Quant Memo
Core

Living Inside Your Risk Limits

A risk limit only works if you treat it as a wall you plan around, not a line you find out you've crossed after the fact.

Prerequisites: Sizing A New Trade From Scratch

Every trader is handed a set of numbers on day one: a maximum position size, a maximum loss for the day, a maximum amount of capital at risk across the whole book. The numbers themselves are simple. What separates a trader who lasts from one who doesn't is whether those numbers are treated as a planning constraint that shapes every decision beforehand, or as a fence you discover only when you've already walked into it.

What a limit actually is

A risk limit isn't a punishment waiting to happen — it's a pre-agreed statement of how much the desk is willing to lose finding out whether a given idea is right. A trader given a $2m gross limit and a $150k daily loss limit isn't being told "don't lose more than this," exactly; they're being told "this is the size of bet the firm has decided this strategy deserves, given what's known about it so far." Living inside the limit well means using the full $2m when a trade is good and sized correctly, not treating $1.2m as the real ceiling out of caution, and not treating $2m as a target to hit regardless of trade quality.

A day inside the limits

A trader running a $3m gross limit starts the morning flat. A signal fires on Stock A, sized to $900k based on conviction and volatility — well inside the limit, leaving $2.1m of room. Two hours later a second, unrelated signal fires on Stock B, sized to $1.1m. Gross is now $2m, still inside the $3m ceiling. Early afternoon, a third opportunity appears in Stock C, sized on its own merits to $1.3m — but adding it would take gross to $3.3m, over the limit.

The undisciplined version of this trader shaves the sizes of A and B slightly after the fact to "make room," or just breaches the limit by a bit because $3.3m against a $3m limit feels immaterial. The disciplined version either sizes C down to fit inside $1m of remaining room, or skips C and waits for A or B to be reduced first. The limit was set before any of the three trades existed and doesn't get renegotiated because a good idea showed up late in the day.

SituationLimit-respecting moveLimit-eroding move
New signal would breach gross limitSize it to what's left, or passTake full size, breach "just a little"
Existing position moves against you, eating marginReduce elsewhere or add hedgesWiden the internal limit to match the position
A trade you're confident in is capped by the limitAccept the cap; conviction isn't a limit overrideAsk for an exception mid-day, informally, without review
Day's P&L is down $140k against a $150k daily loss limitCut risk, stop initiating new tradesKeep trading, reasoning "it's not breached yet"

A risk limit sized correctly in advance and respected in the moment is a tool for taking full advantage of good ideas within a known worst case — not a target to graze against or a formality to work around once a good trade shows up.

Soft limits and the discipline of stopping early

Good desks build in a soft limit below the hard one — say, act as if the ceiling is 80% of the real number. That gap exists because positions move between when you check the limit and when the trade actually executes, and because a trader who treats the hard number as the real number has no room left when a genuinely good, unplanned opportunity appears. Using only 80% of your capacity on an average day isn't leaving money on the table; it's the reason you have room to act decisively on the days that actually matter.

What breaks when limits become negotiable

The moment a trader breaches a limit once and nothing happens — no conversation, no consequence — the limit stops functioning as a limit. It becomes a suggestion that gets re-evaluated trade by trade, which defeats the entire purpose: limits exist precisely so that in-the-moment conviction doesn't get to override a decision made with a clearer head, in advance, about how much the desk can afford to lose finding out if this trader's judgment is any good.

Related concepts

Practice in interviews

Further reading

  • Green, Managing a Trading Desk
ShareTwitterLinkedIn