By Gregory C. Eaves

April 23, 2026

Preamble: The Method

This document applies the Aristotelian first principles method to nine investment theories. For each theory, five phases are executed: (1) Assumption Autopsy — every assumption is surfaced and labeled; (2) Irreducible Truths — only what is verifiably true when all assumptions are stripped away; (3) Reconstruction from Zero — three distinct approaches built solely from those truths; (4) Assumption vs. Truth Map — a side-by-side table showing where conventional thinking diverges from first principles; (5) The Aristotelian Move — the single highest-leverage insight that conventional analysis would never produce.

The nine theories covered are: (1) Modern Portfolio Theory (MPT), (2) Post-Modern Portfolio Theory (PMPT), (3) Behavioral Finance and Behavioral Portfolio Theory (BPT), (4) The Black-Litterman Model, (5) Liability-Driven Investing (LDI), (6) Goals-Based Investing (GBI), (7) Factor Investing, (8) The Efficient Market Hypothesis (EMH), and (9) Core-Satellite Strategy.

This document does not synthesize a unified theory. That is Phase Two of the work. This document is the raw material: the first principles extracted from each theory that will serve as the building blocks for the unified theory to follow.

THEORY 1 OF 9

Modern Portfolio Theory (MPT)

Origin & Overview

Developed by Harry Markowitz (1952), MPT is the foundational framework of institutional investing. It introduced the concept of the efficient frontier — the idea that portfolios can be optimized for maximum return at a given level of risk by combining assets whose returns are not perfectly correlated.

PHASE 1: ASSUMPTION AUTOPSY

MPT embeds a dense web of assumptions, most of which are never stated explicitly:

• Investors are rational and risk-averse: All investors are assumed to maximize expected utility and always prefer less risk for the same return. This is borrowed from neoclassical economics, not observed human behavior.

• Risk = Standard Deviation: Volatility (σ) is treated as the complete and sufficient measure of investment risk. This is a mathematical convenience, not a verified truth.

• Returns are normally distributed: MPT assumes a bell curve. Fat tails, skew, and kurtosis — the very events that destroy portfolios — are assumed away.

• Correlations are stable: MPT uses historical correlations as if they are fixed. In crises, correlations spike toward 1.0, precisely when diversification is needed most.

• Markets are frictionless: No taxes, no transaction costs, infinite liquidity, unlimited short-selling. Borrowed from academic abstraction, not reality.

• A single-period investment horizon: MPT optimizes for one time period. Real investors have multi-decade, dynamic, evolving horizons.

• Expected returns can be estimated: MPT requires precise inputs for expected returns. These are notoriously unstable and difficult to estimate reliably — small changes produce radically different optimal portfolios (the ‘garbage in, garbage out’ problem).

• All investors have the same information: Homogeneous expectations. In practice, information is asymmetric, delayed, and interpreted differently.

• Diversification eliminates unsystematic risk: This is true only if the correlation assumption holds — which it does not in stress scenarios.

• The efficient frontier is the right target: MPT assumes the goal is mean-variance optimization. This is an assumption of goals, not a verified investor objective.

PHASE 2: IRREDUCIBLE TRUTHS

When every assumption is removed, these are the propositions that cannot be denied:

1. Capital is finite and depletable. Every allocation decision has an opportunity cost.
2. Different assets generate different cash flows with different timing and certainty.
3. Holding multiple assets whose values do not move in perfect lockstep reduces the variance of the combined portfolio relative to any single asset.
4. Future returns are uncertain. No mathematical model eliminates this uncertainty.
5. Higher potential return requires exposure to higher uncertainty.
6. Losses and gains of equal magnitude are not equivalent in their real-world impact (a 50% loss requires a 100% gain to recover).
7. Time matters: the sequence and timing of returns affects terminal wealth independently of average return.
8. An investor’s capacity to bear loss without changing behavior is finite and context-dependent.
9. Mathematical optimization is only as valid as its input assumptions.

PHASE 3: RECONSTRUCTION FROM ZERO

Using only the irreducible truths above, three distinct portfolio approaches are constructed as if no prior theory existed:

Approach A: Variance-of-Outcomes Architecture

Starting from Truth 3 (imperfect correlation reduces variance) and Truth 6 (losses and gains are asymmetric), build a portfolio not around maximizing expected return per unit of volatility, but around minimizing the probability of catastrophic drawdown. The portfolio is constructed by simulating the full distribution of outcomes (not just mean and variance), targeting a maximum tolerable loss threshold, and adding assets only if they reduce the probability of breaching that threshold — not just because they reduce standard deviation.

Approach B: Cash Flow Architecture

Starting from Truth 2 (assets generate cash flows) and Truth 7 (sequence of returns matters), build a portfolio entirely around cash flow streams rather than price volatility. Each asset is evaluated on the reliability, timing, and growth of its cash flows. Portfolio construction is a ladder of cash flow certainty, not a mean-variance optimization. Price volatility becomes irrelevant except as a buying/selling opportunity.

Approach C: Capacity-Constrained Architecture

Starting from Truth 8 (loss tolerance is finite and context-dependent) and Truth 1 (capital is finite), define the investor’s actual behavioral loss threshold first — the point at which they would change behavior — and work backward. Build the minimum-complexity portfolio that stays within that threshold at all plausible market scenarios. No optimization model. Just a stress-tested boundary condition.

PHASE 4: ASSUMPTION vs. TRUTH MAP

Where conventional thinking diverges from first principles:

PHASE 5: THE ARISTOTELIAN MOVE

The Move: Replace variance as your risk metric with maximum tolerable drawdown expressed in dollars, not percentages.

