Behavioral Finance & Microstructure
The Alchemy of Finance
Soros argues that market prices do not just reflect the fundamentals, they change them, and that this two-way loop drives booms and busts. For an investor the payoff is a question to ask of any rally: is the price still being confirmed by the thing it prices, or is it now propping that thing up?
The big picture
Soros, who ran the Quantum Fund and studied at the London School of Economics under the influence of the philosopher Karl Popper, wrote the book to set out the theory behind his trading. Its starting point is that people who take part in a market cannot know it fully, because their own views are part of what they are trying to judge. He calls the resulting two-way link reflexivity: expectations move prices, and prices then move the fundamentals the expectations were about — through the cost of capital, the ease of borrowing, the value of collateral, the currency a company can use to buy another. Equilibrium, the resting state standard theory assumes, becomes the exception rather than the rule.
The book tests the idea on history — the conglomerate boom of the late 1960s, the lending boom to developing countries that ended in the 1982 debt crisis, the strong-dollar cycle of the early Reagan years — and then on itself, with a trading diary kept from August 1985 through 1986. Why it matters now: any market where rising prices make money cheaper for the companies being priced, from AI infrastructure to private credit, is a candidate for the same loop. The useful habit is to track price and fundamental side by side and watch which one is leading.
The 3 strategic pillars
-
Fallibility and reflexivity
Participants act on a necessarily incomplete picture, and their actions change the situation they are judging.
Thinking plays two roles at once: it tries to understand the market, and it acts on the market. When the second role is strong, the gap between belief and reality does not close on its own; it can widen and change the reality. A share price that doubles lets a company issue stock or buy rivals on better terms, which lifts reported earnings, which seems to justify the price. The pack measures the two directions separately: does the price move lead the fundamental (P→F), and does the fundamental lead the price (F→P)?
-
The boom-bust sequence
A boom needs both a real trend and a misreading of it; the misreading reinforces the trend until the two can no longer be reconciled.
Soros sketches a recurring order: an unrecognised trend starts; it becomes self-reinforcing and survives tests along the way; conviction grows while the gap between expectation and reality widens; a twilight phase follows in which prices keep rising but belief weakens; then a reversal and a crash that overshoots below the trend. The stages are not fixed in length and not every boom completes them. The pack's labels (inception, acceleration, twilight, reversal, crash) are a stand-in built on lagged correlations and a price-to-fundamental gap, not Soros's own method, which he never wrote down as a formula.
-
Credit, collateral and the real-time experiment
Lending is where reflexivity bites hardest, because the value of the collateral depends on the willingness to lend against it.
Banks lend against assets whose prices rise because banks are lending; when lending stops, collateral values fall and force further cuts. The same logic runs through currencies and budget deficits in the book's account of the early 1980s. The trading diary then shows the theory in use: a working hypothesis, a position sized to it, and a record of when the hypothesis was revised. For a reader the transferable part is the method — write the thesis down, name the feedback you expect, and check it against the data rather than against the price alone.
What a Closelooknet reader does with it
The working use is a two-line chart and two numbers. Pick a price you care about and the fundamental it is supposed to reflect — an index and its earnings, a lender's share price and its loan book, a sector's price and the equity it has issued. Track the gap between them and, over a rolling window, whether price moves have been leading fundamental moves and the other way round. A rally where both directions are positive and the gap is still modest is being confirmed; a rally where the gap is stretched and the fundamental has stopped following is running on belief. The mistake it prevents is reading a rising fundamental as independent proof of a price that may be producing it.
The bridge to the Closelooknet approach
The glossary entry on reflexivity gives the short definition. Credit is where the loop is easiest to see: credit spreads are the price of the collateral chain, and the credit-stress dashboard tracks them. For the AI cycle, the capex-cliff question is a reflexive one — spending is funded partly by the equity values the spending supports. Money Temperature measures how hot the crowd is running across eight instruments, the market-level view of the same excess. On the shelf, Perez places the frenzy-and-crash phase inside a long technology cycle, and Misbehaving covers the individual biases that feed the misreading.
