Reading the Evidence tab

Performance › Evidence is the tab to open when someone asks "does it work?". A control at the top right of the tab switches between two views. Proof so far, the view this article reads with you, is what used to be the separate Evidence page; older links to that page still open it here. Backtests (what the rules alone would have done) is read in Evidence before money.

Proof so far answers in three parts and never mixes them: what is proven about the desk today, what only time can prove, and what none of it proves. Every card leads with a claim of a few words, one figure, the amount of history behind it and the date it was measured; How we know under the figure opens the rest. A figure that does not have enough history yet says how far along it is instead of printing a number, and a figure that was computed but is not fair to publish says withheld and why.

Where

Performance › Evidence, the Proof so far view

Who can open it

Everyone signed in to the book, including people you shared it with as read only

What it reads

This book's own daily results, its rules twin if one exists, and the platform's committed test records

What it changes

Nothing. It is a read-only view

The one sentence

The line under Only time proves says how old the live record is; the newest machinery record's date sits beside Proven today

What is proven today

Four cards, each a fact about the machinery as of the date printed on the card, each re-computable from the platform's own committed records (we can re-run any of them in front of you). A card's claim is built from its record's own counts, never typed by hand. A card whose record is not loaded says so by name and states nothing.

• Every idea tested stays on the record. The claim says how many trading ideas were tested and how many beat simply owning the same stocks; the figure under it repeats that count and says how many were retired as dead and stay on the record. Behind How we know: every signal the platform ever backtested is on a register, the dead ones included, measured on the stocks as they were listed at the time and after trading costs. It counts how many were measured, how many were retired, how many survive, and how many configurations were ever tried, because every extra try makes a lucky result more likely and the register corrects for it: the best edge anyone measured is printed with its deflated Sharpe, the figure that discounts all those tries (0.95 or more would make luck an unlikely explanation). The latest research round's sentence is built from the round's own counts (a base signal plus the additions judged against an equal-weight basket of the same names), so the day a signal survives the card says that instead.
• Every role is examined. Each desk role (the PM, the Researcher, the Critic, the Trader, Consult, and the whole desk together) runs a competence suite: does it read what it must before it acts, use its tools correctly, handle a failed call or a stale price, stay grounded in the numbers it was given. The claim says that every role passes, that one role scores below its pass mark, or that the live model fails one of its own tests; the figure is the live run's score out of 1.00 with the count of scripted runs that pass. A failing live gate is never rounded up: it outranks green scripted rows in the claim and in the pill. Behind How we know, the table lists each role's sessions and score; scripted rows measure the harness with a stand-in, live rows measure the real model on the same recorded day, and a failing check is named.
• Readiness is a list of checks. The readiness ledger defines "safe on paper", "production quality" and "real money" as executable checks. The claim and the figure say how many are proven today, the pill says whether real money is a yes or a not yet, and whoever runs this installation can open the full ledger from the card. Behind How we know: how many pass today, how many are known gaps, how many are checked and failing today ("not yet"), and how many have no proof written yet, by area, in a sentence that names each count in words. "No proof yet" always counts as not proven.
• The live mandate run. The latest live run puts the model through every posture on one recorded market day and counts sessions that stayed inside the posture's limits and any breach; the claim is built from those counts (the live model "stayed inside its limits in 40 of 40 sessions", or "broke its limits" so many times when it did). Behind How we know: the count per posture, and how a change to a role's behavior ships. It is trialed live, several runs with the change against the same number without it, and only a proven change ships; the inconclusive and dead trials are counted too.

None of these four is a claim about money. They say the instrument is built correctly and reports honestly.

What only time proves

Four figures, each in its own card with a state pill; the fourth (up and down capture) prints once SPY has had at least ten up days and ten down days in the record. Each card's claim names its subject and its state ("still collecting", "too early to judge", "too close to call") and carries a direction word only when the figure has been judged from its interval. Under the claim: one figure, then the sample line and the date.

• Return vs SPY at the book's volatility (on screen: against the S&P 500 at equal risk). The book's return since its first funded day next to SPY's return over the same days, with SPY scaled so its day-to-day swings match the book's. Behind How we know: the difference between the two with its 95% interval (whole weeks of daily differences resampled, the same method the weekly Desk minus Rules figure uses). A direction is read only from that interval and only after 63 trading days: the "Reads:" sentence says "not yet distinguishable from SPY at the same risk" while the interval includes zero, "ahead" or "behind" only when the whole interval sits on one side, and "too short to judge" before 63 days whatever the figure shows. The note under it says what SPY itself returned, the scaling used and how many weeks the interval resampled.
• Worst fall from a high vs SPY. The deepest peak-to-trough fall of the book against SPY's over the same days, and the gap between them. Two single paths carry no interval, so this card never draws a conclusion from the gap at any age, and its claim says so.
• Desk minus Rules, per week (the card that sets the model against its own rules). If a rules twin runs beside this book (a rules-only paper copy trading the same signals mechanically, with no AI PM), the weekly difference between the two is the one honest measure of what the Portfolio Manager's judgment adds. It prints as a percent a week with its likely range and one of three readings: not yet distinguishable from zero, the desk is ahead, the rules are ahead. Whether the model keeps the Portfolio Manager job is decided at 13 counted weeks; How we know says so in those words.
• Up and down capture vs SPY (on screen: share of the market's moves caught): on the days SPY rose, how much of its gain the book kept on average; on the days it fell, how much of its loss the book took. Two averages with no interval, so no verdict; 100% on both would mean the book simply moved with the market. (Computed on daily returns; the monthly convention needs a year of months.)

When a figure prints, and when it is judged, the fold under the four cards, repeats the floors in one paragraph.

The states

Every card carries its sample line ("71 trading days, Sep 15 to Dec 23, 2026" or "9 weeks counted") and "as of" the last day in the sample, under its figure.

What this does not prove

The last section is four plain sentences that carry no number and no register fact, so they can never contradict the cards above: a backtest is not a live result; passing evals prove correct behavior, not profitable judgment; the scoreboard is the forward record, reads collecting or early until it has its sample, and reads a direction only from an interval; paper fills are simulated and live-broker slippage is measured separately before real money.

At the foot of the view, the owner of the book finds Readiness for live trading, the way to the go-live scorecard, and everyone finds How to read this tab, which opens this article.

Where the numbers come from

• The live figures are computed from this book's own end-of-day results and SPY's daily closes over the same days, the same definitions the rest of Performance uses. The matched-volatility line uses one volatility estimate for the whole platform (an exponentially weighted one over at least 20 sessions).
• Desk minus Rules is the latest weekly attribution report written into this book; the Track record tab and the Backtests view are the two older halves this view sits on top of.
• The machinery cards summarize committed records (the register, the eval suites, the live mandate run and its trials) as of the date printed on each card, and the readiness counts come from the latest readiness run loaded for this installation (Admin › Readiness for whoever runs it). The register summary is regenerated in the same command that regenerates the register, and the platform's own change gate refuses a summary that is older than its sources.