity falls even if the central bank does nothing.
Cross-border flows. Reserve accumulation by surplus countries exports liquidity; reserve drawdowns to defend currencies remove it.

A publicly available proxy exists for the largest bloc: net liquidity approximately equals the central bank’s balance sheet, minus the government’s cash account held at the central bank, minus the overnight reverse-repo balance — the last two being pools that drain reserves away from markets. Howell’s empirical claim is that changes in this measure lead risk-asset prices by roughly twelve months, because liquidity first expands or contracts the balance-sheet capacity of the dealer banks standing in the middle of every market, and only later shows up in prices.

The first principle underneath is that money is credit — book entries that a system creates and destroys — and prices are set at the margin by credit-funded buyers. Earnings belong to the real economy; the willingness to pay a multiple for those earnings belongs to the credit cycle. Howell measures the switch that turns that willingness on and off.

The Singapore translation

Singapore requires reading the same tide through a different instrument. The US central bank steers through interest rates; the Monetary Authority of Singapore steers through the exchange rate — managing the Singapore dollar against an undisclosed trade-weighted basket of currencies, adjusting the slope, width, and centre of its policy band. For a small economy where trade dwarfs domestic output, the exchange rate is simply the more powerful lever, and local interest rates largely follow global ones as a consequence.

Two practical readings follow. First, a strengthening Singapore dollar on a steeper appreciation path is MAS tightening — a local tell that the authority sees inflation pressure, often rhyming with a global liquidity drain. Second, a portfolio split between SGD and USD assets is exposed to the same global tide through two different doors: a US tightening cycle hits USD assets directly through rates and multiples, while it reaches Singapore through trade, credit conditions, and the currency. An SGD-based allocator holding a large USD sleeve is implicitly short their own home currency — a position most holders never chose consciously, and one that a strengthening SGD quietly taxes.

In plain English

Watch two dials, both free. Globally: is net liquidity rising or falling — is the reservoir filling or draining? Locally: is the Singapore dollar on a strengthening or easing path? Together they give a posture — lean with the tide or brace against it — roughly a year before the tide fully shows up in prices. The point is not precision timing; it is refusing to judge any asset’s prospects without first checking the water level.

Where this breaks

This chapter carries a practitioner-judgement label deliberately: the liquidity-leads-prices relationship is an empirical regularity with a strong record, not a law. The lead time wobbles, the measurement is contested at the edges, and crowded knowledge of an indicator erodes its edge.

The deeper failure mode is structural, and Howell’s framework inherits it honestly. The cycle assumes the central bank can always ease — that opening the tap reliably lifts assets. Late in a long debt cycle, that assumption can fail: when debts are high enough and rates can fall no further, newly created money stops flowing into productive lending and flows instead into other currencies and inflation-hedge assets. Central bankers call it pushing on a string. In that regime the normal signal inverts — more liquidity, but it flees the currency rather than lifting the economy — and the assets that respond are the hard, supply-capped ones rather than the productive ones. Knowing which regime is operating matters more than the liquidity number itself.

The action: add the two dials to whatever gets checked monthly. Net liquidity for the global tide; the Singapore dollar’s path for the local one. Neither requires a subscription, and both answer the question that should precede every allocation decision — what is the water doing?

Markowitz and broken correlations

Sun, 05 Jul 2026 00:00:00 GMT

Diversification is the only widely available tool that reduces risk without reducing expected return — and it switches off at the exact moment it is needed most. Both halves of that sentence are true, both are measurable, and holding them together is what separates using Markowitz from being used by it.

The problem

An allocator holding one asset carries all of that asset’s risk. Common sense says spreading across several assets helps, but before 1952 nobody had said precisely how much, or shown that the mix could be engineered. Harry Markowitz’s insight was that a portfolio’s risk depends not just on each asset’s volatility but on how the assets move together — the correlations — and that mixing imperfectly correlated assets produces a whole calmer than its parts.

Given a menu of assets, there is a mix that delivers any target return at the lowest possible volatility. The line of those best mixes is the efficient frontier; any portfolio below it takes needless risk for its return. This is the chassis most institutional portfolios are still bolted onto, seventy years on.

The insight

The frontier is only as good as its two inputs, and both are guesses. Expected returns are backward-looking or modelled, always uncertain — and a 1% error in one asset’s estimate can swing its “optimal” weight by 20 to 30 percentage points. Optimisers are, in the trade phrase, error maximisers: they pour weight into exactly the assets whose rosy estimates are most likely wrong.

The second input is the correlation matrix, and it is less stable than it looks. Twenty assets means 190 pairwise correlations to estimate, and all of them shift under stress. Correlations are measured mostly in calm markets, because most history is calm — and in a panic, frightened holders sell everything at once to raise cash, so assets that spent a decade moving independently fall together.

The umbrella folds when it starts to rain.

2022 is the textbook case. The stock-bond correlation, negative for two decades, flipped positive in an inflation shock; “balanced” 60/40 portfolios — sold for a generation as automatically safe — behaved in realised risk almost like all-equity books. Illustratively, equities down around 18% and “safe” bonds down around 13% blend to roughly minus 16%, with the protective half doing almost none of its job. The model assumed the diversifier would diversify; the regime decided otherwise.

Two further failure modes complete the set. Markowitz assumes returns follow the bell curve, and real returns have fat tails — extremes far more frequent than the Gaussian allows, so the frontier under-sizes disaster for any asset whose defining feature is violent moves. And the model is regime-blind: a frontier optimised on risk-on data is dangerous in risk-off weather, which is why serious practitioners recompute per regime rather than once.

In plain English

Markowitz draws a flawless map of a calm harbour. In a storm, all the boats start moving together, and one or two of the boats turn out to be fireworks. The map remains genuinely useful — for calm days, with boats whose behaviour is understood — but it must never be mistaken for a promise about storms.

Three working rules fall out.

Seatbelt, not airbag. Treat diversification as the everyday seatbelt: it works in normal bumps and disengages in the crash, so no position should be sized as if its diversification will hold in a panic.
Start at minimum variance. Prefer the minimum-variance end of the frontier as a starting point, because it leans least on the fragile return estimates.
Cap every position. Cap any single position — a common institutional rule is 20% — as a hard, unconditional limit that blunts the error-maximiser pathology without requiring the inputs to be right.

There is also a quieter test hiding here: whether “different” holdings are actually different. A hypothetical Singapore portfolio holding local REITs and local blue-chip banks is not two bets — both ride the same domestic rate curve and the same handful of lenders. Diversification across names is not diversification across risks; the correlation question has to be asked at the level of what actually drives each holding.

Where this breaks

Beyond the crisis-correlation failure already described, mean-variance has a structural blind spot: it optimises inside a measuring unit it never questions. The mathematics is anchored to a “risk-free” rate on government debt, and in periods of sustained currency debasement that anchor carries a guaranteed real loss. There is no slot in the model for “the ruler is shrinking.” Nor is there a slot for binary events — default, reclassification, custody failure — that arrive as a jump rather than a drift.

Used honestly, Markowitz is a discipline for the middle of the distribution, paired deliberately with tools built for the tails. The next two chapters supply those tools: Kelly for how much to bet, and CVaR for how bad the bad case really is. Before moving on, one exercise: for any two holdings believed to be diversifying each other, name the single event that would make them fall together. If the event is easy to name, the diversification is conditional — size accordingly.

Positioning and flows: reading who is in the trade

Sun, 05 Jul 2026 00:00:00 GMT

A price does not move because someone is right. It moves because someone transacts — and how far it moves depends on who is left on the other side to absorb the order. This is the single most under-taught idea in markets: opinions do not set prices, flows do.

A million people privately convinced an asset is cheap move nothing; one forced seller into a thin market moves everything.

That reframe has a name: positioning. Where the map (Part I) asked what season is it, and the mental tools (Part II) asked what is the risk worth, this part asks a colder question — who already owns this trade, and who is left to buy it?

What positioning actually is

Positioning is the standing inventory of bets: not what people say, but what they hold. It matters because every buyer is a future seller and every short is a future buyer. A market where “everyone” is long has quietly converted its entire crowd of bulls into a reservoir of pending supply.

The mechanics of crowding are almost boringly structural. Professional trend-following capital adds longs as price rises and shorts as price falls — that is its design. So a strong trend mechanically recruits its own crowd, and the longer the trend runs, the closer it comes to pulling in the last marginal buyer.

At that point nothing needs to go “wrong” with the story; the fuel is simply spent, because the only way a price keeps rising is if someone new keeps paying more.

The deflating question, usable on any trade the taxi driver, the group chat, and the headlines all agree on: who is left to buy? If no next wave of buyers can be named, the position is not early. It is late.

Crowded trades and the narrow exit

Crowded trades do not unwind politely, and the reason is architectural: everyone entered through a wide door over months, and everyone exits through a narrow one in days. When a shock hits a crowded position, the natural buyers are already fully invested — so the first sellers find no bid, prices gap, and the gap itself forces further selling from margin calls and risk limits.

