Reading the playbook

The playbook is the layer above the daily book. Per class, one artefact a day says what the market actually is, which measured edges apply, which strategy families are paper-testing with what numbers, and which are suppressed, with the reason. The daily book tells you what the ranking says today. The playbook tells you what to trade in what market, and everything it says traces to a measurement, never a backtest promise.

Three markets, three shapes

fx is a range market. VR60 reads 0.41, 98.1 per cent of 30 day windows close within +-10 per cent of where they opened, and the drift sits near zero. A market that keeps coming back rewards the fade, not the breakout.

Equities are a tails market. The published spread reads +6.6 per cent, re-measured +2.2 per cent after a split correction, and the day hits land 76 per cent in the tails. The middle of the distribution is noise; the top and bottom deciles carry the signal, so equities get bands, not fades.

Perps are crash reversion with an asterisk. Tails invert in the published window, which is the crash reversion pattern, but the measured year does not reproduce it. What ships there is sentiment, and the fade stays on watch.

fx: the fade, measured

The fx edge is the fade at the 2.0 z-score band on the majors: a long when the close is 2.0 standard deviations below its trailing 60 session mean, a short when it is 2.0 above, held 30 sessions or until the price returns to the mean, whichever comes first. Measured at 55.7 per cent hits and +27.9 bps net of fees across 1,214 trades, null p 0.00. Beyond the fade, fx is not measured yet: 2 backtest runs and 5 paper sessions is a measured nothing, so the fx families list stays empty until it earns one.

Equities: the tails bands

The tails bands ship with the daily book and read their numbers from the same T1 summary the book uses: hit rates, winsorised means and the null p per leg, at 7 day and 30 day holds. The reason equities get bands rather than fades is the shape: the rank correlation with the next month is slightly negative, so a fade at the centre of the distribution pays nothing, and the signal lives in the tails.

Perps: sentiment ships, the fade watches

Sentiment ships on the perp universe: +4.5 per cent over a 30 day hold on the liquid third, positive in quiet and defensive regimes, and it inverts in a constructive one, so the regime decides how the read is used. The fade does not ship. It fails its null after fees in all eight configurations measured, and the one constructive-bucket survivor is too thin to stand on: 37 overlapping days in two blocks. Watch means measured and not cleared, so the fade stays on the watch list until a forward window says otherwise.

Paper status is the honesty line

Every strategy family is in-sample until its forward twin. The playbook says so on every row: in-sample before the twin exists, then paper day N of 30 once it runs. The screen shows paper status rather than a backtest number dressed as a promise, because a backtest that has not met its forward twin is a question, not an answer.

Why sub-day families are suppressed

The 5m, 15m and 1h scalps, mom 1h and brk stay off the screen regardless of pnl, and each suppressed row prints its reason. The sub-day scalps print sub-day churn: trade-more families whose edges die to costs. The trend and breakout families print contradicts the anatomy: they bet on continuation where the measured behaviour is reversion. The measured doctrine says the edge lives at weeks: the fx fade holds 30 sessions, sentiment holds 30 days, the tails bands measure at 7 and 30 days. A sub-day family contradicts that doctrine, so the playbook suppresses it rather than show a grade that would flatter it.

How to read it

Open the playbook before the book. The anatomy line says which shape you are trading. The edges say what is measured, with the number. The families say what is proving itself, with its paper day. The suppressed list names what the doctrine rules out, with the reason beside each name. Read only, like the rest of the terminal: the playbook states the market and the measurement, and the decision stays yours.