Every other reform of MPT tinkers at the edges — better correlations, better return estimates, downside variance. The Aristotelian move is to abandon the premise that risk is a statistical property of an asset. Risk is a relational property between an asset’s behavior and an investor’s specific capacity and goals. Define the dollar amount you cannot afford to lose in any 12-month period without changing your behavior or your life plan. Build the entire portfolio around never breaching that number under any historical or plausibly simulated scenario. Let everything else — asset allocation, diversification, position sizing — be determined by that single constraint. This is the move MPT cannot make because it requires acknowledging that the objective function it optimizes is not the investor’s actual objective.

THEORY 2 OF 9

Post-Modern Portfolio Theory (PMPT)

Origin & Overview

Developed by Brian Rom and Kathleen Ferguson (1991), PMPT arose as a direct critique of MPT’s use of standard deviation. It replaces symmetric variance with downside deviation — measuring risk only below a target return threshold — and uses the Sortino Ratio instead of the Sharpe Ratio.

PHASE 1: ASSUMPTION AUTOPSY

PMPT corrects MPT’s symmetric risk measure but imports most of its other assumptions intact:

• Downside deviation is the correct and complete measure of risk: PMPT replaces σ with downside deviation. This is better than MPT, but still a single statistical measure attempting to capture a multidimensional reality.

• A target return threshold (MAR) can be specified meaningfully: PMPT requires a Minimum Acceptable Return. Where this number comes from — and whether investors can actually specify it — is an inherited assumption, not a derived truth.

• Return distributions are non-normal but still stationary: PMPT accommodates skew and kurtosis but still assumes the distribution is stable over time. Regime changes are not addressed.

• Optimization remains the goal: PMPT still seeks to optimize a risk-adjusted return ratio. The assumption that optimization of a ratio corresponds to investor welfare is unexamined.

• The Sortino Ratio is a meaningful decision tool: Like the Sharpe Ratio, the Sortino Ratio compresses multi-dimensional risk into a single number, losing critical information about the shape and depth of downside outcomes.

• Historical downside data predicts future downside risk: PMPT still relies on historical return data. The distribution of future downside events may not resemble history, especially in novel market environments.

• Investors care only about returns below a threshold, not the path to those returns: PMPT ignores sequence-of-returns risk, the psychological impact of the path, and the difference between a brief dip and a prolonged drawdown of the same magnitude.

• The framework is investor-agnostic: PMPT, like MPT, presents a universal framework. It does not account for the radical variation in investor circumstances, tax situations, liquidity needs, or behavioral tendencies.

PHASE 2: IRREDUCIBLE TRUTHS

When every assumption is removed, these are the propositions that cannot be denied:

10. Losses below an investor’s required return level are categorically worse than equivalent gains above it.
11. Return distributions are asymmetric and exhibit fat tails — especially on the downside.
12. A single statistical ratio cannot fully describe the risk profile of an investment.
13. The relevant risk threshold is investor-specific, not universal.
14. The depth, duration, and sequence of losses matter independently of their statistical frequency.
15. Future return distributions may differ structurally from historical distributions.
16. Capital that falls below a critical threshold may trigger behavioral responses that permanently alter the investment outcome.

PHASE 3: RECONSTRUCTION FROM ZERO

Using only the irreducible truths above, three distinct portfolio approaches are constructed as if no prior theory existed:

Approach A: Threshold-Relative Portfolio Construction

Starting from Truth 1 and Truth 4 (losses below threshold are categorically worse; threshold is investor-specific), build a portfolio where every position is evaluated solely on its contribution to the probability of falling below the investor’s specific required return. No universal optimization. Each investor’s portfolio is custom-constructed from their own threshold, not from a shared ratio.

Approach B: Tail-Shape Portfolio Architecture

Starting from Truth 2 and Truth 5 (distributions are asymmetric; depth and duration of losses matter), construct a portfolio evaluated on the full shape of its downside distribution — not just the area below a threshold. Focus on the worst 5% of outcomes: how deep, how long, how recoverable. This is portfolio construction by stress scenario, not statistical optimization.

Approach C: Regime-Adaptive Portfolio

Starting from Truth 6 (future distributions may differ from history), build a portfolio that explicitly defines behavior under multiple regime assumptions: normal environment, crisis environment, inflationary environment, deflationary environment. Evaluate downside risk separately in each regime. The portfolio that performs acceptably across all regimes — not the one that optimizes a single historical ratio — is the target.

PHASE 4: ASSUMPTION vs. TRUTH MAP

Where conventional thinking diverges from first principles:

PHASE 5: THE ARISTOTELIAN MOVE

The Move: Define the investor’s ruin threshold — not their target return — and make it the primary portfolio constraint.

PMPT improves MPT by measuring downside risk relative to a target. But it still optimizes toward a target return. The Aristotelian move is to invert this: instead of asking ‘what portfolio maximizes my Sortino Ratio relative to my target?’, ask ‘what is the level of loss from which I cannot recover, financially or behaviorally?’ That is the ruin threshold. Build a portfolio where the probability of reaching that threshold in any simulated scenario is effectively zero. The target return becomes secondary — a product of what remains after the ruin constraint is satisfied. PMPT cannot make this move because it inherited MPT’s assumption that optimization toward a positive target is the correct objective. The first principle says the correct objective is survival first, return second.

THEORY 3 OF 9

Behavioral Finance and Behavioral Portfolio Theory (BPT)

Origin & Overview

Emerging from Kahneman and Tversky’s Prospect Theory (1979) and formalized by Shefrin and Statman (2000), BPT replaces the rational investor with a psychologically realistic one. It acknowledges mental accounting, loss aversion, overconfidence, and the construction of layered ‘mental account’ portfolios rather than mean-variance optimal ones.