Selling begets selling. The academic literature calls this a liquidity spiral (Brunnermeier–Pedersen); traders call it an air pocket.

The short-volatility episode of early 2018 is the clean specimen. After an unusually long calm, speculative positioning in volatility futures sat at near-record net shorts — a crowding that was measurable for months. Equities then fell a few percent in one day, volatility spiked, short-volatility holders were forced to cover, and covering pushed volatility higher still, forcing more covering. The most popular short-volatility product lost the overwhelming majority of its value in a single session.

The same anatomy sits under 2000 (a euphoric crowd in profitless internet names while policy tightened), 2007 (an entire financial system long housing credit while spreads quietly widened), and 2021 (a leveraged retail crowd concentrated in a handful of names at peak liquidity). In each case the underlying story had some merit. The crowdedness, not the story, is what made the trade lethal — and the cruel corollary, formalised in the limits-of-arbitrage literature (Shleifer–Vishny), is that being right on fundamentals and wrong on timing is financially indistinguishable from being wrong.

The COT report: the crowding census anyone can read

The Commitments of Traders report — a free weekly census, published by the US futures regulator, of who holds what in US-listed futures — is the only systematic positioning data most private investors will ever access. Reading it well means knowing that the bettors are not equals. The report splits them into three categories, and the category read determines the signal.

Commercial hedgers are producers and users offsetting real-world exposure — an oil producer selling forward, an airline buying fuel forward. They are habitually early and “wrong” on timing, because they hedge when they feel economic pain, but rarely wrong on direction over a medium horizon. Heavy commercial shorts in a commodity often mean producers are locking in prices they consider high — a quietly contrarian, bullish-of-caution tell. This is the closest thing the data offers to informed money.

Large speculators — funds, managed-futures programmes, professional desks — are the crowd that makes the crowd, structurally trend-following. Small speculators, below reporting thresholds, were historically the contrary tell that piled in at tops; honest reading requires noting that this signal has decayed as retail access and tooling improved.

The one construction worth doing: take large-speculator net position (longs minus shorts) and rank it against its own trailing three-year range. A reading in the top or bottom decile is a positioning extreme. Two disciplines attach to that word. First, an extreme is a statement about fragility, never a timing trigger — crowded trades can stay crowded for months. Second, the report reflects Tuesday’s positions released Friday, a built-in three-day lag; in a fast market it is stale on arrival, which makes it a weeks-scale context frame, never a trade signal.

Decomposing the flow: money talks, opinions don’t

Positioning is the stock; flow is the current — and the same one-percent rally means different things depending on who drove it. Professional desks decompose flow by participant type; each component has a public proxy an ordinary reader can track.

Institutional directional flow shows up in fund flows: weekly net money into broad vehicles — an S&P 500 tracker, a gold tracker, in Singapore the STI ETF. Price rising with inflows is institutional sponsorship and tends to be durable; price rising on outflows is momentum running on fumes. Dealer hedging flow is the market’s hidden thermostat: when option dealers are net long options they sell rallies and buy dips, absorbing shocks; when net short they buy rallies and sell dips, amplifying them. This one force explains why some days drift and others snap. Retail sentiment flow reads through the put/call ratio and positioning surveys — most useful at its extremes.

Which is where the classic overbought/oversold logic belongs. Extreme readings — sentiment euphoric, speculative longs at multi-year highs, everyone already in — are contrarian caution, because the marginal buyer is exhausted. Extreme capitulation is contrarian interest, because forced sellers eventually run out. Neither extreme says act today; both say the setup for a violent move against the crowd is loaded.

Where this breaks

Four failure modes, each documented, each expensive to ignore.

The census samples one pond. COT data covers US-regulated, exchange-listed futures. For equities, rates and classic commodities that is most of the deep water; for crypto it is a puddle beside an ocean, because the real leverage lives in offshore perpetual futures, self-custodied holdings and on-chain lending — none of which reports to any census. A calm official positioning read can sit directly on top of record unreported leverage; there, funding rates and on-chain leverage are the true crowd-meters.

Positioning complements regime; it does not replace it. The 2020 oil collapse — the front contract briefly printing below zero — was not a crowding unwind; speculative positioning beforehand was only modestly long. It was demand destruction plus a supply shock, and the warnings lived in the macro chain, not the crowd-count. When the driver is fundamental, regime reading outranks positioning.

Proxies get spoofed by structure. Heavy fund inflows can be a pension’s mechanical monthly contribution, not conviction; a put-buying spike can be a fully invested fund buying insurance, not fear. Decomposition reveals who; it cannot always reveal why.

The detonator is often discontinuous. Flow data is continuous; the catalyst that breaks a crowded modern trade is increasingly a regulatory or policy step-change — a ruling, a ban, a redefinition — which is invisible to flow by construction. A crowded position held into a scheduled regulatory decision is maximum fragility meeting an unpriceable spark: the most dangerous configuration in markets.

The implication

Positioning loads the gun; it never pulls the trigger. No one can time the spark — but anyone can decline to stand in the powder room. Before any position, the two questions this part installs are: who is on the other side of this trade, and who is left to buy after me? What to do with the answers — direction, size, and the capital that never moves at all — is the discipline layer, next.

Regime speed and interweaving: reading the weather without pretending precision

Sun, 05 Jul 2026 00:00:00 GMT

Knowing the four seasons and the five chains is necessary but not sufficient, because two timing errors destroy more wealth than any wrong asset choice. Flip too fast and every market wobble becomes a costly false alarm; flip too slow and the storm is ridden all the way down. This chapter covers the three facts that govern timing — regime persistence, market speed tiers, and cycle nesting — and then the honest method for deciding which season is actually active.

Persistence: seasons have momentum, and the momentum is lopsided

Weather has momentum: a sunny week rarely flips to a blizzard overnight. Regimes behave the same way, and Hamilton’s regime-switching framework put numbers on it. Estimated on modern market history, the probability that an expansion month is followed by another expansion month runs near 0.92; the probability a contraction month is followed by another contraction month runs nearer 0.78.

Two consequences hide in that asymmetry. First, summer is stickier than winter — bull regimes persist for years; bear regimes burn out faster. Second, because the entry into winter is slow to confirm but the regime rarely snaps back once confirmed, patience after confirmation is statistically justified — while waiting for total certainty before acting is how investors ride a confirmed storm to the bottom.

This is the arithmetic behind an asymmetric decision rule that recurs throughout this curriculum: one credible alarm justifies leaning defensive; leaning toward risk requires everything to agree.

Speed tiers: the same storm arrives at three different times

A regime change does not hit all markets at once. It travels through them in a fixed order of speed.

Currencies move first. Foreign-exchange markets are the deepest and most heavily traded on earth, so stress shows up there within days.
Credit and equities move second, digesting the news over weeks.
The real economy moves last — defaults, layoffs, and recessions arrive months after the fast markets have already repriced.

The 1997 Asian crisis ran exactly this sequence: currency pressure first, equity outflows weeks later, corporate failures months after that.

The practical use is a two-way glance. Watch the fast tier for warning — it sees trouble first, though it also produces the most false alarms. Watch the slow tier for confirmation — it lies least, but it speaks last.

A signal appearing in the fast tier alone is a hypothesis; the same signal echoed in a slower tier is a regime.

Interweaving: three cycles, one price

The seasons of the first chapter describe the short cycle — the ordinary boom and bust of credit that turns every five to eight years. That cycle is nested inside a long debt cycle spanning decades, in which borrowing capacity builds across many short cycles until the central bank’s power to engineer a recovery runs down. And that cycle sits inside a still slower one: the multi-decade rise and decline of whichever currency the world saves in — a seat held before the dollar by the British pound, and before that the Dutch guilder.

Picture a swing moving back and forth on a playground that is itself slowly subsiding. The swing is the short cycle; the subsidence is the long one.

Every reassuring recovery statistic — the 0.78 that says winters end — was measured while the ground was solid, meaning while the central bank still had room to cut and stimulate. Japan after 1990 is the recorded case of the ground giving way: the short-cycle mechanics all present, the recovery mechanism unplugged for a generation. The nesting is why regime reading can never be fully mechanical: the same signal means different things at different depths of the longer cycles.

The regime flicker

One further interweaving deserves its own warning: a regime flicker is not a regime change. In August 2015 a surprise currency devaluation made every fast-tier indicator lurch as if winter had arrived; within weeks the move had fully reversed. It was a positioning spasm — forced selling by crowded traders — not a season change.

Distinguishing weather from a passing squall is exactly what confirmation rules are for.

The honest method: a small dashboard, strict rules, and a neutral default

Recognising the active regime does not require forecasting. It requires reading a handful of observable dials and obeying rules set in advance.

The dials:

the direction of credit spreads (widening or tightening)
the direction of central-bank policy (easing or tightening, in deeds not adjectives)
inflation relative to target (forcing policy or not)
the dollar’s trend (strengthening squeezes the world)
market breadth (whether declines are narrow or general)

None requires a terminal — all are reported in the financial press weekly.

