Reading Your Daily P&L

The screen says the book is up $340,000 today. Is that a good day? You cannot answer without knowing three more things: how much risk the book was carrying, where the number came from, and whether $340,000 is inside or outside the range the risk model expected. A trader who only looks at the headline number is one good day away from missing that the desk was actually taking on risk it shouldn't have.

Start with risk-adjusted, not absolute

$340,000 on a book with $40m of daily VaR is unremarkable — roughly 0.85 percent of the risk budget, well within a single standard deviation. $340,000 on a book with $4m of daily VaR is extraordinary, and extraordinary numbers deserve suspicion before celebration. The first question every morning is not "how much did we make" but "how much did we make relative to how much we could plausibly have made or lost, given the risk we were running."

Split it before you judge it

The next step is decomposing the total into pieces that separately make sense:

Component	What it captures	What a healthy number looks like
New-trade P&L	Gains or losses on positions opened and closed today	Small relative to carry, unless it's an execution-heavy day
Carry / mark-to-market	Change in value of positions held over from yesterday	The bulk of most days' P&L on a book that isn't churning
Financing and fees	Borrow costs, commissions, funding	Should be a steady, boring drag — a surprise here means a booking error
FX / currency	Translation effects on non-base-currency positions	Should roughly match your hedge, not be a source of P&L on its own

A $340,000 gain that's $310,000 carry and $30,000 new trades reads as "the book did what it was supposed to do." The same $340,000 built from $500,000 of new-trade gains offset by a $160,000 carry loss is a completely different day — it means the existing book is bleeding and only fresh trading is masking it, which is worth knowing before it compounds.

Check it against what risk predicted

Every book has a predicted daily volatility from its risk model — say $1.2m for a given book on a given day. A $340,000 move sits comfortably inside one standard deviation. A $3.1m move on that same book is a nearly three-sigma event, and a three-sigma event on the day the model predicted $1.2m means either something genuinely unusual happened, or the risk model's inputs are stale — correlations that broke, a position that grew without the model noticing. Either conclusion is worth chasing down the same day, while the trades are still fresh in memory.

A worked morning

The book made $340,000. Breaking it down: $295,000 carry, $60,000 new trades, minus $15,000 financing and fees. Predicted daily vol from the risk model was $410,000. The number is unremarkable in size and sensible in composition — carry dominates, as it should on a book that mostly holds positions rather than churns them, and it sits well inside one standard deviation of the model's prediction. You note it, move on, and the review takes two minutes.

Contrast with a day the book makes the identical $340,000, but predicted vol was $95,000. Same headline, completely different read: this book made 3.5 standard deviations of "predicted" P&L, which means either a position materially larger than the risk system thinks, or a correlation the model is not capturing. That day's review does not take two minutes — it takes an hour, and it starts before you've decided whether to be pleased.

Reading it across time, not just on the day

A single day's read is only the first layer. The same $340,000 gain means something different depending on what the previous five days looked like — the continuation of a steady trend, or a sharp reversal after a run of losses that had started to worry the desk. Keeping a simple running log of the daily breakdown — carry, new trade, financing, versus predicted vol — turns each day's two-minute check into something cumulative: patterns that are invisible in any single day's number, like carry P&L quietly trending down over three weeks even while daily totals stay positive, become visible once you can see the log rather than just today's line.

What a healthy routine looks like

Most experienced traders develop a version of this review that takes only a few minutes on an ordinary day precisely because they've internalized what "ordinary" looks like for their own book — they know roughly what carry P&L should be given the positions held, they know the usual size of the financing drag, and anything outside that range jumps out immediately rather than needing to be calculated fresh each morning. That fluency is built by doing the full breakdown deliberately, every day, including the boring ones — skipping the review on quiet days is exactly how a trader loses the baseline needed to recognize when a day stops being quiet.