PHASE 1: ASSUMPTION AUTOPSY

BPT is the most self-aware of the theories about its assumptions, yet it still carries significant embedded ones:

• Behavioral biases are stable, catalogable, and predictable: BPT assumes that investor psychology follows identifiable, consistent patterns. But biases interact, vary by context, and shift over time. Human irrationality is not a fixed set of equations.

• Mental accounts are the correct unit of analysis: BPT models investors as maintaining separate mental buckets (safety layer, aspirational layer). This is descriptively accurate for many investors but prescriptively adopted as the right framework.

• Loss aversion is universal and stable: BPT inherits from Prospect Theory the 2.5x loss aversion coefficient. This ratio varies significantly across individuals, cultures, and contexts.

• Biases are deviations from a rational norm that should be corrected or accommodated: BPT treats rational MPT as the benchmark and behavioral patterns as deviations. This assumes the rational benchmark is the right target, which is itself unproven.

• Understanding biases helps investors make better decisions: Awareness of a bias does not reliably reduce it. This is an empirically contested claim.

• A portfolio that accounts for psychological comfort is necessarily less efficient: BPT assumes a trade-off between psychological satisfaction and financial optimality. This assumes financial optimality is well-defined — which requires all the MPT assumptions it ostensibly rejects.

• Market prices reflect aggregate behavioral patterns: BPT uses behavioral patterns to explain market anomalies but assumes these patterns are persistent and exploitable — itself a contested empirical claim.

PHASE 2: IRREDUCIBLE TRUTHS

When every assumption is removed, these are the propositions that cannot be denied:

17. Human beings make decisions under uncertainty using heuristics, not optimization algorithms.
18. Losses are experienced more acutely than equivalent gains — this is a documented neurological and psychological reality.
19. Investors simultaneously hold multiple distinct financial goals with different time horizons, risk tolerances, and emotional weights.
20. Investor behavior has direct consequences for investment outcomes, independent of market behavior.
21. The psychological sustainability of a portfolio strategy determines whether it will be maintained through adverse conditions.
22. Awareness of a cognitive pattern does not reliably prevent its expression in behavior.
23. The framing of a financial decision changes the decision made, even when the economic content is identical.

PHASE 3: RECONSTRUCTION FROM ZERO

Using only the irreducible truths above, three distinct portfolio approaches are constructed as if no prior theory existed:

Approach A: Behavioral Durability Architecture

Starting from Truth 5 (psychological sustainability determines whether a strategy survives), design a portfolio not to be mathematically optimal but to be behaviorally durable. Every asset and position size is tested against the question: ‘Will the investor hold this through a 30% decline?’ If not, it is resized or removed regardless of its theoretical efficiency. The portfolio that is actually maintained dominates the optimal portfolio that is abandoned.

Approach B: Goal-Compartment Portfolio

Starting from Truth 3 (investors hold multiple distinct goals with different emotional weights), build a portfolio that maps explicitly to discrete, named goals — not to an aggregate utility function. Each compartment has its own assets, risk tolerance, and time horizon. No inter-compartment optimization. The investor experiences the portfolio as a set of purpose-built tools, each of which they can evaluate on its own terms.

Approach C: Decision-Environment Design

Starting from Truth 7 (framing changes decisions) and Truth 6 (awareness does not prevent bias), focus not on building the right portfolio but on designing the right decision environment. Automate contributions, rebalancing, and tax-loss harvesting to remove the investor from the decision loop on high-frequency, high-bias-risk decisions. Reserve human judgment for low-frequency, high-stakes decisions where deliberation is possible.

PHASE 4: ASSUMPTION vs. TRUTH MAP

Where conventional thinking diverges from first principles:

PHASE 5: THE ARISTOTELIAN MOVE

The Move: Design the portfolio’s decision architecture before designing the portfolio itself.

Every behavioral finance intervention assumes that if you show investors their biases, they will invest better. The evidence does not support this. The Aristotelian move is to treat the investor’s decision environment as the primary design problem, and the portfolio as the output of that environment. Define which decisions the investor will make manually (none, if possible), which will be automated, and under what conditions human override is allowed. Then build the portfolio to function within that decision environment. This is the move behavioral finance cannot make on its own because it still treats the investor’s psychology as a problem to be educated away rather than a fixed constraint to be engineered around.

THEORY 4 OF 9

The Black-Litterman Model

Origin & Overview

Developed at Goldman Sachs by Fischer Black and Robert Litterman (1990), the Black-Litterman model attempts to solve MPT’s sensitivity to expected return inputs. It starts from market equilibrium (implied by market-cap weights) and allows investors to express specific views, blending them with equilibrium returns in a Bayesian framework.

PHASE 1: ASSUMPTION AUTOPSY

Black-Litterman is a sophisticated patch on MPT’s input-sensitivity problem. Its assumptions are deeply embedded:

• Market-cap weights represent equilibrium: BL assumes that the current market capitalization weights reflect the aggregate expectations of all informed investors in equilibrium. This is an assumption of market efficiency, not a verified truth.

• Equilibrium returns can be reverse-engineered from market weights: The reverse optimization step assumes that the CAPM holds and that the risk aversion parameter can be calibrated. Both are contested.

• Investor views can be expressed as probabilistic statements about return differences: BL requires views in a specific mathematical form (absolute or relative). Most real investor insights do not map cleanly to this format.

• Confidence in views can be quantified: The Omega matrix (view uncertainty) requires investors to assign precise confidence levels to their views. This is typically arbitrary in practice.

• Bayesian blending of equilibrium and views is the correct mechanism: The Bayesian combination assumes the prior (equilibrium) and the likelihood (investor views) are both well-specified probability distributions. Neither is in practice.

• The result is a stable, usable portfolio: BL produces portfolios that are less extreme than raw MPT optimization, but they are still sensitive to the confidence levels assigned to views.