The rules do the real work.

A signal counts only when it persists for two consecutive readings — a deliberate lag, accepted as the price of filtering out squalls like 2015.
One confirmed alarm is sufficient reason to lean defensive.
Leaning toward risk demands that all the dials agree.
And when they conflict — some flashing warning, others calm — the honest answer is neutral: no strong tilt in either direction.

Neutral is not indecision; it is the accurate statement that the evidence does not currently support conviction. Precisely dating a regime turn in advance is not possible; no one knows how, and anyone claiming to is selling something.

The limitation

Every element of this method has a known cost. The confirmation lag means being slightly late at every turn — deliberately, in exchange for far fewer false flips. Early signals can run eighteen months ahead of prices, and an early defensive lean underperforms painfully before it is vindicated: early is indistinguishable from wrong until confirmation arrives.

Recency bias works against the reader in both directions — long summers make persistence feel like a law just before it breaks, and deep winters make despair feel permanent at the exact bottom. And beneath all of it sits the structural caveat: the persistence numbers were estimated on a single monetary era, and the long cycles that frame it are still turning.

The one action that survives all of these limitations: write the triggers down in advance. A rule recorded in calm weather — if spreads widen a third of a point for two consecutive readings, the tilt goes defensive — survives the moment of fear or euphoria; a judgement improvised inside that moment rarely does. The next chapter supplies what the rules act on: the full menu of what an investor can actually own, season by season.

Regimes as weather: why the season matters more than the stock

Sun, 05 Jul 2026 00:00:00 GMT

Most people are sold financial products before they are taught financial concepts, so their decisions run on hope rather than understanding. This curriculum reverses that order. It is a map: markets have rules, seasons, and a causal logic, and before the map is in hand, price movement feels random — after it, most financial news resolves into a small number of recognisable patterns.

The map begins with the one concept that outranks every other in this guide: the market regime.

The problem: the first question most investors ask is the wrong one

The instinctive first question is “which stock should be owned?” It is the wrong first question, because the environment an asset sits in determines more of its return than anything about the asset itself.

In a year when the macro tide is rising, most equities rise; when the tide turns, most fall together. The gap between owning the “best” stock and an average one is small next to the gap between owning equities in the right season and the wrong one.

The tide

A worked illustration — deliberately hypothetical, modelled on the 2009–2014 recovery. A central bank cuts rates from 4% to 0.5% and holds them there for years. Broad equities return roughly 16% a year across the stretch, while the underlying businesses grow their earnings at roughly 6%.

Ten of those sixteen points came from the tide — investors willing to pay more for the same earnings as cheap money spread through the system — and only six from the businesses themselves. Most holders attribute the whole sixteen to their own selection skill. The tide gave them most of it.

This is the finding Hamilton (1989) formalised: markets do not draw returns from one stable distribution. They switch between a small number of persistent states, and which state is currently active is the single most important fact about any period. Howell’s global-liquidity research reaches the same destination from the flow side: the volume of money moving through the financial system — the tide — sets the level at which every boat floats.

The insight: four seasons, two dials

A regime can be read off two dials: growth (accelerating or decelerating) and inflation (rising or falling). Two dials, two directions each, four combinations — the four seasons of markets.

Spring — growth turning up, inflation low or falling. The recovery quadrant. Policy is easy, fear is still high, and risk assets recover before the economy does. Credit heals first, then equities, then hiring; the headlines are still dark when the season has already changed.

Summer — growth solid, inflation contained. The expansion quadrant. Equities compound, volatility falls, lending is generous, and everything financial appears easier than it is. Summer is also the season in which most financial products are designed and marketed — a fact worth carrying into every brochure printed during one.

Autumn — growth still standing, inflation rising. The overheating quadrant. The central bank leans against rising prices, borrowing costs climb, long-dated bonds suffer, and commodities and real assets take the lead. Autumn is where classic stock-bond diversification quietly stops working, because rising rates hurt both halves at once.

Winter — growth contracting. The contraction quadrant — with one further distinction that decides survival. In a deflationary winter (2008 is the reference case), government bonds are the classic refuge: rates fall, bond prices rise, and the cushion works. In an inflationary winter (2022), that refuge becomes a trap: bonds and equities fall together, and only cash-like instruments, commodities, and gold held their ground.

Same season on the calendar; opposite survival kit.

The one-sentence core of this entire part: you are not picking stocks; you are reading the weather, then choosing what to wear.

Plain-English translation

“Is this a good investment?” is an incomplete question. The complete question is: “is this a good investment in this season — and what happens to it when the season turns?”

Every asset class has seasons in which it is rewarded and seasons in which it is punished; none is good in all four. The fourth chapter of this part walks the full menu, season by season.

The local dials

A Singapore-based reader can watch the dials locally. Six-month T-bill and Savings Bond yields are a live thermometer of the tightening dial.

Singapore’s central bank runs its policy through the exchange rate rather than an interest rate, so a strengthening SGD path is the local version of a rate hike, aimed squarely at imported inflation. And because the local index is weighted heavily toward banks and property trusts, it experiences autumn — the rising-rate season — very differently from a tech-heavy foreign index.

The habit this builds for 2026 reading: when a central-bank statement holds rates steady but describes inflation as “persistent,” that is autumn language, whatever the accompanying reassurance says. And when a fund’s marketing email promises equity-like returns “in all market conditions,” it is claiming a season-proofness that no asset class has ever demonstrated. The claim itself is the red flag.

The limitation

Seasons do not announce themselves. Regime boundaries are obvious in hindsight and contested in real time, so the honest output of a regime read is a posture — lean defensive, lean toward risk, or stand neutral — never a dated forecast. No one can call the turn of a season in advance with precision; a claimed timing edge here is itself a red flag.

There is also a quieter caveat: nearly every statistic behind the quadrant model was estimated on the post-1971 era of freely created money — a single monetary epoch, not a law of nature. The third chapter of this part returns to what that means.

What comes next is the machinery: the four macro forces and the five domino chains that push markets from one season into the next — the difference between knowing that weather exists and knowing how a storm actually forms.

Tail hedging and the yield illusion

Sun, 05 Jul 2026 00:00:00 GMT

Fire insurance is cheapest when there have been no fires. After the fire starts, the same cover costs several times more — if it can be bought at all. Options on a portfolio behave identically, and that single fact generates both the most under-used defensive tool in investing and the most common disguise for hidden risk.

The problem

The Markowitz chapter established that diversification is a seatbelt that disengages in the crash: in a genuine panic, correlations converge and the “balanced” portfolio falls as one block. Something else has to play the role of airbag — a position that pays off more the worse things get, bought before the crash rather than during it.

The obstacle is not availability but psychology and pricing. Downside protection is cheap when markets are calm and volatility is low, and it multiplies in cost — three to ten times — once fear arrives. The useful version must therefore be bought when it feels most pointless, and paid for in years when it does nothing.

The insight

The instrument is the put option: the right, without obligation, to sell an asset at a fixed strike price before an expiry date. Illustratively, holding an index at a level of 100, a put struck at 90 might cost 2. If the index falls to 70, the put is worth roughly 20, cutting the net loss from 30 points to about 12. If the index never falls below 90, the 2 is gone — a steady bleed in calm years, as the option’s value melts toward expiry.

Vineer Bhansali’s institutional result is the trade-off stated honestly: a budget of roughly 1 to 2% of the portfolio per year, spent on deep out-of-the-money protection, can cut a catastrophic drawdown roughly in half — from the neighbourhood of minus 40% to the neighbourhood of minus 20%. That is buying back half the disaster for a couple of percent a year. The insurance is nearly free precisely when the sky is clear, which is also precisely when nobody wants it: in mid-2007, with expected volatility unusually low, protection against a 20% fall cost a fraction of a percent; by late 2008 the same protection had multiplied many-fold.

Retail allocators almost never do this, for three predictable reasons:

Loss-aversion weights the small certain premium more heavily than the large uncertain payoff.
The evaluation window is too short — the hedge “loses” three quarters in four and pays once a decade.
Options are mis-taught as speculation rather than insurance.

The yield illusion — the same trade, reversed

Now run the logic backwards, because an entire product category does. A covered call — holding an asset and selling someone else the right to its upside beyond a threshold — collects a steady premium every month. Marketed as “income,” “yield enhancement,” or “monthly payout,” it feels like the opposite of speculation. Structurally, it is the insurance seller’s side of the trade just described: the holder has sold convexity, keeping a capped upside and the entire downside in exchange for a regular fee.

That is a short volatility position in disguise. The premium is not free income; it is rent paid for absorbing tail risk, and the bill arrives in exactly the market where the downside shows up uncapped while the upside that would have healed it has been sold away. The same decomposition applies to most “enhanced yield” structured products: a bond, plus a sold option, plus fees, wrapped so the sold option is invisible. The one-line test for any yield materially above the sovereign benchmark: what risk is being sold to earn this, and would that risk be acceptable if it were named?

There is no free yield — only paid-for risk wearing a nicer suit.

