01 Win-rate −8 pts
The edge thins
Same seed. Eight fewer winners in a hundred — not a new world, a thinner one.
—
from — on the base tape
P5 — base —
—
Edit any of these and you are testing your own limits — press Run again to see the verdict move. The four dials above stay the simple first read.
A living grid of closes lives on the Fills tab. Drop a CSV there and the blur sits — those averages write the three dials above. This drawer does not take a second file.
A CSV of closes belongs on the Fills tab. The four dials plus the limits above are this run. Saved already keeps prior readings on this device.
Set the four numbers, then run. Your survival odds and the equity fan resolve here — computed on your device, from your hand.
The number above is this reading. Wine is that same fail share — not a lighter scan. The curve only moves risk, same engine, fewer paths per tick.
Wine pin = this reading. The curve sweeps other bet sizes on the same engine — not a second book.
Wine is this reading. The curve is the same engine at other bet sizes, not a second book. Computed on this device · not investment advice.
Drop a CSV of closed trades — or try the sample. When it locks, these cells become the file. Nothing is sent.
Stays on this device. Nothing is sent anywhere. Need a numeric PnL or R column. The free run uses averages — not a replay of the fills.
The verdict is one number from this reading. Underneath it: two forces. First, the order a two-point deck happens to arrive in. Then, why a drawdown is so much heavier than it looks.
—
—
Two books. Same win rate, same +0.30R average, same 6% of the account on every trade. One takes a small clean payoff. One takes a hero payoff. Their advertised average is identical — $4.32 million after 250 trades. Watch who actually arrives. This act is an illustration, not this run’s verdict.
The charts above show where the runs go. Here is the harder truth underneath them — what living inside this system is like even when the edge is real, and the one number everyone trusts that says nothing about whether you survive.
—
—
Three honest stresses, computed live from the same engine and the same seed family as the run. Each card is base survival against the stressed tape, with the P5 floor on both. Wine marks the base reading — drawn the same whether a stress still holds or not.
If it only lives on the kindly tape, it is not a plan.
Same seed. Eight fewer winners in a hundred — not a new world, a thinner one.
—
Same seed. Each winner pays three-tenths of R less. The losses stay exactly as they were.
—
Same seed. One extra −2R every ~20 trades — a regime sketch, not a claim about a named market.
—
One pair of sliders. The seed stays put. Light numbers on drag; the full tape on settle. This is how you see what robustness builds — an edge that still stands when you write the tape worse on purpose.
A plan is what remains after the tape is less kind. That is what robustness builds — not a kinder story, a spine that still stands.
Illustrative · dated 2026·08·15 · not advice. Shock day is a simplified regime (one extra −2R every 20 trades), not a claim about a named market. Wine marks the base reading, drawn the same on a hold or a move. Reads for any trader — prop, independent, desk, or own book.
Optimize · one lever · your floor
You already have a run of this reading. Name a P5 dollar floor, pick one lever, and the lab finds the smallest setting that puts the 5th-percentile finish on that dollar — or tells you this lever cannot. Apply writes the number. Run confirms it. Drawdown and daily limits stay on; the profit-target stop is lifted so the floor can grow.
Name the floor in dollars. The lever below is what must move to hold it.
Reading the lever…
The scan starts from this run. Apply writes one number. Strategy Run is the confirmation.
Apply does not re-run. Confirm on Strategy so the diagnosis and this lab stay one fact.
This reading already has a free verdict. This sheet is the prescription — then the stamped Situation Report · $29 · one-time · one month on one checkout.
The same 2,000 paths. A different survival — and a floor you can plan around.
A simulation, and a sum you could do by hand. Same reading.
The free tool shows a survival percentage. The document you keep is corroboration: 2,000 seeded paths of the improved strategy, next to arithmetic that never ran the simulation.
Situation Report · $29 · one-time · one month · Gumroad is the merchant of record · 2 devices · we enforce · 14-day refund · free reading stays free · full licence strip on /buy
The free reading above stays free. Pay only if you want the stamped Situation Report.
The key from your purchase email opens the Situation Report on this device. Nothing about your trades is sent — only the key and the product id.
This is the product’s existing document — not a second plan. The five acts mount here from the same engine that named the move above.
Illustrative simulation from your own inputs · not investment advice · computed on your device · fixed seed, reproducible run. The move shown is the input most likely to move your survival on these paths — consider testing it in your own trading before you rely on it. No outcome is promised. No alert or watch service is offered. Prices are those charged at checkout.
Saved · this account
Type the three numbers the platform already shows. The instrument moves as you type. Read starts the same engine from here — this reading, not day zero. Nothing is sent.
The four dials stay put. That is the strategy you named. Tracking never silently retunes risk.
Equity, days used, high-water. You type what the platform already shows. No import required.
Survival % and P5 from here, not from day zero. The spine only grows when you read again.
Type the three numbers. The rail is the remaining book — floor, now, target.