• Views are independent of each other: The model allows for correlated views, but the calibration of those correlations is practically difficult and rarely done rigorously.

• The model is appropriate for institutional investors with quantitative views: BL was designed for professional asset managers with formal research processes. Its application to individual investors or qualitative views is a stretch.

PHASE 2: IRREDUCIBLE TRUTHS

When every assumption is removed, these are the propositions that cannot be denied:

24. Any portfolio optimization model is only as valid as its input assumptions.
25. Market prices contain information about aggregate expectations, but do not represent a single true equilibrium.
26. Investors have information and beliefs that differ from the market consensus — these differences can be the source of genuine return opportunity.
27. The strength of a belief and the evidence supporting it are distinct and should be tracked separately.
28. Combining two uncertain estimates produces a third estimate that is no more certain than its inputs.
29. Mathematical sophistication in portfolio construction does not eliminate the fundamental uncertainty of future returns.
30. The complexity of a model and its reliability are not positively correlated.

PHASE 3: RECONSTRUCTION FROM ZERO

Using only the irreducible truths above, three distinct portfolio approaches are constructed as if no prior theory existed:

Approach A: Explicit Uncertainty Portfolio

Starting from Truth 6 and Truth 7 (mathematical sophistication does not reduce uncertainty; complexity ≠ reliability), build a portfolio that explicitly acknowledges and quantifies input uncertainty at every step. Rather than producing a single ‘optimal’ portfolio, produce a range of portfolios across plausible input scenarios. The position sizes that are stable across that range are the portfolio. Positions that only appear in extreme input scenarios are excluded.

Approach B: Conviction-Weighted Deviation Portfolio

Starting from Truth 3 and Truth 4 (investor beliefs can be return-generative; strength of belief ≠ evidence for it), build a portfolio that starts from the market-cap baseline and permits deviations from it only in proportion to (a) the investor’s evidence base, not just conviction, and (b) the historical reliability of the investor’s judgment in that category. No view is expressed without a track record or evidence base to calibrate it.

Approach C: Minimum Necessary Complexity Portfolio

Starting from Truth 7 (complexity ≠ reliability), systematically reduce the number of free parameters in the optimization. Use the market portfolio as the default. Allow deviations only for views with high evidence and high conviction. Size deviations conservatively. Accept that a simple, robust portfolio dominates a complex, fragile one over time.

PHASE 4: ASSUMPTION vs. TRUTH MAP

Where conventional thinking diverges from first principles:

PHASE 5: THE ARISTOTELIAN MOVE

The Move: Replace view confidence calibration with a track record requirement: no active deviation from the market portfolio is permitted without a documented, auditable history of accuracy in that specific type of view.

Black-Litterman’s core insight is correct: start from market equilibrium and deviate only where you have genuine information advantage. But it then asks investors to quantify their confidence — a number that is almost always fabricated. The Aristotelian move is to replace that fabricated confidence with a real one: your historical accuracy rate in this specific type of prediction. If you have never successfully predicted sector rotation, you have zero confidence weight to assign — regardless of how compelling the narrative feels. This requires building a personal prediction log before making any active allocation. The model cannot make this move because it assumes investors can introspect their confidence. The first principle is that actual track records are the only honest calibration.

THEORY 5 OF 9

Liability-Driven Investing (LDI)

Origin & Overview

LDI emerged from defined benefit pension fund management in the 1990s-2000s and became prominent after regulatory changes (notably PPA 2006 in the US). The core idea: the fund’s liability stream — future pension payments — is the benchmark, not a market index. The portfolio is constructed to match or hedge those liabilities.

PHASE 1: ASSUMPTION AUTOPSY

LDI is more honest about its objective than most theories, but still carries assumptions that are rarely examined:

• Liabilities can be precisely measured: LDI requires a present value of future liabilities. This requires actuarial assumptions (mortality, inflation, discount rate) that carry significant uncertainty and are sensitive to interest rate changes.

• The discount rate for liabilities is the appropriate liability measure: Selecting a discount rate for future liabilities is a regulatory and accounting choice, not a financial truth. Different rates produce radically different liability measurements.

• Fixed income instruments are the natural liability hedge: LDI heavily favors bonds because their cash flows match liability timing. This is conditionally true — but only if bond credit quality and duration are maintained, and only if the liability assumptions hold.

• Duration matching is sufficient for liability hedging: Duration matching addresses interest rate sensitivity but not convexity mismatches, spread risk, inflation risk, or longevity risk.

• The funded status is the correct measure of portfolio success: LDI defines success as closing the gap between assets and liabilities. This is a specific institutional context assumption — not universally applicable even to all pension funds.

• The liability stream is knowable in advance: Actuarial projections of future payments carry substantial model risk, especially over 30-50 year horizons.

• Liability hedging and return generation are separable: LDI typically divides the portfolio into a ‘hedging portfolio’ and a ‘return-seeking portfolio.’ This separation assumes the two components do not interact in ways that undermine the hedge.

PHASE 2: IRREDUCIBLE TRUTHS

When every assumption is removed, these are the propositions that cannot be denied:

31. Every investor — individual or institutional — has future obligations that must be met from their asset pool.
32. The relevant measure of investment success is not absolute return but the ability to meet specific obligations at specific times.
33. The mismatch between the timing and certainty of assets and obligations is the fundamental source of financial risk.
34. Reducing the sensitivity of the asset-liability gap to interest rates, inflation, and longevity is a verifiable risk reduction.
35. Future obligations are uncertain — both in amount and timing — and this uncertainty compounds over long horizons.
36. Not all risks in a portfolio contribute equally to the probability of meeting obligations.
37. The cost of failing to meet obligations is asymmetric and potentially catastrophic.