In plain English

There are two different tails, and they need different airbags. For the deflation crash — the 2008 shape, where everything risky falls and cash is king — the protection is long puts and short-dated government bonds, assets a central bank stands behind. For the inflation and debasement tail — where money itself is being diluted and bonds fail alongside equities — the protection is hard, supply-capped assets, gold first. Owning only one airbag leaves one door open; 2022 punished portfolios that had insured only against the first tail.

For an allocator without options access, the structural equivalent of a tail hedge is unglamorous: a larger-than-comfortable reserve at the most liquid end of the portfolio, plus a small hard-asset sleeve. It bleeds less elegantly than a put ladder, but it serves the same function — guaranteed purchasing power at the moment everyone else is forced to sell.

Where this breaks

Tail hedging is filed here as practitioner judgement because the cost side is genuinely contested. Serious researchers have argued that systematically bought puts are expensive enough, often enough, that the long-run drag exceeds the crash payoffs — that the insurance is structurally overpriced because everyone fears the same tail. The debate is unresolved; what is not disputed is the asymmetry of when protection is cheap, and the arithmetic that a halved drawdown compounds enormously better than a full one.

The illusion side has no such caveat. A covered call is a short volatility position whether or not the brochure says so, and a “yield” that exceeds the sovereign rate is being paid for a risk, named or unnamed. The action: decompose before buying. Any income product should be reducible to its parts — what is owned, what has been sold, who must stay solvent, and what the all-in fee is. A product that resists decomposition has answered the question.

The asset menu: everything an investor can actually own

Sun, 05 Jul 2026 00:00:00 GMT

The financial industry sells tens of thousands of products. Underneath the packaging there are roughly nine things an investor can actually own — and every product is one of them, or a combination of them, wearing a wrapper and a fee. This chapter is the menu: for each class, where the return genuinely comes from, how it behaves in each season from the first chapter, and the kind of investor it suits.

One decomposition tool first, because it applies to every line on the menu. Any return has at most four honest sources:

the risk-free rate (the baseline for lending to the safest borrower — a T-bill)
a risk premium (extra pay for carrying a named economic risk)
genuine skill (rare, and assumed absent until proven across a full cycle)
hidden borrowing or complexity (returns that look like skill but are gearing in a suit)

A product that cannot say which source pays its headline number is quoting an unaudited figure.

The shield: assets that keep money

Cash and short-term bills — T-bills, Savings Bonds, money-market funds. The return is the risk-free rate, nothing more, and the price barely moves when rates change. Cash looks lazy in summer, when it lags everything, and quietly loses to inflation over long stretches.

Its season is autumn and both winters — not merely because it holds value, but because it is the only asset that can be deployed at the bottom, when everything worth owning is briefly cheap. For a Singapore-based reader, local T-bills and Savings Bonds carry no foreign-currency risk, and the exchange-rate-based policy regime gives the local currency a built-in buffer against imported inflation. Suits: every investor, as the floor beneath everything else.

Long-dated government bonds. The return is the risk-free rate plus a term premium — extra pay for lending over decades rather than months. The cost of that premium is duration: roughly, the percentage the price falls for each one-point rise in rates, and a thirty-year bond carries duration near eighteen.

Bonds are the classic refuge in a deflationary winter, when rates fall and their prices rise. In autumn and in an inflationary winter they are the trap on the menu: 2022 delivered double-digit losses on “safe” government paper. Suits: holders who know which kind of winter they are hedging.

Gold. No cash flow, no yield, no promise — which is precisely the point. Gold’s price is best read as a weather report on trust in money itself: it is rewarded during inflationary winters, currency stress, and the debasement phases of the long debt cycle, and it drifts through calm summers while yielding nothing. Suits: the investor who wants one asset in the shield that no borrower and no policy decision stands behind.

The engine: assets that compound

Broad equities. Ownership of businesses — the only class on the menu with an open-ended claim on human productivity. The pay is the equity risk premium, historically some four to six points a year over the risk-free rate, but realised only across a full cycle; it goes negative for three-to-five-year stretches without warning.

One structural note: growth-heavy equities are long-duration assets in disguise — most of their value sits decades away, so rising rates can hit them harder than a long bond. Seasons: spring and summer lead; autumn compresses their valuations; winters cut them. Suits: anyone with a decade and the temperament to not sell in the winter — most efficiently through a broad, low-cost index fund, since the environment drives most of the outcome anyway.

Investment-grade credit. Lending to strong companies for a modest spread over governments. It behaves like a calmer cousin of equities in most seasons and wobbles in deep winter. Suits: ballast inside the engine, not a source of excitement.

High-yield credit. Lending to fragile companies for a fat spread — and that spread is not income. It is an insurance premium collected for absorbing defaults, which arrive in clusters at the worst times; net of those clusters, across a full cycle, the class barely beats its investment-grade cousin with far more pain. Suits: only investors who recognise they are being paid to carry a specific risk, not receiving a gift.

Property and listed property trusts. Rent plus borrowed money. The rising-rate season deals the class a double hit: property values are marked down as required yields rise, while refinancing costs cut the distributions at the same time.

A distribution yield that rises because the price is falling is a warning light, not a raise. Suits: income-oriented investors willing to read lease lengths and debt structures — not those who treat the class as a bond substitute, which it is not.

The convex wing: assets that can multiply or die

Commodities. The raw-materials complex earns nothing in most seasons and shines in exactly one: autumn into inflationary winter, when it is one of the few classes that rises with the general price level. No compounding engine, high volatility. Suits: a deliberate inflation instrument, held for the season rather than for the decade.

Crypto. Two claims must be held at once. By design, the largest cryptoasset is the hardest money ever engineered — a fixed supply ceiling no authority can expand. In observed behaviour, it has traded as the highest-octane risk asset on the menu: its correlation to technology stocks rose from roughly 0.1 to roughly 0.7 into 2022, and it falls hardest whenever liquidity drains.

The class also carries a cost most menus omit — custody and regulatory drag: a lost private key is an irreversible total loss, an exchange failure confiscates blameless holdings, and a legal reclassification can reprice a token overnight. Suits: only capital sized so that a complete loss changes nothing — a position small enough to die quietly.

Illiquid alternatives — private equity, private credit, structured notes. The extra pay is the illiquidity premium, some two to four points for surrendering the exit. The premium is real, and so is its cost: the exit door shrinks fastest exactly when cash is most needed, and reported valuations smooth over storms that liquid markets price daily.

A smooth line is not the absence of risk; it is often the absence of honest measurement. Suits: investors whose liquid foundations are already complete, never as a first building block.

Asset class	Strongest season	Weakest season	Role
Cash & short bills	Autumn, both winters	Long summers	Floor, plus dry powder for the bottom
Long government bonds	Deflationary winter	Autumn, inflationary winter	Conditional recession hedge
Gold	Inflationary winter, currency stress	Calm summers	Trust-in-money hedge
Broad equities	Spring, summer	Both winters	The compounding engine
Investment-grade credit	Spring, summer	Deep winter	Ballast
High-yield credit	Spring, early summer	Winter (default clusters)	Paid risk, never “income”
Property & listed trusts	Stable-rate summers	Autumn (double hit)	Income with hidden rate sensitivity
Commodities	Autumn, inflationary winter	Deflationary winter	The inflation-season instrument
Crypto	Liquidity-rich summers	Any liquidity drain	Convex wing, sized to survive zero
Illiquid alternatives	Long summers	Any crisis	Premium harvest with a locked door

The limitation

Three warnings keep the menu honest.

Class labels drift: a young asset’s season membership can migrate as it matures, and its marketing narrative usually runs a decade ahead of its behaviour.
Calm-weather statistics betray: the correlations that make a menu look diversified are measured in fair weather and converge toward one in a crash — the earthquake logic documented by Longin and Solnik, and lived by every 60/40 holder in 2022.
And a menu is not a portfolio: knowing the dishes says nothing yet about proportions, and proportions — not selections — are where survival is decided.

That is exactly where the map hands over to the tools. The next part opens with the single most misleading number in finance — the “average” return — and why an average can drown an investor who never learned to read the depth beneath it.

The discipline layer: direction, size, and the untouchable floor

Sun, 05 Jul 2026 00:00:00 GMT

Most large losses are not analytical failures. They are the collapse of three separate decisions — which way to lean, how much to commit, and what must never be committed at all — into a single impulse, usually at the exact moment conviction feels strongest. Every durable school of risk practice, from institutional risk management to the systematic-trading literature, rediscovers the same defence: keep the three decisions structurally apart, so that a strong view cannot silently become a large position, and a large position can never reach the capital that guarantees survival.

This chapter describes that architecture as a set of concepts. It is a summary of a school of thought, not a product; nothing in it names a trade, a size, or a forecast.

Direction is a decision; size is a discipline; the floor is untouchable.

Direction: the three-state posture

The first separation gives direction exactly three values: leaning toward risk, leaning away from it, or neutral. Not a dial from one to a hundred — three states, because the coarseness is the point. A continuous dial invites continuous fiddling; three states force an explicit, reviewable decision each time the lean changes.

