Investing

Investing Mental Models: Key Ideas


Premise

Edge lives in the gap
between value and price.

Seven mental models for markets, risk, and edge — each a single idea to play with. The thread: returns come from the distance between what is true and what is priced, and the discipline to act only when you can name that distance.

Position Sizing

Size the position to the conviction.

Most investors diversify their conviction away — they find a mispricing, believe it deeply, then allocate a standard 2% anyway. The edge is calibrating capital to clarity, then bearing the volatility concentration brings.

  • Undersize and you survive, but the right call is too small to matter.
  • Oversize and the thesis can be right while a drawdown still ruins you.
  • Strong opinions, loosely held: cut the size the moment the thesis fractures.
Drag to size the bet. Watch the portfolio path move between under-allocation and ruin.
starting capital VALUE REPEATED DECISIONS

Enough to matter, not enough to ruin.

On Waiting

Most of the return is late.

Ninety-nine percent of the time, you make money by waiting. The reward structure is inverted from the rest of work: most of the return is generated by doing nothing for long stretches.

  • The hard thing is not finding a 10× — it's not selling once you have it.
  • Selling early locks in the visible gain and forfeits the nonlinear tail.
  • Know the terminal value, or every large gain reads as "take profits."
Move the exit. See what you capture — and what compounding you leave behind.
early late missed VALUE

Taking profits early is the expensive instinct.

Market Efficiency

Eventually efficient. Not immediately.

Two propositions that sound contradictory but aren't. Prices take time to reflect information; over enough time they converge on fundamentals. The lag between the two is the entire basis for active investing.

  • Not immediate: biases, constraints, and capital flow open windows.
  • Eventually: mispricings get arbitraged; narratives give way to earnings.
  • The gap between eventually and immediately is where returns live.
Sweep time forward. The mispricing — the gap between price and value — is the opportunity.
fair value market price not immediate eventual

The gap narrows as information diffuses.

The ASI Endgame

Free intelligence collapses the gap.

A thought experiment: superintelligence arrives, rational, with capital. Markets become immediately efficient. The very gap returns lived in shrinks toward zero — and alpha thins.

  • Every edge exists because your counterparty's model is incomplete.
  • Incompleteness is the resource you harvest — the one AI depletes.
  • Edge migrates to what intelligence can't reach: private information, relationships, regulatory position, compute and energy.
Raise AI penetration. The delay window closes, public alpha thins, and edge migrates to what code can't reach.
info known priced in DELAY = ALPHA private info relationships regulatory position compute & energy EDGE MIGRATES TO

The last durable trade may be owning the efficiency itself.

Three Things

Where, why-not-yet, and when.

To make money, understand three things about a mispricing. Without all three, a gap can persist indefinitely — and an indefinitely deferred payout has a present value approaching zero.

  • Eventual valuation — what the math says it's worth at terminal state.
  • The nature of the gap — ignorance, time preference, constraints, or real uncertainty.
  • The catalyst — earnings, a narrative shift, regulation, capital reallocation.
Pick a question. The first tells you where, the second why not yet, the third when.
the gap price today eventual value catalyst

A mispricing without a catalyst can wait forever.

Outcomes vs. Process

Judge the bet on process, the book on outcomes.

The litmus test of a single position is the risk-return tradeoff, not the result — judging a strategy by one outcome confuses noise for signal. But a portfolio is a population of samples, and over enough bets the law of large numbers is your ally.

  • A good bet can lose; one outcome is too noisy to judge a process.
  • "A 10% chance of a 100× payoff — take it every time, and be wrong 9 of 10." — Bezos
  • Over a large book, positive expected value must show in aggregate.
0 bets · avg —
Each bet: a 90% small loss or a 10% large win — positive expected value. Watch the running average converge.
break-even E[V] > 0 RUNNING AVG NUMBER OF BETS

The single bet can be right and lose. The book should not.

On Contrarianism

Contrarian without edge is just confident and wrong.

The valuable contrarian position has a specific, articulable reason the consensus is mispriced — usually that the market is right about the facts but wrong about the interpretation, or right near-term but wrong about the terminal state.

  • Being contrarian without an edge is just being wrong with more confidence.
  • Consensus exists for a reason — it's usually right about direction.
  • That's why momentum works, and buying all-time highs can be sound.
Hover or click the quadrants. Only one corner — against consensus, with a real edge — pays.
WITH CONSENSUS AGAINST EDGE NO EDGE informed agreement valuable contrarian the crowd confidently wrong

The market correct on facts, wrong on interpretation.

Takeaway

Find the gap.
Size it. Wait.

Edge is rented, not owned — it exists only while someone else's model is incomplete. Name the gap, size the bet to your conviction, hold through the compounding, and judge the book, not the bet. Then watch the resource you harvest get scarcer.