PHASE 3: RECONSTRUCTION FROM ZERO

Using only the irreducible truths above, three distinct portfolio approaches are constructed as if no prior theory existed:

Approach A: Personal Liability Mapping

Starting from Truth 1 and Truth 2 (all investors have obligations; success means meeting obligations), apply LDI logic to individual investors. Map every known future obligation (housing, education, healthcare, retirement income) as a liability stream. Build the portfolio specifically to fund each liability. No generic target-date fund. A custom liability-matched portfolio for each investor’s specific obligation schedule.

Approach B: Obligation Uncertainty Portfolio

Starting from Truth 5 (obligations are uncertain and compound over time), build a portfolio that stress-tests not just asset values but liability estimates. Model the range of possible obligation outcomes (longevity, healthcare inflation, unexpected expenses) and ensure the portfolio survives the worst plausible combination. The goal is not to match the expected liability — it is to fund the 95th percentile liability.

Approach C: Asymmetric Risk Weighting

Starting from Truth 7 (failing to meet obligations is catastrophic and asymmetric), weight all portfolio risks by their contribution to the probability of obligation failure — not by their contribution to tracking error or standard deviation. Risks that are small in probability but catastrophic in obligation-failure terms get maximum attention. Small risks with large return potential but no obligation-failure implication get minimum attention.

PHASE 4: ASSUMPTION vs. TRUTH MAP

Where conventional thinking diverges from first principles:

PHASE 5: THE ARISTOTELIAN MOVE

The Move: Map every financial decision to a specific obligation, and refuse to take risks that are not compensated by a corresponding reduction in the probability of obligation failure.

LDI institutionalized the correct insight that investment success means meeting obligations — not beating an index. But it then re-introduces complexity (duration matching, hedging portfolio, return portfolio) that obscures the original truth. The Aristotelian move for any investor — individual or institutional — is to build an explicit obligation schedule first: when do I need money, how much, and how certain am I of that amount? Then, for every risk in the portfolio, ask: does bearing this risk reduce the probability of meeting an obligation, or does it only increase expected return? Risks in the second category are only permissible after all obligations in the first category are fully funded. This is the move that most investors — including many pension funds — never make because it requires accepting that return generation is a secondary objective.

THEORY 6 OF 9

Goals-Based Investing (GBI)

Origin & Overview

Popularized in the 2000s-2010s by practitioners including Ashvin Chhabra and wealth management firms, GBI organizes an investor’s portfolio around specific, named life goals — retirement, education, legacy, lifestyle — rather than around a single aggregate portfolio optimized for risk-adjusted return.

PHASE 1: ASSUMPTION AUTOPSY

GBI is intuitively appealing but contains assumptions that are frequently unexamined in practice:

• Goals can be clearly articulated in advance: GBI requires specific, quantified goals with timelines. Real goals are often vague, change over time, and conflict with each other. The assumption that investors can specify their goals clearly is a significant one.

• Goals are stable over time: GBI builds portfolios around goals as if they are fixed. Life circumstances change, priorities shift, and new goals emerge. The framework is largely static in a dynamic reality.

• Mental account separation improves outcomes: GBI assumes that organizing investments by goal reduces harmful behavioral interference. Evidence for this is mixed — mental accounts can also lead to sub-optimal cross-goal resource allocation.

• Each goal can be funded independently: GBI treats goals as separable. In practice, goals compete for the same finite capital pool, and funding one goal’s portfolio optimally may come at the expense of another.

• A probability-of-success metric meaningfully captures goal achievement: GBI typically uses Monte Carlo simulation to show ‘85% probability of meeting retirement goal.’ The assumptions embedded in that simulation are often invisible to the investor.

• Goals can be translated into specific portfolio allocations: The mapping from ‘I want to retire comfortably’ to a specific asset allocation involves multiple layers of assumption that are rarely made explicit.

• Goal prioritization is straightforward: When goals compete for limited capital, GBI requires prioritization. The framework does not provide a rigorous basis for this prioritization — it is typically intuitive and unexamined.

PHASE 2: IRREDUCIBLE TRUTHS

When every assumption is removed, these are the propositions that cannot be denied:

38. Investors have multiple, distinct financial objectives with different timelines, magnitudes, and emotional weights.
39. A single aggregate portfolio metric (total return, Sharpe Ratio) cannot capture whether an investor is on track for all of their distinct objectives.
40. Capital is finite; every dollar allocated to one goal reduces what is available for others.
41. The probability of meeting a financial goal depends on both the portfolio construction and the investor’s behavior in managing toward that goal.
42. Goal clarity — knowing specifically what you need, when — reduces decision error.
43. Some goals are non-negotiable (survival), others are aspirational (wealth growth) — these categories require different risk frameworks.
44. Time horizon is not a single number but a vector: different goals have different timelines that create different liquidity and risk requirements simultaneously.

PHASE 3: RECONSTRUCTION FROM ZERO

Using only the irreducible truths above, three distinct portfolio approaches are constructed as if no prior theory existed:

Approach A: Goal Hierarchy with Survival Floor

Starting from Truth 6 (goals are categorically different in urgency), establish a strict hierarchy: survival goals (housing, healthcare, basic income) are funded first to near-certainty. Lifestyle goals are funded second. Aspirational goals receive only capital remaining after the first two tiers are fully funded. No aspirational return-seeking until survival is secured. This is not a lifestyle preference; it is the logical consequence of the asymmetry of need.

Approach B: Dynamic Goal Rebalancing

Starting from Truth 2 (goals evolve) and Truth 3 (capital is finite), build a goals-based system with a mandatory annual review that re-prioritizes goals, re-estimates costs, and reallocates capital. Goals are not static containers. They are dynamic estimates that are revised as life evolves. The portfolio management system is built for revision, not set-and-forget.