The load-bearing state is neutral, and it is the most misread. Neutral is not indecision. It is the default the whole system falls back to whenever the evidence is stale, missing, or in conflict — the posture that says, honestly, the inputs do not currently justify a lean either way. In a disciplined practice, neutral is where capital rests most of the time, and arriving there is an output of the process, not a failure of it.

Behind this sits a premise worth stating plainly: conviction is a resource. It depletes with use, it must be budgeted, and spending it on marginal setups leaves none for the rare configurations that deserve it. The behavioural-finance literature documents the opposite tendency — overconfidence expands to fill the available capital — which is precisely why the default must be structural rather than a matter of mood. A framework that always has a confident opinion is the one to distrust; the same test applies to any advisor, newsletter, or model.

One more boundary completes the separation: a posture is a direction, never a size and never a trade. The lean answers only “which way, if any?” How much, and through what instrument, are downstream questions with their own disciplines — because the historical disaster pattern is exactly a strong directional view stampeding into an oversized position in an illiquid instrument.

Size: evidence builds, size follows

The second separation governs how a position grows: size is earned by evidence, never declared by confidence. The practice starts small — small enough that being wrong is tuition, not damage — and adds only as independent lines of evidence come into agreement.

Independence is the operative word. Three signals derived from the same underlying driver — price momentum read on three timeframes, say — are one signal counted three times. Genuine agreement means unrelated families of evidence pointing the same way: the macro climate, the flow and positioning picture, the behaviour of the asset itself. Each additional independent confirmation justifies an increment of size; a contradiction from any one of them justifies none. The conditions combine as a conjunction, not an average — one failed condition holds everything, no matter how bright the others look, because enthusiasm cannot out-vote a red light.

Two quantitative bars anchor the discipline.

A floor — minimum return-per-unit-of-risk. Before size increases, the expected payoff must clear a stated bar relative to the risk taken — the logic of the Sharpe ratio applied as an admission test rather than a report card. An opportunity that cannot clear the bar is not a small opportunity; it is no opportunity.
A ceiling — fractional Kelly. Even a genuine, measured edge justifies only fractional commitment, because the edge itself is estimated with error, and the penalty for oversizing an overestimated edge is compounding ruin.

Between the floor of “not worth the risk” and the ceiling of “more than half-Kelly,” the honest sizing range is narrower than instinct wants it to be.

The floor: a wall, not a dial

The third separation is the strictest: a fixed portion of capital that is never deployed, regardless of how compelling any signal looks. Not a target, not a guideline — a floor defined in advance, whose breach triggers an automatic defensive stance rather than a debate.

The reasoning is structural, and it connects directly to the positioning chapter. The whole edge of holding capital is not being the forced seller — and the best prices in any cycle appear precisely when others are forced to sell. Only capital that is unencumbered on that day can act. A reserve that can be tapped “just this once” for an exceptional opportunity has re-joined the queue for the exit door; one bad regime later, it is being sold at the bottom alongside everyone else’s.

Hence the formulation: the floor is a wall, not a dial. A dial degrades silently — nudged once for a great setup, nudged again for a better one, gone by the time it is needed. A wall fails loudly: it either holds, or its breach forces the entire practice defensive, blind to which asset did the breaching. In the barbell framing (Taleb), the floor is the hard end that makes the risk-taking end survivable at all. The academic point hiding in the plain language: survival is a property of the whole book against the floor, not of any single position.

Freshness is a signal; blank is the discipline working

Two further habits complete the layer, and both concern honesty about information rather than markets themselves.

First, every number carries an age, and an undated number is suspect by default. A live exchange rate goes stale in minutes; an illiquid asset’s appraised value can be months old and look reassuringly calm precisely because nothing has traded. Disciplined practice therefore treats freshness as a signal in its own right: fresh evidence can support a lean, stale evidence is down-weighted, and missing evidence forces the honest answer — neutral. The distinction that protects capital is that no data and data that says zero are entirely different objects; a risk gauge reading zero invites action, a risk gauge that is blank because the feed is dead forbids it.

Second, and following directly: a blank is the discipline working. A flat posture in a loud market, a position still small while headlines scream, a reserve untouched beside a “once-in-a-decade” opportunity — each reads, from outside, as inactivity. Each is the machinery functioning: declining to manufacture a confident output from inputs that cannot support one. Systems that always produce a confident answer are optimised for engagement; systems that sometimes output nothing yet are optimised for survival. The honest blank is evidence of a working system; the confident unsourced number is evidence of the opposite.

Where this breaks

The discipline layer has real costs, and pretending otherwise would violate its own honesty rule. A neutral default forfeits returns whenever a strong trend runs on while the evidence bar is still being cleared; evidence-built sizing guarantees arriving late to fast moves; a permanent reserve floor is a measurable drag on compounding through long calm periods. These are not bugs to be engineered away — they are the premium paid for an insurance policy that pays out in forced-selling episodes.

Two harder limits deserve naming. The layer cannot rescue a bad universe: disciplined sizing of a structurally worthless asset produces a slower loss, not a gain. And the three-state posture inherits whatever errors sit upstream in the regime read — a wrong answer to “what season is it?” propagates into a confidently wrong lean, which is why the regime chapter comes first and why stale regime inputs must resolve to neutral rather than to yesterday’s answer.

The implication

Direction is a decision; size is a discipline; the floor is untouchable. The practical translation is a sequence, not a mood: establish the floor first, read the season, let evidence — not conviction — build the size, and treat every blank and every neutral as information. Before any capital moves, the ten-questions checklist in the tools section operationalises this chapter, one question per failure mode.

The five domino chains behind most crises and recoveries

Sun, 05 Jul 2026 00:00:00 GMT

Financial crises look like lightning — sudden, unforecastable, each one unique. They are closer to weather fronts. A small set of causal chains, running through the same four forces, produces most of the storms and most of the recoveries. The four forces:

credit (how willingly lenders lend)
liquidity (how much money is moving through the financial system)
growth (whether the real economy is expanding)
the dollar (the currency most of the world borrows and trades in)

The triggers differ every time. The order in which the dominoes fall does not. What follows traces the five chains step by step, each with its reference case — not because these are the only possible sequences, but because a reader who holds these five can decompose most financial history, and most financial headlines, into mechanism rather than mystery.

Chain one — the tightening spiral

Lenders turn cautious, or a central bank raises the price of money. The weakest borrowers must now refinance at worse terms, so the extra yield risky borrowers pay over governments — the credit spread — starts to widen.

Wider spreads make investment projects uneconomic; companies slow hiring and spending; earnings estimates fall; risk assets reprice downward; falling asset prices make lenders more cautious still. The spiral feeds itself.

The first domino is the quietest. Adrian and Boyarchenko showed that financial conditions predict not the average of future growth but its bad tail — the worst plausible outcome fattens while the consensus forecast barely moves.

In practice: a widening of roughly a third of a percentage point in investment-grade spreads over a few weeks, with no obvious news, is the earliest survivable warning. In 2006–2007 that signal blinked for over a year while growth forecasts stayed benign.

Chain two — the recovery arc

The spiral produces panic; panic produces a policy response. The central bank cuts rates, and in severe cases directly backstops markets that have frozen.

Liquidity expands while sentiment is still fearful. Risk assets bottom and turn up before the economy improves — which is why waiting for good news guarantees missing the recovery — and credit, equities, then employment heal in that order.

March 2020 is the reference case, and it carries the chain’s key diagnostic: the cure must match the disease. That crash was a funding freeze — markets seizing up for cash — which is exactly what a central bank can fix, because it can create cash without limit. When the backstop reached credit markets directly, the arc was underway within weeks.

The balance-sheet recession

The chain has a failure condition, and Japan after 1990 is its proof. In a balance-sheet recession — Koo’s term for a slump in which households and firms are so over-indebted that they repay debt at any price of money — rate cuts reach zero and nothing reignites. Money is created but not borrowed; the transmission fails; equities took decades to recover in nominal terms and never did in real terms.

The recovery arc is conditional, not guaranteed, and the condition is whether anyone still wants to borrow.

Chain three — the inflation trap

Inflation rises far enough that the central bank is forced to tighten even as growth weakens — the one configuration in which policy cannot rescue markets, because rescuing them would feed the inflation. Rates rise; the present value of distant cash flows falls; and here is the step most investors never see coming: bonds and equities fall together, because both are claims on faraway money. A share is the discounted value of decades of future earnings — functionally a long-dated bond wearing a different costume.

The usually negative stock-bond correlation flips positive, and the classic 60/40 portfolio loses on both legs at once. In 2022 a balanced portfolio of that shape lost roughly 16% — one of its worst years on record — with no banking crisis and no recession headline to blame. The hedge became the hazard, and an entire generation of allocators was surprised because their whole investing lives had shown only the friendly correlation.

