How your posture changes what the desk does

Choosing a posture is not choosing a set of limits with a mood attached. It is telling the desk what job it has. Pick Aggressive and the Portfolio Manager's instructions open with: maximise return over the next two months inside the hard limits; take qualified risk; being under-invested while signals are positive is a failure of this mandate, not prudence. Pick Moderate and they open with trade like a normal active manager: invested most days, a handful of names plus the core sleeve, a qualified setup is a starter the same session. Pick Preservation and they say protect capital, act rarely, most days end with no trade by design. The sentence is followed by the numbers on the Profile card (the target share of the account to keep invested, the blend score that obliges an entry, the risk per idea, the smallest starter, what a caution regime or a data release does to size, whether it adds to winners, how long it holds, and how much of what is invested sits in broad funds rather than picked stocks (all of it under Preservation, falling to a little over half under Aggressive and a tenth under Speculative)), and those same numbers are what the code sizes orders with. The mandate, the card and the arithmetic cannot disagree, because all three read the same set of numbers.

What a session does with it

Every trading session the PM is handed, before anything else, a decision inputs block: how invested the book is against the posture's target band (and the gap in dollars if it is under), the market regime, every Guidance rule in force (hard rules marked as enforced), the Quant desk's open recommendations, the Researcher's latest brief with its checkable calls, and one row per name the desk could act on: the blend score against the posture's act line, the top algorithm votes with their weights, Jim's setup read, whether the armed trigger is met, the posture's dollar size for a starter, and the brief items, desk recommendations and Guidance rules that touch that name. A name whose trigger is met and whose blend clears the act line, that can be sized and that no hard rule blocks, is qualified.

Under Conservative, Moderate, Aggressive and Speculative, a book that is under its target band with qualified names on the block owes entries that session. The PM either proposes them, sized by the posture's dollars on the row, or names the one hard reason it could not: trading is halted, a hard Guidance rule forbids it, the regime is risk-off, the book has no room left under its caps, or the Executor's own preview refused the order (a hollow quote, an earnings blackout). A measured reason is accepted too, when it comes with its number: the expected move is too small for the stop it needs, the cost of getting in and out is at least the expected edge, or the week's allowance of quick round trips is used up. "Caution", "a print tomorrow", a bare "too small" and "waiting for more confirmation" are not reasons outside Preservation: caution and events make the starter smaller (halved under Conservative and Moderate, trimmed by a quarter under Aggressive), they do not cancel it. Conservative alone stands aside when a caution regime and a release inside a day coincide. If a session ends with nothing proposed and no accepted reason, it is asked once more in the same run; if it still abstains, the Critic files a mandate not followed finding that the next session must answer, and it shows up on the Overview mind card like any other Critic flag.

Every proposal, and every decision to pass on a name, carries the inputs it used. In the Inbox and in the session view on Desk › Portfolio Manager they appear as Why chips under the thesis: supports XOM trigger met, supports brief call, sizes it aggressive sizing, against regime caution, veto guidance rule. A proposal that cites nothing that supports it, or a pass that ignores a desk recommendation or a soft Guidance note on the same row, is sent back once before it is accepted. A hard Guidance rule is never argued with: an order against it is refused in code before the batch is recorded, and the Executor would refuse it again.

How many new ideas a session may open

The posture also sets the pace. A session may open at most one new idea under Preservation and Conservative, three under Moderate, five under Aggressive and eight under Speculative; the Critic counts what a session opened against that number and files a finding when it goes over. Pace is a ceiling, never a quota: a session that finds nothing worth owning proposes nothing, and that is not a breach. If you have told the desk to take more risk on Strategy › Guidance, the session may use its whole order allowance for new names while the instruction stands (three under Conservative and two under Preservation; the other postures already allow it), and the Critic counts against that number (Tell your desk).

A name the Quant ranks near the top is not an entry just because of its rank. It still has to clear the act line, fit inside every limit and pass the Executor's preview, and normally its trigger has to be met too. There is one exception: when the book sits well under its target band in a risk-on market and no trigger has confirmed that session, the PM may open small starter positions from the top of the ranking. Those cards say Starter position and explain that the book is under-invested.

How to tell it is working

Switch a paper book from Moderate to Aggressive on Strategy › Profile and the next intraday session should read differently in three places: more names qualify (the act line drops from 0.15 to 0.10), each starter is larger (0.75% risk per idea, a 5% floor, an order cap of 10% of the account and never under $1,000, only a quarter off in a caution tape), and the exposure line on the PM's card reports a bigger gap to a 70–90% band, so more entries are owed. Switch to Preservation and the same tape produces no entries and a card that says why per name. The platform's own tests run exactly this comparison (the same day under the four standard postures) and hold each to its band; the Critic reads the mandate at the top of every review and judges the session against it rather than against generic caution.