Approach C: Minimum Viable Goal Funding

Starting from Truth 1 (goals are often vague) and Truth 5 (goal clarity reduces error), before building any portfolio, build a goal specification process. Require minimum viable goal definitions: what is the minimum acceptable outcome (not ideal) for each goal? Fund the minimum acceptable outcomes first. Additional capital funds the upgrade from minimum to ideal. This prevents the common failure of funding aspirational goals while leaving survival goals under-funded.

PHASE 4: ASSUMPTION vs. TRUTH MAP

Where conventional thinking diverges from first principles:

PHASE 5: THE ARISTOTELIAN MOVE

The Move: Define the minimum acceptable outcome for every goal before defining the ideal outcome, and fund all minimum outcomes to near-certainty before allocating any capital to ideal outcomes.

GBI correctly identifies that investors have multiple goals. But it then optimizes toward ideal goal outcomes — which requires assuming more risk than necessary. The Aristotelian move is to distinguish between the minimum acceptable outcome (which must be funded with near-certainty) and the ideal outcome (which is aspirational). Fund the floor of every goal before funding the ceiling of any goal. This requires investors to ask the question that GBI frameworks almost never ask: ‘What is the worst version of this goal that I could still live with?’ The answer to that question determines the risk-free portion of the portfolio. Everything above the floor is aspirational and can bear risk. This move is uncomfortable because it requires investors to confront unpleasant minimum scenarios — which is exactly why conventional goal-setting focuses on ideal outcomes instead.

THEORY 7 OF 9

Factor Investing

Origin & Overview

Rooted in the Fama-French Three-Factor Model (1992) and extended through subsequent research (Carhart 1997, Fama-French Five-Factor 2015), factor investing identifies systematic, persistent return premiums associated with specific characteristics: market beta, size (small cap), value, profitability, investment, and momentum.

PHASE 1: ASSUMPTION AUTOPSY

Factor investing is empirically grounded but rests on a framework of assumptions that are increasingly contested:

• Identified factors represent genuine, persistent return premiums: Factor premiums were identified from historical data. Their persistence going forward is not guaranteed — especially as they become widely known and arbitraged.

• Factor premiums are compensation for risk: The risk-based explanation (factors persist because they compensate for undiversifiable risk) competes with the behavioral explanation (factors exist because of persistent mispricing). The distinction matters enormously for whether premiums will persist.

• Factors can be implemented efficiently in a portfolio: Theoretical factor exposures identified in academic research are often significantly diluted in real-world factor ETFs due to index construction, rebalancing frequency, and capacity constraints.

• Factors are stable across time and markets: Factor premiums have shown significant variation across decades and geographies. The value premium, in particular, had an extended period of underperformance in the 2010s.

• Multi-factor portfolios are additive in their return premiums: The assumption that combining multiple factors produces additive premiums ignores factor correlation and the possibility that factors share the same underlying economic risk.

• Factor exposures can be measured and targeted accurately: Factor scores assigned by different providers to the same security vary significantly, reflecting model choices rather than empirical certainty.

• The academic factor zoo does not represent overfitting: Hundreds of ‘factors’ have been published. Many represent data mining rather than genuine return drivers. Identifying which are robust and which are spurious is an unsolved problem.

• Factor timing is impossible or not worth attempting: Most factor investing assumes static factor exposure. This assumes that factor premiums are not time-varying — a contested empirical claim.

PHASE 2: IRREDUCIBLE TRUTHS

When every assumption is removed, these are the propositions that cannot be denied:

45. Securities with different characteristics have historically generated different long-run returns.
46. Some return differences are explainable by systematic, structural features of the investment (e.g., a company being cheap relative to its fundamentals).
47. Return premiums that are well-documented and widely known are subject to arbitrage pressure that may reduce or eliminate them.
48. There is a difference between a return premium that compensates for genuine economic risk and one that results from a persistent pricing error.
49. Implementation costs, taxes, and portfolio turnover have a direct, computable impact on the realized value of any return premium.
50. Historical return data from any finite sample contains both signal and noise, and distinguishing them is not straightforward.
51. Diversifying across multiple systematic return sources reduces dependence on any single one persisting.

PHASE 3: RECONSTRUCTION FROM ZERO

Using only the irreducible truths above, three distinct portfolio approaches are constructed as if no prior theory existed:

Approach A: After-Cost Factor Portfolio

Starting from Truth 5 (implementation costs directly reduce realized premiums), build a factor portfolio entirely on the basis of after-tax, after-cost, net-of-implementation-friction factor exposures. Theoretical factor premiums are irrelevant. Only the premium that survives actual implementation — given the investor’s tax situation, account type, and holding period — is counted. This eliminates the majority of ‘factor’ products on the market.

Approach B: Structural Characteristic Portfolio

Starting from Truth 2 (some return differences reflect structural features), focus only on factor exposures with a clear, durable economic logic: cheapness relative to fundamentals (value), profitability, and low leverage. Exclude momentum and other factors that lack a clear fundamental mechanism. Build a simple, low-turnover portfolio around structural characteristics rather than statistical factor scores.

Approach C: Factor Humility Portfolio

Starting from Truth 3 and Truth 6 (known premiums face arbitrage; historical data contains noise), take the full market portfolio as the baseline and deviate only toward factors with (a) a clear economic rationale, (b) persistence across multiple independent data sets and geographies, and (c) sufficient expected premium to survive realistic implementation costs. Size factor deviations conservatively, acknowledging that the premium estimate from historical data is almost certainly overstated.

PHASE 4: ASSUMPTION vs. TRUTH MAP

Where conventional thinking diverges from first principles:

PHASE 5: THE ARISTOTELIAN MOVE

The Move: Calculate the minimum factor premium required to break even after all implementation costs in your specific tax situation before investing in any factor product.