Chain four — the reserve drain

A country pegs or manages its currency while borrowing heavily in dollars. The dollar strengthens; servicing the debt gets harder; the country spends its foreign-exchange reserves — its emergency dollar savings — defending the currency. When reserves run low, the defence fails, the currency breaks, foreign capital flees, dollar debts become unpayable overnight, and defaults cascade into the banking system and the real economy.

The 1997 Asian crisis is the reference case, and it teaches the chain’s signature: the order of visibility.

The currency market cracks first — it is the deepest, fastest market.
Equities reprice weeks later as foreign capital exits.
The real economy — defaults, bank failures, recession — grinds through last, months behind.

A fixed exchange rate is a promise backed by finite reserves; when the reserves are gone, the promise breaks violently, never gently.

Chain five — the dollar cascade

The dollar is not merely a currency; it is the world’s funding currency — the money most cross-border debt is written in. When the dollar surges, every borrower of dollars everywhere is squeezed simultaneously: their debts grow in local terms while their revenues do not. The result is a global, synchronised tightening that no single country chose.

The cascade has a recognisable ordering: the most fragile, most speculative assets crack first. Emerging-market currencies, crypto, and the riskiest credit fall before major equity markets do — and that ordering is itself the confirmation that a genuine global tightening is underway rather than a local accident. In 2022, crypto — marketed through the prior summer as an uncorrelated diversifier — fell by more than half alongside technology stocks, because both were the same bet on abundant liquidity wearing two costumes.

How the chains interlock

Two connections matter more than any single chain. First, one chain’s rescue seeds the next chain’s crisis: the unprecedented liquidity flood that reversed the 2020 crash became the monetary fuel for the 2022 inflation trap. The correct question about any rescue is what it will cost two years later.

Second, every chain that ends in recovery silently assumes the central bank still has room to act. The long-term debt cycle framework holds that this room decays as debt accumulates across decades of short cycles — and Japan is the recorded case of a chain-two engine with the recovery mechanism effectively unplugged. The chains are a reliable grammar within a monetary era; whether the era itself persists is a separate and harder question.

The limitation

The chains are a grammar, not a timetable.

They say what order the dominoes fall in, never on what date; more than one chain can run at once (2022 was chains three and five interleaved); and which chain is truly active is often settled only in hindsight. Anyone narrating a live crisis with total confidence about which chain it is has confused a map with a forecast.

The working test, from here on: take any financial headline — a rate decision, a currency wobble, a credit downgrade — and ask which domino is this, in which chain? Most of the time there is an answer, and finding it converts news from noise into position on a map.

What the chains do not say is how fast the dominoes fall. That is the next chapter.

The liquidity ladder

Sun, 05 Jul 2026 00:00:00 GMT

In a crisis, the question is never “what is the portfolio worth?” It is “how much of it can actually be reached?” Those two numbers can differ by a quarter or more of the total, and the gap is invisible in every calm-weather statement.

The problem

Liquidity is the ease of converting an asset to cash without moving its price. It is the most taken-for-granted property in investing, because in normal markets nearly everything is sellable and the property looks free. It is not free, and it is not constant.

Pástor and Stambaugh established the two facts that matter. First, liquidity is a priced dimension of assets: stocks more sensitive to market-wide drying-up of liquidity earn higher average returns, meaning investors demand compensation for holding what becomes hard to sell in stress. Second, and the sharper edge: market liquidity evaporates system-wide in crises, exactly when holders need it. The extra yield on hard-to-sell assets is not a gift; it is an insurance premium being received, and the claim arrives in the crash.

History keeps the receipts. In mid-2007 the first freeze appeared not in equities but in structured-finance paper — the hardest-to-sell instruments stopped trading months before the stock market noticed. The bottom rungs of the system’s ladder failed first, which is what makes them an early-warning system as well as a risk.

The insight

The practical rendering is a liquidity ladder: every holding placed on a rung by how fast, and at what cost, it converts to spendable cash. Not by how safe it is — by how reachable it is. The two properties are cousins, not twins.

Singapore supplies an unusually clean worked example, because its sovereign instruments span the ladder deliberately, from most reachable to least:

Cash and bank deposits — the top rung: reachable today, paying the least.
Singapore Savings Bonds — one rung down: redeemable any month at par with no capital penalty, a design that makes them, unusually among bonds, near-liquid savings with a term yield.
Treasury bills — short-dated and high-quality, but built to be held to maturity of six or twelve months; an early exit means selling at whatever the market offers that day.
Listed equities and broad index funds — the middle: sellable in minutes in normal markets, but at whatever the crisis price is.
CPF balances — near the bottom by design: retirement-locked, compounding steadily, and unreachable for a mid-life emergency regardless of balance.
Multi-year locks — the true bottom rung: structured notes, private deals, property, where exit is measured in years or penalties.

An illustrative audit shows why the ladder must be drawn explicitly. A hypothetical Singapore portfolio holding 15% in a five-year structured note and 10% in CPF has a quarter of its wealth unreachable in any crisis, whatever the statement says its value is. If the reachable rungs cannot cover a genuine emergency, the crash forces a fire-sale of the middle rungs at the worst prices of the decade — the illiquid bucket reaching up and burning the liquid one.

In plain English

Three questions arrange any portfolio onto the ladder. If cash were needed this week, what fraction is reachable without a price concession? What fraction is reachable only at whatever a stressed market offers? And what fraction cannot be reached at all — locked by contract, penalty, or statute? The answers should be known before they are tested, because the test only ever arrives in a crisis.

The yield ordering falls out naturally: each step down the ladder should pay more, because each step down sells an option — the option to change one’s mind.

A locked product paying no more than a Savings Bond is mispriced against the holder. And any “premium” yield should be read through this chapter’s lens first: some of what is marketed as skill or strategy is simply the illiquidity premium, collected while the weather is calm.

Where this breaks

The ladder’s rungs are drawn in calm weather, and stress redraws them downward. Assets sellable-in-minutes remain sellable in a crash, but at prices that convert paper diversification into realised loss; corporate bond funds have discovered that daily redemption terms can sit on top of weekly-liquidity holdings. The measured liquidity premium is also unstable — estimates vary by period and method — and a premium harvested with borrowed money has historically been the fastest route from “extra yield” to forced seller. The honest rule: the ladder describes today’s market; assume every rung is one level worse in the storm.

One connected system

This chapter closes Part II, and the eight tools are best held as a single machine rather than a toolbox.

The compounding arithmetic set the objective: maximise the geometric, not the average.
Markowitz mixes the assets, knowing the mixing weakens in storms.
Kelly sizes every position so that no single loss is fatal.
CVaR measures the depth of the bad case honestly, where VaR only fences off the normal one.
Howell reads the tide that decides whether risk is being paid or punished this year — with the Singapore dollar as the local dial.
Tail hedging buys the airbag while it is cheap and names the products that quietly sell it.
Factors supply the patient, rules-based edge that replaces stock-picking.
The liquidity ladder guarantees the whole structure can survive a crisis without being forced to dismantle itself at the bottom.

Run in that order, they form a checklist that needs no mathematics: what compounds, what diversifies, what size, how deep is the worst case, which way is the tide, where is the airbag, which habits are being harvested, and how much is reachable in a storm. Every question has appeared in this part; every one can be asked of any portfolio in an afternoon.

What the system deliberately does not supply is the trigger — when to act, how much conviction a signal deserves, and what is never for sale regardless of signal. That is a matter of discipline rather than analysis, and it is where the curriculum goes next.

The problem with average returns

Sun, 05 Jul 2026 00:00:00 GMT

An average return can describe an investment that loses money. That is not a paradox or a rounding error — it is the first fact of portfolio mathematics, and every tool in this part of the curriculum exists because of it.

The problem

Picture two ponds. In pond A, the water rises 50%, then falls 40%. In pond B, the water never moves. Most people, asked which pond holds more water at the end, pick A — it went up more than it came down. The arithmetic says B wins, decisively.

Run it with 100 dollars. Up 50% takes it to 150; down 40% takes it to 90. The portfolio is down 10%, even though the simple average of the two years is plus 5%. Returns multiply, they do not add, so a percentage gain and an equal percentage loss never cancel. Lose 40% and the climb back to even requires plus 67%, because the recovery starts from a smaller base.

This is the difference between the arithmetic mean (add the yearly returns, divide by the number of years — the number brochures quote) and the geometric mean (the rate wealth actually compounds at). Every marketing document leads with the first. Every account balance is built by the second.

The insight

The gap between the two numbers has a name and a formula. Over many periods, geometric mean is approximately the arithmetic mean minus half the variance — variance being volatility squared. That subtraction term is volatility drag: the permanent wealth bled away purely from bouncing around.

Two illustrative assets make the point. Asset X averages 15% with 20% volatility; its compound rate is roughly 15% minus 2%, so 13%. Asset Y averages 20% with 40% volatility; its compound rate is roughly 20% minus 8%, so 12%. Y carries the higher headline and compounds slower, because doubling the volatility quadruples the drag.

