We published the condition that would end this strategy — and handed the off-switch to someone else.

Most people should close this tab.

The most exclusive thing a page can do is tell you, plainly, whether it is for you.

• ▹ You evaluate managers by their failure modes, not their best month.
• ▹ You'd rather see a published kill-condition than a glossy deck.
• ▹ You think in decades and doublings, not quarters.

If you're looking for a hot tip or a guaranteed number, close this tab. This is the opposite of that.

When it wins and when it loses, the amounts are roughly equal — but it wins almost twice as often.

That sentence is the whole strategy. Everything else on this page is either proof that it is true, or architecture that keeps it alive.

Wins and losses are similar in size — and wins arrive meaningfully more often. The gap between them is the edge: measured live, size-aware, not modeled.

Historical, size-aware, across the live record; the sample is pre-asymptotic. These describe the strategy's measured behavior, not a forecast of any investor's outcome.

Don't take the number. Watch it print.

An edge can be validated three ways: watch it print, try to break it, and publish the condition that would kill it. The first two are below. The third is §04.

Delete the three best trades and the edge still stands — expectancy holds, profit factor stays above breakeven. That one fact dismantles "you just got lucky" before it's spoken.

Delete the three best trades — the most common way a small sample flatters itself — and expectancy still holds, profit factor still clears breakeven.

What survives a small sample is not a win rate — it's the architecture beneath it: a published falsifiability gate, a witness countersignature, and operator capital on every fill. Those are the durable facts. The raw figures live — and accumulate — in the open.

Any performance shown is net of fees. Past performance is not indicative of future results. The sample is pre-asymptotic — statistical confidence grows with trades.

A few dollars of edge per trade is a rounding error. Run it twenty-seven times a month, and it stops being one.

The edge isn't large. It's repeatable. That's the difference between a savings account and a snowball.

1. One trade — Risk one unit; on average, expect a small positive edge on top. Invisible to the naked eye.

2. One month (~27 trades) — The same small edge, many times over. It begins to show up against the capital at risk.

3. One year (~330 trades) (extrapolated, illustrative — not a forecast) — Many repetitions of one small edge. Then position-sizing on a growing base bends the line further: the math stops being linear.

"Find a long hill of wet snow." The edge is the snowball. The repetitions are the hill.

Cadence is context, not a forecast. The roughly twenty-seven trades a month and roughly three hundred thirty a year are extrapolated and illustrative. No return is promised or implied.

Why being early is rational — and honest about stage.

You are underwriting an edge that is real but pre-asymptotic. In exchange for taking that earlier-stage risk — before the track record is long enough to be priced in, and before capacity closes — founding members receive terms and access that do not repeat.

1. The edge is capacity-constrained — it degrades past a size the market can absorb. This is exactly why Medallion closed to outside capital.

2. Therefore capacity is the scarce asset, and early capacity is the scarcest of all.

3. Therefore founding economics aren't a "discount." They are compensation for being early to a thing that closes.

The proof phase was the point — and it's complete.

The record was run deliberately at proof scale: enough to establish the edge exists, in public, without asking compounding capital to ride it before it was verified.

"Proven" refers to the validated edge and process, never to future results. Scaling does not reduce the risk of loss — capital can still be impaired (§06). Nothing on this page is an offer or a forecast of return.

Anyone can show a good record. Almost no one will tell you, in advance and in code, the exact condition under which they will admit it stopped working — and hand the off-switch to someone else.

Three mechanisms against the three ways capital dies.

The buffer means a loss can't ruin you. The Sit-Out means a bad streak can't compound. The Gate means a dead edge can't bleed you.

This is the part no one else publishes. Ready for the rest?

The Sit-Out and the Gate are process commitments, not guarantees. The Sit-Out bounds drawdown by design and adherence; it does not guarantee a maximum — markets gap. The Gate manages the risk of riding a decayed edge; it does not eliminate the risk of loss.

The operator runs the engine. You set the dial. Two separate decisions.

The edge produces what it produces. Your outcome is governed by one variable you control — how much capital rides on each trade. Return is a function of that dial. It is not a forecast, and it is not the operator's promise.

This tool sizes risk, not reward. It maps a chosen fraction of the Kelly criterion to a contract count and the drawdown that sizing has historically required — it does not project, forecast, or imply any return or profit, and none is shown. Figures derive from a young, pre-asymptotic live record; the Kelly fraction is computed from a payoff ratio and win frequency that will vary, and full Kelly is shown only as a mathematical ceiling no disciplined operator runs. Drawdown is scaled from the worst loss the record has already lived, not a maximum — deeper is possible and capital can be impaired (§06). Nothing here is an offer or solicitation; offers are made only through definitive documents to verified investors.

The investor picks the dial. The operator runs the engine. The drawdown you are willing to sit through is the only real input you bring.

A measured edge is worth nothing until capital rides it.

Exploiting an edge is two disciplines: run it many times at a risk-per-trade you choose (§02.5, §04.5) — and pick the vehicle your capital actually rides it through. Two are live today.

These are related strategies, described in plain terms; see each page for details. Nothing here is an offer or a solicitation, and nothing forecasts a return — offers are made only through definitive documents to verified investors. The founding cohort on this page is the single path to participate at founding terms.

Everyone quotes Buffett's first rule. Almost no one operationalizes it.

"Rule No. 1: Never lose money. Rule No. 2: Never forget Rule No. 1." — Warren Buffett

"Never lose money" is not a feeling — it is a set of disciplines. Below are the four we operate by. Each is evidenced earlier on this page, not asserted here. They are how we pursue Rule No. 1; they bound and manage the risk of loss — they do not eliminate it.

Each ring protects the inner core. Survival is the precondition for compounding. These four disciplines are how we pursue Rule No. 1 — and the very next section states plainly where loss remains possible.

"Never lose money" appears here only as Buffett's rule and the firm's operating discipline — never as a statement about your capital's safety, and never a promise that losses won't happen. The exposure structure bounds risk on the exposed slice; it does not remove it, and capital can still be impaired — see §06.

The asymmetry, stated plainly.

The form of the advantage is described here. Numbers, fees, and terms live behind verification.

Categories of advantage are described here, not priced. Specific terms, fees, and any target returns are reserved for the gated offering documents and are made only to verified accredited investors.

The manager who states the downside first is the one you believe about the upside.

Real scarcity. Nothing invented.

No countdowns, no "act now," no manufactured spots remaining. Two caps, both structural — and one of them is a hard number.

"Capacity is roughly 500 ES-equivalent contracts. Founding members reserve a slice of a finite book before it fills — provisioned throughput, not a promotion. After the cohort closes at a fixed member or committed-capital threshold, terms reset to standard. This is the structure, not a countdown." The cohort thresholds are confirmed in the briefing.

The 500-contract figure is an operational capacity limit on the strategy — the size the market can absorb before the edge degrades — not a performance claim, a target, or a statement of assets under management. Member-count and committed-capital thresholds are disclosed only in the verified briefing. Nothing here is an offer or solicitation.

Alignment beats charisma with this audience.