Factor investing is sold on the basis of gross historical premiums — value has returned X%, small cap has returned Y%. These are before taxes, before fund costs, before trading friction, before the inevitable period of underperformance that will test behavioral durability. The Aristotelian move is to compute the break-even premium for your specific situation: your marginal tax rate, your account type, the fund’s expense ratio, estimated turnover and its tax impact, and a realistic holding period. If the expected net premium does not survive this calculation with a meaningful margin of safety, the factor exposure is not worth taking. In practice, this eliminates most actively managed factor products and many factor ETFs — and it is exactly the calculation that factor product manufacturers do not want you to do.

THEORY 8 OF 9

The Efficient Market Hypothesis (EMH)

Origin & Overview

Formalized by Eugene Fama (1970), EMH posits that asset prices fully reflect all available information. It comes in three forms: weak (prices reflect all historical prices), semi-strong (prices reflect all public information), and strong (prices reflect all information, including private). The practical implication is that consistent excess returns are not achievable without taking on additional risk.

PHASE 1: ASSUMPTION AUTOPSY

EMH is one of the most debated propositions in finance. Its assumptions are fundamental and frequently conflated:

• ‘Fully reflect’ is well-defined and measurable: EMH requires prices to ‘fully reflect’ available information. What this means precisely — and how you would test it — is contested. The hypothesis is difficult to falsify because any test requires a joint hypothesis about what the correct price should be.

• Information is homogeneous and equally accessible: EMH assumes all investors process the same information with the same speed. In reality, information access, processing speed, and analytical capacity are extremely heterogeneous.

• Arbitrage is costless and unlimited: EMH relies on rational arbitrageurs to eliminate mispricings. In practice, arbitrage is costly, risky (limits of arbitrage), and capacity-constrained.

• Market prices are the best estimate of intrinsic value at all times: This conflates price (what someone will pay) with value (what the asset is worth). These diverge persistently, especially in the short term.

• Consistent excess returns require taking on measurable risk: EMH’s auxiliary claim is that any excess return must be explained by risk. This requires a complete and correct risk model — which does not exist.

• The hypothesis applies uniformly across all markets and asset classes: EMH likely holds to different degrees in different markets. Large-cap US equities may be more efficient than small-cap emerging markets.

• Investor rationality is sufficient for market efficiency: EMH requires enough rational investors to arbitrage away mispricings. If rational investors face institutional constraints that prevent arbitrage, prices can deviate from fair value even with rational actors present.

PHASE 2: IRREDUCIBLE TRUTHS

When every assumption is removed, these are the propositions that cannot be denied:

52. Asset prices are set by the marginal transactions of buyers and sellers, each acting on their own information and objectives.
53. Information that is widely known and easy to act on is rapidly incorporated into prices.
54. Information that is costly to obtain, difficult to analyze, or restricted in use is incorporated into prices more slowly and incompletely.
55. The cost of gathering, analyzing, and acting on information determines how efficiently that information is priced.
56. Consistently outperforming the market requires either an information advantage, an analytical advantage, a behavioral advantage, or the acceptance of risks that the market has priced incorrectly.
57. Most investors — after costs — will underperform the market over long periods.
58. The degree of market efficiency varies across asset classes, market conditions, and time periods.

PHASE 3: RECONSTRUCTION FROM ZERO

Using only the irreducible truths above, three distinct portfolio approaches are constructed as if no prior theory existed:

Approach A: Cost-Minimization Portfolio

Starting from Truth 6 (most investors underperform after costs), if no consistent information or analytical advantage exists, the only guaranteed way to improve net returns is to minimize costs. Build the lowest-cost, broadest-market-exposure portfolio possible. Every basis point of cost is a certain drag on return. No market timing, no active selection, no factor tilts beyond those available at negligible cost.

Approach B: Advantage Inventory Portfolio

Starting from Truth 5 (outperformance requires a specific, identifiable advantage), before making any active investment decision, explicitly identify which type of advantage is being claimed: information, analysis, behavioral, or risk pricing. If no clear advantage can be named, the default position is passive. Active positions are sized in proportion to the evidence for the claimed advantage — not in proportion to conviction.

Approach C: Efficiency-Calibrated Allocation

Starting from Truth 7 (efficiency varies across markets), build a portfolio that is fully passive in highly efficient markets (large-cap US equities) and selectively active only in demonstrably less efficient segments (private markets, small-cap emerging markets, distressed debt) — and only if the investor has a verifiable advantage in those segments and the expected net-of-cost excess return is positive.

PHASE 4: ASSUMPTION vs. TRUTH MAP

Where conventional thinking diverges from first principles:

PHASE 5: THE ARISTOTELIAN MOVE

The Move: Before any active investment decision, write down your specific, verifiable advantage over the marginal market participant for that investment — if you cannot, index.

EMH is most commonly used as either a blanket argument for passive investing (ignoring that efficiency varies) or dismissed as obviously wrong (pointing to Buffett). Both uses miss the point. The Aristotelian move is to treat EMH as a diagnostic tool: for every active position, force yourself to articulate the specific mechanism by which you expect to outperform the aggregate judgment of all other market participants. Not a narrative — a mechanism. Who is on the other side of this trade? Why are they wrong? What information or analytical edge do you have that they lack? Why has the market not already priced this in? If you cannot answer these questions with specificity, the position is speculation, not investment. The practical implication is not that all active investing is irrational — it is that active investing without a named, verifiable advantage is.

THEORY 9 OF 9

Core-Satellite Strategy

Origin & Overview

A practical portfolio construction framework that emerged in the 1990s-2000s, combining a large passive core (broad market index exposure, typically 60-80% of the portfolio) with a smaller satellite portion (active, thematic, or alternative investments) intended to generate alpha or provide specific exposures not available in the core.