The extreme case is instructive. An asset that returns plus 300% one year and minus 75% the next has an “average” of plus 112% — and a compounded result of exactly zero. Every 100 dollars becomes 400 dollars, then becomes 100 dollars again, before fees, with a stomach-churning ride in between. The headline was a mirage; the drag was the truth. A hypothetical 2021 buyer of a token with exactly that profile was not unlucky — they were reading the wrong number.

The core of it:

Return is only half a number; the other half is what you risked to get it.

From here forward, no return figure in this curriculum stands alone; every one is paired with the risk taken to earn it. That pairing is what “risk-adjusted thinking” means, and it is the shared foundation of everything from Markowitz to Kelly.

In plain English

Smooth-and-slightly-lower beats volatile-and-higher over any real horizon. A big loss is a permanent tax on future compounding — a 50% drawdown demands a 100% gain to undo, which is why avoiding the catastrophic loss is mathematically more valuable than catching the extra upside. The question to carry into any pitch is not “what did it average?” but “what did it compound, and what was the worst stretch along the way?”

One more consequence hides in the safest corner of the portfolio. Cash is not zero-risk; it is a small, certain, invisible loss to inflation every year — a drag applied to the measuring ruler itself rather than to the asset. The allocator hiding entirely in cash has not escaped the compounding asymmetry; they have chosen the leak they cannot see over the one they can. “Safe” means matched to when the money is needed, not left in currency forever.

Where this breaks

The volatility-drag formula is an approximation, and it inherits the assumptions of the return series it is fed. It describes risks that live inside the historical distribution of prices — the bouncing. It says nothing about the risks that live outside it: an issuer defaulting, an asset being reclassified by a regulator, custody failing outright. An asset can have low measured volatility and a catastrophic hidden tail, and this chapter’s arithmetic will grade it as safe. Later chapters — particularly the move from VaR to CVaR — exist to patch exactly that blind spot.

Nor does risk-adjusted thinking pick assets by itself. It is a lens, not a menu. The regime map in Part I still decides which assets are in season; this chapter decides how their track records should be read.

The immediate action is a single substitution: wherever a product, a pitch, or a headline offers an average return, ask for the compound return and the worst peak-to-trough loss over the same period. Any seller who cannot or will not produce those two numbers has answered a different question — and that answer is also information.

From VaR to CVaR

Sun, 05 Jul 2026 00:00:00 GMT

The risk number most funds quote is the height of a fence. The number that matters is how far the ground drops on the other side. Those are different measurements, they have different names, and confusing them is how well-modelled portfolios produce unmodelled disasters.

The problem

VaR — Value at Risk — answers the question “how much could be lost in a normal bad period?” Formally: 95% of the time, losses will not exceed some stated amount. It sounds like a worst case. It is the opposite: it is the edge of normal, the boundary of the ordinary 95%, with total silence about the remaining 5%.

The silence is the flaw. Two portfolios can share an identical VaR while one falls slightly past the fence in a bad month and the other falls two hundred times as far. VaR cannot tell them apart, because it only marks where the tail begins, never how deep it runs. In 2008, every major bank ran a VaR model; the models were roughly right about the 95th percentile and catastrophically silent about the 1st, which arrived far worse than the fence implied.

The insight

CVaR — Conditional Value at Risk, also called Expected Shortfall — asks the question VaR refuses: given that this is one of the worst 5% of outcomes, what is the average loss across those scenarios? It is the mean of the tail, not its edge.

VaR is the fence; CVaR is the average depth of the fall past the fence.

The deeper reason mathematicians insist on CVaR is called coherence. Artzner and colleagues laid out four common-sense axioms a risk measure should obey; the load-bearing one is sub-additivity — combining two portfolios must never show more risk than the two measured separately, because diversification can help but should never mathematically hurt. VaR violates this: merging two unrelated positions can make measured VaR rise, which is nonsense, and means a firm cannot safely add VaR figures across its book. Acerbi and Tasche showed expected shortfall satisfies all four axioms. This is not aesthetic tidiness — it means CVaR aggregates honestly, while stacked-up VaR can lie.

A pair of illustrative portfolios shows what the tail-average catches that the fence misses. Portfolio A posts a calm plus 0.7% every month for thirty months — a beautiful smooth line — then falls 45% in month thirty-one. Portfolio B bounces noisily, up 8% one month, down 7% the next, worst single month minus 15%, but never breaks and always recovers. Most beginners point to the smooth line as the safer one. A’s average loss in its worst outcomes is catastrophic and permanent; B’s is uncomfortable and survivable. The smoothness was the disguise, not the safety.

In plain English

Replace “what is the average return?” with “what is the average loss in the worst months — and is that survivable without selling at the bottom?” That substitution is the entire chapter in one move. For any fund or product, ask for the CVaR or, in plainer clothes, the ten worst months on record, and average them: that is the felt risk. If the ten worst months cannot be produced, something important has been learned anyway — the risk is being hidden, not absent.

The wiggle that can be seen is rarely what ruins anyone; the hole that cannot be seen is. A return series that never wiggles is often a series that cannot wiggle — because the asset is rarely re-priced, or because incoming money and borrowed money are papering over the cracks. Smoothness in a risky world is a claim that should raise the level of scrutiny, not lower it.

Where this breaks

CVaR is estimated from a sample of historical returns, and the whole machine assumes the future tail resembles the past one. Three classes of loss break that assumption outright.

Custody and counterparty failure — an exchange collapsing, an issuer’s reserves proving fictional, an irrecoverable private key — is a total loss triggered by something other than a price move, appearing in zero return histories because it is not a return but a disappearance.
Regulatory reclassification or an outright ban is a step-function: an overnight repricing no continuous distribution contains.
A short, calm sample produces a shallow tail estimate no matter how sophisticated the formula — a risk number is only as deep as the worst event in its data.

The honest hierarchy is therefore three levels, not two. VaR marks the edge of normal. CVaR averages the depth beyond the edge, for risks that are in the sample. And the tail-of-tails — legal, custodial, sovereign — was never in the sample at all, and requires a different defence entirely: the insurance logic of the tail-hedging chapter, and the pre-commitment logic of the discipline layer.

The action is a two-question overlay on every holding. First: can this position lose everything from something that is not a price move? Second: can a regulator or a government reprice it overnight? A yes to either means the position’s true risk is not in any chart it will ever be sold with.

Backtests always look good

Sun, 05 Jul 2026 00:00:00 GMT

Every backtested track record a reader is shown is a survivor, and the surviving is the trick. The chart arrived because it looked good; the thousands that looked bad were never sent. Before examining any specific performance claim, it pays to understand the machine that manufactures them — because the machine guarantees a brilliant-looking output whether or not any skill exists anywhere in the process.

Built after the fact

A backtest is a search through data that already happened. The builder chooses indicators, thresholds, entry rules, exit rules, and date ranges, then adjusts until the historical curve looks right. With enough adjustable parts, any past can be fitted perfectly — including its random noise — and the coincidences captured in the fit will not repeat.

The mathematics of luck

The mathematics is unforgiving. Test enough strategies against the same history and the best one looks skilled even when every strategy is pure noise: the expected best Sharpe ratio (return earned per unit of risk taken) rises with the number of attempts through luck alone. Around a thousand noise strategies produce a luckiest specimen with a Sharpe ratio near 3.7 — a figure a genuinely skilled fund would celebrate — with zero real edge.

Bailey and López de Prado formalised this as backtest overfitting: the curve in the marketing material is the winner of a private tournament the reader never saw, and the polish of the curve measures the intensity of the tournament, not the quality of the idea.

The silent graveyard

Survivorship bias runs the same selection one level up — on funds and accounts rather than parameters.

Consider a firm that launches sixteen funds with essentially random strategies. After one year, roughly eight are up by chance; the eight losers are quietly closed. After four years of the same pruning, one fund has been up every single year — purely by coin flip — and that fund gets the glossy campaign while fifteen closures vanish from the record.

The graveyard is silent by construction. Closed funds stop reporting, deleted accounts stop posting, and abandoned model portfolios leave no trace, so the visible sample is systematically unrepresentative of the attempted population. Any performance figure that arrives without a count of the failures alongside it is a numerator with the denominator amputated.

One trial is not a track record

A single result — one portfolio, one year, one anecdote — carries almost no information about skill, because skill is a property of a distribution and one draw does not reveal a distribution. The base rate makes this concrete: across decades of SPIVA-style scorecards, roughly 80–90% of professional active funds fail to beat their simple benchmark index after fees over horizons of ten years or more. Sharpe’s arithmetic explains why this is not a fixable flaw: active managers collectively are the market, so the average active dollar must trail the average passive dollar by exactly the difference in costs — an accounting identity, not an opinion.

Against a prior that low, one reported win moves nothing.

A coin that landed heads once is not evidence of a two-headed coin.

Two 2026 specimens

A hypothetical but familiar artefact: “my AI portfolio returned 34% in six months,” with a screenshot. Three mechanisms operate at once.

