Splitting P&L Into New Trades And Carry
Every day's P&L is really two P&Ls glued together: what you made deciding today, and what you made just holding what you already owned. Mixing them up makes good books look lucky and bad books look fine.
Prerequisites: Reading Your Daily P&L
Ask a portfolio manager how the book did this week and you'll get one number. Ask what drove it and the honest answer splits in two: P&L from decisions made this week — new positions, adds, trims, exits — and P&L from decisions made earlier that are simply still sitting there. The second kind is called carry P&L, and if you don't separate it from new-trade P&L, you cannot tell whether you're a good picker of new ideas or just holding a few old winners that happen to keep drifting up.
The split matters because the two numbers get judged differently. New-trade P&L tells you whether this week's decisions were good. Carry P&L tells you whether your existing book is still working — and a book can carry well on autopilot for months after the manager has stopped adding value.
Doing the split
For any position, carry P&L is the price change on the shares you already held coming into the day, times yesterday's size. New-trade P&L is everything else — the shares you bought or sold today, marked from your fill to the closing price.
In words: carry asks "what did my existing shares do today," and new asks "how did today's trading do relative to where I could have closed it."
Worked example
You come into Tuesday holding 40,000 shares of a stock, bought earlier, closing Monday at $61.00. Tuesday it closes at $62.10 — that move alone is carry P&L: , so $44,000.
Mid-Tuesday you also add 15,000 shares at $61.50, which close at $62.10. That's new-trade P&L: , so $9,000.
Total P&L on the position for Tuesday is $53,000. A manager who only looks at the total sees a good day. Split it and you see the add contributed $9,000 on $925,000 of capital deployed — a fine but unremarkable trade — while $44,000 came from a decision made days ago that this week required no skill to collect.
Why it changes what you do next
If new-trade P&L is consistently weak while carry P&L is strong, the book is living off legacy winners, and the manager's live decision-making isn't being tested by recent numbers at all — a run of flat new trades can hide behind a rallying core holding for a long time. If it's the other way round, carry losses from stale positions are eating gains from good fresh calls, which usually means positions are being held too long after the thesis has already played out.
Separate what today's trading earned from what yesterday's positions earned by simply holding still. A book that only carries well isn't being actively managed — it's just long something that's working.
Don't confuse carry P&L here with "carry" the strategy (rolling yield, financing cost, dividend accrual). This carry is purely "P&L from size you already had," whatever asset class you're in — the arithmetic, not the trade type.
Related concepts
Practice in interviews
Further reading
- Grinold & Kahn, Active Portfolio Management (ch. 6)