PHASE 1: ASSUMPTION AUTOPSY

Core-Satellite is often presented as a commonsense compromise between passive and active investing. Its assumptions deserve scrutiny:

• The core and satellite portions are meaningfully separable: In practice, the satellite positions often have significant factor exposures that overlap with the core, creating unintended concentration rather than genuine differentiation.

• The satellite portion can generate excess returns: The alpha-generation premise of the satellite inherits all the problems of active management: persistent outperformance is rare, and identifying it in advance is harder than it appears.

• The optimal core/satellite split is approximately 60-80/20-40: This split is convention, not derivation. It is borrowed from industry practice, not from any first-principles analysis of the investor’s specific situation.

• The core provides stability while the satellite provides return: This role assignment is an assumption. The satellite can — and frequently does — add volatility without adding return.

• Satellites provide genuine diversification from the core: If the satellite holds stocks, sector ETFs, or factor funds, its correlation to the core is often high, especially in crises — precisely when diversification is most needed.

• The framework is self-contained and complete: Core-Satellite does not specify how to select satellites, how to evaluate their performance, when to replace them, or how to manage the interaction between core and satellite risk.

• A modest satellite allocation limits the damage of poor satellite selection: A 20-30% allocation to an active satellite that significantly underperforms has a meaningful impact on overall portfolio results — the ‘limit the damage’ assumption is optimistic.

PHASE 2: IRREDUCIBLE TRUTHS

When every assumption is removed, these are the propositions that cannot be denied:

59. A portfolio can simultaneously pursue market-rate returns (through broad exposure) and targeted return opportunities (through specific positions).
60. The cost of active management must be exceeded by the benefit it provides for active management to be value-additive.
61. Positions in a portfolio interact: the risk and return of the combined portfolio is not simply the weighted average of the component risks and returns.
62. Complexity in a portfolio introduces behavioral risk: more positions require more decisions, which creates more opportunity for harmful behavioral interference.
63. The appropriate allocation to active or thematic bets is determined by the quality and strength of the underlying thesis, not by convention.
64. Cost minimization in the core is achievable and certain; return generation in the satellite is uncertain and costly.
65. A clear distinction between the purpose of each portfolio component improves decision-making and performance attribution.

PHASE 3: RECONSTRUCTION FROM ZERO

Using only the irreducible truths above, three distinct portfolio approaches are constructed as if no prior theory existed:

Approach A: Conviction-Sized Satellite

Starting from Truth 5 (satellite allocation should reflect thesis quality), eliminate the conventional 20-30% satellite allocation and replace it with a conviction-sizing rule: each satellite position is sized in proportion to the evidence base and historical accuracy of the underlying thesis. If no position clears a minimum evidence threshold, the satellite allocation is zero and the entire portfolio is core. The satellite is not a standing allocation — it is earned.

Approach B: Purpose-Differentiated Portfolio

Starting from Truth 7 (clear purpose distinction improves decisions), define the core not as ‘passive market exposure’ and the satellite not as ‘active alpha.’ Instead, define the core as ‘funding specific obligations with certainty’ and the satellite as ‘pursuing specific, named return opportunities that are not available in the core.’ Every position must be able to state clearly which purpose it serves — and be evaluated only against that purpose.

Approach C: Minimum Core with Gated Satellite

Starting from Truth 6 (cost in core is certain; return in satellite is uncertain), maximize the core allocation until all financial obligations are fully funded with high probability. The satellite is funded only from discretionary capital — capital whose loss would not affect any funded obligation. The satellite is then managed with full risk tolerance because it genuinely represents discretionary risk capacity.

PHASE 4: ASSUMPTION vs. TRUTH MAP

Where conventional thinking diverges from first principles:

PHASE 5: THE ARISTOTELIAN MOVE

The Move: Set your satellite allocation to zero unless you can name, in writing, the specific mechanism by which each satellite position will outperform — and track that prediction formally.

Core-Satellite is the institutionalization of investment FOMO: the feeling that a purely passive portfolio leaves return on the table. The satellite portion exists, structurally, to satisfy the need to ‘do something.’ The Aristotelian move is to treat the satellite as a null hypothesis: the default allocation is zero, and every satellite position must actively earn its place by meeting a minimum standard of evidence. That standard: a written thesis stating the return mechanism, the expected return above the core, the time horizon, and the exit criterion. The position is reviewed against this thesis, not against the market. If you cannot write the thesis, you have no basis for the position. This move is uncomfortable because it reveals that most satellite positions are not based on genuine investment theses — they are based on narrative, recency bias, or professional anxiety about underperforming a benchmark.

What Comes Next: Toward a Unified Theory

This document has done the demolition work. Nine investment theories have been stripped to their foundations. Across all nine, a set of recurring irreducible truths has emerged. These truths will serve as the building blocks of the unified personal investment theory.

The recurring first principles that have survived every phase of this analysis are: (1) Capital is finite and every allocation has an opportunity cost. (2) The relevant measure of success is meeting specific obligations and goals, not relative performance against a benchmark. (3) The probability and magnitude of catastrophic loss is more important than average expected return. (4) Investor behavior — not portfolio theory — is the primary determinant of actual investment outcomes. (5) Mathematical sophistication does not reduce the fundamental uncertainty of future returns. (6) Implementation costs, taxes, and behavioral friction are certain; alpha is not. (7) The correct portfolio is the one that is maintained through adversity, not the one that is mathematically optimal in a frictionless simulation.

The unified theory will be built from these seven propositions alone. Everything else is convention, assumption, or borrowed framework. The construction begins in the next document.

Categories: Uncategorized