The post exists because the number is good — losing portfolios do not get posted, so the selection happened before the reader arrived.
Six months is one draw: in a rising market, a random basket of popular names performs well, and no benchmark is offered against what a broad index did over the same window.
And the AI framing adds authority without adding evidence — a language model generated the portfolio, but it did not generate a track record, and the anecdote remains a single trial however sophisticated the generator.

Behind the one visible post stands the usual graveyard: the prompts, portfolios, and screenshots that lost, unposted.

The fund marketing email

Equally familiar: “our machine-learning strategy returned 22% annually (backtested to 2010).” The parenthesis is load-bearing. A strategy built in the present and tested back to 2010 was constructed knowing every crash, every winning sector, and every regime change since 2010 — it passed an exam while holding the answer key.

Machine learning enlarges this problem rather than solving it: modern pipelines can evaluate millions of parameter combinations, so the winning configuration is drawn from a vastly larger pool of attempts, and luck’s ceiling scales with search capacity. The more powerful the machine, the better the backtest looks — and the weaker the inference it supports.

Where this skepticism breaks

The filter has a real limitation: it also rejects genuine, not-yet-proven skill. A new manager with true edge and a short record is indistinguishable, from the outside, from the lucky survivor — the discipline simply refuses to pay for the distinction before evidence exists.

Live, dated, independently audited records spanning at least one full drawdown, with results on data the strategy never saw during construction, do constitute evidence and deserve to be weighed as such. The rule allocates the burden of proof; it does not claim skill is impossible.

The two questions

Two questions dissolve most performance claims before any deeper diligence is spent.

First: is this record live or backtested? Live means real money, dated trades, independent audit. “Backtested,” “simulated,” “model portfolio,” and “would have returned” all mean built after the fact — a hypothesis wearing the costume of a result.

Second: how many tries existed before this one was shown? How many strategies were tested, how many funds launched, how many portfolios run — and what happened to the ones not in the brochure? An honest operation can answer with a number; a survivorship machine cannot answer at all.

Neither question requires mathematics, market knowledge, or confrontation. They cost nothing, and they relocate the conversation from the quality of the curve — which is manufactured — to the size of the graveyard, which is where the truth about skill actually lives.

The pattern mythology problem

Sun, 05 Jul 2026 00:00:00 GMT

Elliott waves, Gann angles, and harmonic patterns persist for a structural reason, not a foolish one: they are built so that they cannot be wrong in advance — only reinterpreted afterward. A framework with that property never accumulates disconfirming evidence, however many forecasts miss, because each miss is absorbed as a mislabelling rather than recorded as a failure. The mechanism deserves a careful walk-through, because many capable people use these tools, and the problem lives in the structure, not in the users.

The mechanism: degrees of freedom exceed the constraints

Elliott waves

Elliott wave theory holds that markets move in five waves with the trend and three against it, with the pattern nested at every scale. The nesting is the escape hatch.

At any moment several wave counts are simultaneously legal, and the method explicitly maintains “alternate counts” in reserve; when price violates the preferred count, the count is revised — what was wave three becomes wave one of a larger degree — and the framework continues undisturbed. The theory survives every outcome because some valid labelling exists for every outcome.

Gann angles and harmonic patterns

Gann angles attach significance to specific slopes on a chart, but a slope is a ratio of price units to time units, and that ratio is a formatting choice. Rescale the chart and the angle moves; the significant line can therefore always be redrawn through whatever the market did. Harmonic patterns name ratio structures — Gartley, butterfly, crab — each defined with tolerance bands around its Fibonacci ratios; with enough named patterns and wide enough tolerances, some pattern is always close to completing somewhere, on some timeframe.

The common structure across all three: the framework’s flexibility is larger than the space of possible outcomes. A system that can accommodate any price path forbids nothing, and a claim that forbids nothing carries no information about the future — however intricate its geometry.

The falsifiability point

Popper’s criterion says the informative content of a claim is what it rules out. “The index will finish the year consistent with a valid wave structure” rules out nothing, because every path is consistent with some valid wave structure. This is not an insult; it is a measurement.

A claim that cannot fail in advance cannot succeed in advance either — it can only narrate, after the fact, with great elegance.

The empirical record matches the structural analysis. Surveys of technical-trading research (Park and Irwin) find that where chart-based rules are frozen into testable form, performance does not survive honest out-of-sample evaluation after costs.

Practitioners object that freezing the rules destroys the method — that the skill lives in the reading, not the rules. The objection is sincere, and it describes the problem exactly: skill located in an untestable reading cannot be distinguished from luck, by construction. Not necessarily absent — indistinguishable, which for capital allocation amounts to the same operational fact.

Why they persist: the honest account

Persistence is not gullibility. These frameworks perform three genuine functions, and only one of them is the one on the label.

First, narrative satisfaction. A retrospective wave count renders a chaotic year coherent — after the fact, the count always works, and it delivers the real reward of felt comprehension. Markets are mostly noise, and humans are pattern-completing machines; a framework that guarantees a pattern will always find a grateful audience.
Second, structured attention. Wave counting and harmonic mapping force a trader to watch specific levels, predefine scenarios, and wait rather than act on impulse. Users who report benefit are often reporting something real — the benefit of patience and preparation, which almost any structured framework would confer — and attributing it to the geometry instead.
Third, asymmetric scoring. A narrative that always fits feels more informative than a signal that is frequently, visibly wrong. Information runs exactly the other way: only a claim that can miss is capable of hitting. The comfort of the always-fitting story is purchased with its predictive content.

What a falsifiable technical claim looks like

Four properties, all fixed before the outcome:

a measurable variable
a stated threshold
a defined horizon
and an explicit condition under which the claim is wrong

Positioning data clears the bar. “Large-speculator net length in this futures market sits in the top decile of its trailing three-year range” is published weekly, computed by arithmetic, and datable to the hour. The claim it supports — that the trade is fragile because the marginal buyer is already in and fresh demand is thin — is a statement about vulnerability, not timing, and it can be scored across every historical instance of the same reading.

Flow data behaves the same way: fund flows, dealer-hedging pressure, and sentiment ratios at documented extremes are measurable, dateable, and capable of being wrong in public. The positioning-and-flows chapter develops these tools in full.

The difference is not sophistication — a nested wave count is far more intricate than a decile rank. The difference is that the decile rank can lose, and because it can lose, its wins mean something.

The limitation

Falsifiable does not mean reliable.

Positioning extremes measure fragility, not timing, and crowded trades can stay crowded for months
retail-sentiment tells have demonstrably decayed as market access broadened
flow proxies can be distorted by mechanical rebalancing that carries no view at all

Testable signals earn only the right to be evaluated — most fail the evaluation, and that is the system working.

And the argument here proves less than a hostile reading would want: it does not show that no individual using wave analysis has skill. It shows something narrower and harder — that the framework as stated cannot be tested, and untestable skill cannot be separated from luck even by the person who has it.

The sorting question

One question files any chart-based claim into its correct drawer: what future observation would count against this? Where a specific answer exists, the claim is analysis, and it can earn trust the ordinary way — by being right more often than chance, after costs, in public. Where no answer exists, the claim is mythology: sometimes beautiful, occasionally right, and permanently unaccountable.

Kanseitō

Factor investing

The problem

The insight

In plain English

Where this breaks

Kelly and fractional sizing

The problem

The insight

In plain English

Where this breaks

Liquidity cycles and the Singapore translation

The problem

The insight

The Singapore translation

In plain English

Where this breaks

Markowitz and broken correlations

The problem

The insight

In plain English

Where this breaks

Positioning and flows: reading who is in the trade

What positioning actually is

Crowded trades and the narrow exit

The COT report: the crowding census anyone can read

Decomposing the flow: money talks, opinions don’t

Where this breaks

The implication

Regime speed and interweaving: reading the weather without pretending precision

Persistence: seasons have momentum, and the momentum is lopsided

Speed tiers: the same storm arrives at three different times

Interweaving: three cycles, one price

The regime flicker

The honest method: a small dashboard, strict rules, and a neutral default

The limitation

Regimes as weather: why the season matters more than the stock

The problem: the first question most investors ask is the wrong one

The tide

The insight: four seasons, two dials

Plain-English translation

The local dials

The limitation

Tail hedging and the yield illusion

The problem

The insight

The yield illusion — the same trade, reversed

In plain English

Where this breaks

The asset menu: everything an investor can actually own

The shield: assets that keep money

The engine: assets that compound

The convex wing: assets that can multiply or die

The menu at a glance

The limitation

The discipline layer: direction, size, and the untouchable floor

Direction: the three-state posture

Size: evidence builds, size follows

The floor: a wall, not a dial

Freshness is a signal; blank is the discipline working

Where this breaks

The implication

The five domino chains behind most crises and recoveries

Chain one — the tightening spiral

Chain two — the recovery arc

The balance-sheet recession

Chain three — the inflation trap

Chain four — the reserve drain

Chain five — the dollar cascade

How the chains interlock

The limitation

The liquidity ladder

The problem

The insight

In plain English

Where this breaks

One connected system

The problem with average returns

The problem

The insight