Action-Kit — from theory to practice
Tooling & data
| What you need | Where to get it | Cost |
|---|---|---|
| FRED (Federal Reserve Bank of St. Louis) Free macro and credit series — bank loans, corporate debt, credit spreads, house prices — to use as the fundamental side of the tracker | Federal Reserve Bank of St. Louis Downloads as CSV; resample to the same frequency as your price series before loading. | Free |
| Company fundamentals over time Quarterly earnings, revenue, share count and debt for the name or sector you track | Koyfin The free tier covers basic historical fundamentals; longer histories and exports sit in the paid plans. | Freemium |
| Historical prices Month-end or quarter-end closes for the price side | Yahoo Finance historical data Use adjusted closes so splits and dividends do not create false breaks. | Free |
| Spreadsheet Run the pack's tracker and keep your written thesis next to the numbers | Microsoft Excel, LibreOffice Calc or Google Sheets The workbook uses standard functions (CORREL, SLOPE, OFFSET, INDEX). | Freemium |
The formulas
-
Price-to-fundamental gap and its z-score
gap_t = ln(P_t / F_t); z_t = (gap_t − mean(gap_1..t)) / stdev(gap_1..t)- P_t — price level in period t
- F_t — fundamental level in period t (earnings, loans, issuance indexed above zero)
A rising z means price is running ahead of the fundamental relative to the series' own history. It says nothing about why.
-
Lagged cross-correlation in both directions
P→F = corr(rP_{t−k}, rF_t); F→P = corr(rF_{t−k}, rP_t); loop = (P→F + F→P) / 2, each over the last W periods- rP, rF — log changes of price and fundamental, ln(X_t / X_{t−1})
- k — lag in periods (1 by default)
- W — rolling window in periods (12 by default)
Positive in both directions is the signature of a self-reinforcing loop. A common driver such as interest rates can produce the same pattern without any feedback.
-
Rolling feedback slopes
βP→F = slope of rF_t on rP_{t−k}; βF→P = slope of rP_t on rF_{t−k}, over the last W periods- Same inputs as the correlations
The slope adds size to the sign: how much the fundamental moved per unit of lagged price move, and the reverse.
-
Stage rules (the pack's stand-in)
CRASH if trendP ≤ c, shortP < 0, loop ≥ θ; REVERSAL if trendP < 0, shortP < 0, z > 0; TWILIGHT if trendP > 0, z ≥ zH and (trendF ≤ 0 or loop < θ); ACCELERATION if trendP, trendF > 0 and loop ≥ θ; INCEPTION if trendP, trendF > 0 and loop > 0- trendP, trendF — log change over W periods; shortP — log change over m periods
- θ — loop threshold (0.3), zH — stretched gap (1.5), c — crash threshold (−0.15)
Evaluated top to bottom. An operationalisation built for this pack, not a formula from the book; labels lag by roughly one window.
Applied Pack · free members
Soros Applied Pack
A reflexivity feedback-loop tracker for your own price and fundamental series: which way the feedback runs, how stretched the gap is, and a stand-in boom-bust stage label.
- Soros_Reflexivity_Tracker.xlsx — READ ME, an Inputs sheet (window, lag, thresholds), a Series sheet with live formulas for log changes, the price-to-fundamental gap z-score, lagged correlations and slopes in both directions and the stage label, and a Stage Summary sheet (EXAMPLE rows included)
- soros_reflexivity.py — stdlib-only CLI: reads a price/fundamental CSV and prints the same measures and stages per period, with the same results as the workbook
- series_sample.csv — 108 monthly EXAMPLE_ rows of a synthetic boom and bust in the expected format
- README.txt — inputs, formulas, stage rules, limits, how to run, and the educational-use disclaimer
Pack security
Macro-free Excel · plain-text Python you can read before you run it · no installers, no network access — the code works only on files you provide. Served only from closelook.net; we never distribute through download portals or email attachments. How to verify in 30 seconds →
SHA-256 7ab7b9a8d46ec6c234ef13ab1b417054c65011ca61491bffc3aee55ede91b899
The pack is a free-membership download — no card, free forever. Membership also delivers the Daily Pulse and Weekly Signal to your inbox.
Join the Look — free → Already a member? Sign in →
Educational templates — a research diary companion, not investment advice.
Closelooknet publishes a market diary, not investment advice. This condensed read restates the book's ideas in our own words for education — for the author's full argument, go to the source.