Awaiting a read. The figure is this stand’s survival — the same engine as Strategy.
Stand moved — read again to refresh the figure.
Read to name the rule closest to ending the account from this equity.
2,000 paths · remaining calendar only · nothing is sent
Illustrative · dated 2026·08·16 · computed on this device · 2,000 paths · fixed seed from the four dials. Not a watch service, not a firm bulletin, not investment advice. Wine marks the latest look, drawn the same on a hold or a move. Personal by default — a named firm is opt-in and must run that firm’s dated numbers.
Three systems on one desk. The wine mark sits on the survival winner — even when another system prints more on average. That is the number you plan around.
Compute the three systems to see which one stays in the game.
Writes to this device if a store is available. Nothing is sent. No alert is promised.
Illustrative model, computed on your device from the four inputs · dated 2026-08-15 · Monte-Carlo, 2,000 paths, fixed seed · the same method on a firm’s book or the floor you name yourself · not investment advice · PropSurvival is independent of every firm named elsewhere in the product.
Kept on this device only — no account, no network. Some browsers clear idle site data — · ·
Safari can clear this site’s data after about 7 days idle. ·
| Strategy | Firm | Survival | Binding rule | Saved | Path |
|---|
Every other screen hands you a survival number for this reading and moves on. This one shows the workings — the vocabulary that makes the number mean something, the method that produces it, and the proof that it isn't a guess. Rigor you can read is the reliability you're relying on.
Reference reading · dated 2026-08-14 · Monte-Carlo engine · fixed seed · computed on your device · illustrative models · not investment advice
This is the only place the technical terms live — and none of them stays abstract. Each carries the small piece of arithmetic that makes a trader see why their survival is what it is.
Rank one hundred simulated futures of your strategy from worst to best. P5 is the fifth from the bottom — not the average you hope for, but the bad-but-not-freak outcome you plan around. A strategy survives when even its P5 clears the floor.
95 of 100 futures land to the right of the P5 line. The whole product is built around that one line, not the peak.
Every result measured in units of the money you risked on one trade. Risk $200, make $400, that's +2R. One yardstick, so a $50 account and a $50,000 account read the same.
A loss limit measured from your account's high-water mark. Profit and the floor rises with you; give it back and it stays where it climbed to. The floor follows you up — it never comes back down.
Given your average win and loss in R, the win rate that leaves you exactly flat. A 95% win rate can still bleed to red — one loss can give back 25 wins. A bigger average winner lowers the bar you need to clear.
The average R you keep per trade net of cost: (win% × (avg win − cost)) − (loss% × (avg loss + cost)). With the desk's default 0.05R commission that is gross expectancy minus 0.05R. Positive net expectancy is the edge itself. Everything else on this page is about whether you survive long enough to collect it.
The chance the account hits its floor before its target. It doesn't rise smoothly with bet size — a real edge that's near-safe at 1% risk can see ruin jump from ~3% to ~63% at 2%. A cliff, not a ramp.
The share of a trading life spent below a prior high. An edge that ends up profitable essentially every long run still spends about 71% of its life underwater and loses 7 in a row as routine. Normal — not the edge breaking.
The survival number is not a score handed down from a model you can't see. It's the result of dealing your own trades, over and over, against a firm's dated rules — and counting.
Each run deals your trades one by one — a win adds your average winner, a loss subtracts your average loss, both in R — and stops the moment the account either clears the profit target, crosses the trailing drawdown below its high-water mark, or loses more than the daily cap in one session.
Same inputs, same seed, same result. The seed and path count are recorded on every export, so the finding is an instrument reading, not an opinion — anyone can re-run it and land on the same number.
Survival is always measured against a specific rulebook. These are read from a firm's published materials and stamped with the date they were read — never asserted as permanent fact.
| Rule | Reading | As of |
|---|---|---|
| Profit target | +$3,000 | 2026-07-14 |
| Trailing drawdown | $2,500 | 2026-07-14 |
| Daily loss cap | $2,000 | 2026-07-14 |
| Evaluation window | 120 trades | 2026-07-14 |
| Minimum trading days | documented · not in this sim | 2026-07-14 |
| Consistency rule | documented · not in this sim | 2026-07-14 |
Because every value carries a date, a change shows up as a changed date — that's exactly what the Saved workspace watches for the strategies you keep. Trading on your own limits instead of a firm's? The engine binds to whatever floor and target you name — the method is identical.
Verify current terms with the firm before you tradeFive things separate a survival reading from a marketing figure. Each is something you can check yourself — nothing here rests on taking our word for it.
Risk of ruin for one edge as bet size grows, solved two independent ways: a closed-form equation (the line) and thousands of simulated paths (the points). If the simulator were guessing, the points would drift off the line. They don't.
Model: a ±1R walk with a fixed edge, absorbing at ruin and at target · simulated with a fixed seed, reproducible · illustrative model, not investment advice.