Alchemy of Finance Ch. 6: Where the Theory Stops — Falsifiability and What a Retirement Investor Keeps

Investment Background

The previous five chapters presented reflexivity. This chapter marks where it stops.

For this book the chapter is especially necessary, because reflexivity faces a more fundamental problem than the other theories here: whether it can be tested at all.

And to his credit, Soros was far more honest about this than his followers. He states plainly in the book that its theoretical portion is "unsuccessful," and concedes that his framework cannot produce reliable predictions.

The Wall Street Translation

Criticism One: Falsifiability — The Most Fundamental

This is the most serious criticism of reflexivity.

A scientific theory must be able to say: "if such-and-such is observed, my theory is wrong."

Reflexivity struggles to satisfy that requirement.

The problem is that its explanatory power is too great:

• Price diverges from conventional valuation → a reflexive loop is running.
• Price matches conventional valuation → the loop is currently inactive.
• The bubble burst → the loop reversed.
• The bubble did not burst → we are still in the boom phase.

Every possible observation can be accommodated by the framework.

And a theory that explains everything usually predicts nothing.

This is the same class of problem as the overfitting discussed in Chapter 3 of Way of the Turtle: a model with enough degrees of freedom can always fit what has already happened.

It is also the same standard we applied when criticizing the subjectivity of technical analysis in Chapter 6 of Trader Vic. We must apply it to this book too.

Criticism Two: Clear After the Fact, Murky Before It

The eight-stage model in Chapter 2 is the best illustration.

On a historical chart, the eight stages are perfectly clear.

In real time, you cannot know which stage you are in.

Specifically, stage three's "successful test" and a genuine end of trend look identical as they occur. Only subsequent prices distinguish them — and by then the information has no trading value.

Which means the model's value as a forecasting tool approaches zero, however high its value as an explanatory one.

Criticism Three: Soros's Own Record Cannot Prove the Theory

This must be stated carefully, because it is easily misread as an attack on Soros.

His investment record is exceptional; that is not in dispute.

The problem is that his success cannot serve as evidence that reflexivity is correct.

Three reasons:

1. Survivorship bias. Chapter 5 of A Random Walk established the logic — among thousands of macro traders using various frameworks, some will have exceptional long-run records. We hear only from them.

2. Inseparability. Soros simultaneously possessed information channels, a professional team, execution capability, risk management discipline, and capital to absorb errors. There is no way to determine how much of his return came from reflexive insight versus the rest.

3. His failures exist but are discussed less. He was badly wrong in 1987 and on several later occasions. A framework that accompanies both successes and failures has a contribution that is hard to quantify.

What Soros Himself Conceded

This is the most notable part of the chapter, and the reason we respect the book.

Soros states in the book that its theoretical portion is unsuccessful. He calls himself a "failed philosopher." He concedes that:

• Reflexivity cannot produce reliable predictions.
• His investment decisions contain a great deal of judgment the theory does not explain.
• He is frequently wrong, and regards the speed of correcting errors as his real edge — exactly the content of Chapter 3.

A theorist conceding that his theory cannot predict is rare honesty in financial writing.

And it points to where this book's value actually lies.

So What Remains

Accepting all three criticisms, what remains is real and useful to a retirement investor:

One, as a lens rather than a predictor.

Reflexivity's value is that it lets you see a causal structure invisible within the conventional frame.

Even if you cannot use it to time anything, knowing that price may be changing fundamentals changes how you understand phenomena like "this is obviously a bubble, why hasn't it burst." And that understanding affects behavior — primarily by keeping you from shorting a boom and from chasing one.

Two, the fallibility method in Chapter 3 is entirely independent of whether reflexivity is correct.

"Actively hunt for where my thesis is wrong" is a methodological principle depending on no theory of market structure. Its value has independent psychological support — confirmation bias is thoroughly documented.

This is the book's most reliable and most transferable output.

Three, the credit-collateral loop in Chapter 4 is the closest thing here to a testable claim.

1929, 1989, 2007 — the common structure of three crises can be observed and documented. That specific loop has a stronger empirical basis than the general theory.

And its practical corollaries — inventory your credit exposure, use no leverage, hold assets outside the loop — depend on no forecast.

Four, the insight about rigid commitments in Chapter 5.

"The act of defending a commitment can weaken its foundation" has a direct personal-finance application (rigid withdrawal rules versus dynamic guardrails), and equally requires you to predict nothing.

Executable Trading Rules

1. Treat this book as a tool for understanding, never for prediction. The most important meta-rule. Its value is explaining what you are living through, not telling you what comes next.

2. Keep Chapter 3; stay skeptical of Chapters 1 and 2. The fallibility method has independent evidentiary support; the eight-stage model does not. Two parts of one book with entirely different evidentiary status deserve different treatment.

3. Never change your core allocation because "I have identified a reflexive loop." Soros himself acted too early within this framework. You have less information and fewer tools than he did. The correct approach is an allocation that survives every stage.

4. Treat mechanical rebalancing as a substitute defense against reflexivity. Raised in Chapters 1 and 2, stated here as the final conclusion: rebalancing requires you to identify nothing. It automatically trims what has inflated and buys after a bust. It is a defense that does not require the theory to be correct.

5. Apply Chapter 5's lesson to your own plan: do not build rigid commitments. Use withdrawal rules with adjustment mechanisms rather than fixed amounts. This is the book's most direct and least contested contribution to retirement planning.

Relevance to a Retirement Portfolio: Closing

Six chapters, four sentences:

• Chapter 1: price may be changing fundamentals, so there may be no independent value for price to deviate from — a third account, distinct from both efficient markets and behavioral finance.
• Chapter 2: the process has a recognizable life cycle, and busts run far faster than booms because leverage and liquidity are asymmetric in the two directions.
• Chapters 3–4: the most important specific loop is credit and collateral, and the defense is actively falsifying your own assumptions while holding assets outside the loop.
• Chapters 5–6: rigid commitments collapse because of the act of defending them — but the theory itself is hard to falsify, and Soros conceded it cannot predict.

This book's place in the library is specific:

A Random Walk says prices are efficient. Misbehaving and Thinking, Fast and Slow say people are biased. This book says causation may run both ways.

Three accounts of the same phenomenon. And for a retirement investor, the most important fact is that all three point to the same operational conclusion.

• If markets are efficient → buy low-cost index funds, because you cannot beat them.
• If people are systematically biased → buy low-cost index funds, because you have those biases too.
• If causation is reflexive → buy low-cost index funds, because the loop is unpredictable and mechanical rebalancing protects you without requiring prediction.

That convergence is itself the most forceful argument available. Three mutually incompatible worldviews producing one recommendation.

And our standard position is unchanged: a core of low-cost, globally diversified index funds, with a cash buffer covering essential spending, no leverage, and withdrawal rules containing an adjustment mechanism.

Soros made billions with his theory. He also conceded, in the book, that he could not tell you when it would work.

What is most valuable to you is not his returns, but his question: where might I be wrong?