Realised Versus Unrealised P&L

A position is up $400,000 on the screen. Is that $400,000 real? It is real in the sense that it reflects an honest current price — but it is not real in the sense that matters for planning: nothing has actually converted to cash yet, and the number can go back to zero, or below, before you ever get to touch it. That gap between "the mark says I made money" and "I actually have the money" is the entire distinction between unrealised and realised P&L, and collapsing the two together is one of the most common ways traders make bad decisions with good positions.

Two different kinds of certainty

Realised P&L is locked in — you sold the shares, closed the derivative, the cash settled. It cannot be taken away by tomorrow's price action, because tomorrow's price action no longer applies to a position you do not hold. It is, by construction, the only P&L that is actually true rather than estimated.

Unrealised P&L is a mark: today's price applied to a position you are still holding, reported as if you had closed it, when you have not. It is the correct and necessary way to know how a book is doing day to day — you cannot wait for every position to close before finding out whether the book is healthy — but it carries every uncertainty the still-open position carries. It can also depend on how the position is marked (last trade, mid, model), which introduces judgment even before the market moves again.

Worked example

You are managing three positions at Friday's close:

Position	Status	P&L
A (closed Wednesday)	Sold, cash settled	+$180,000 realised
B (still open)	Marked at Friday's close	+$310,000 unrealised
C (still open)	Marked at Friday's close	-$95,000 unrealised

Total book P&L for the week reads +$395,000. That number is correct and is what goes on the daily P&L report. But only $180,000 of it is money the desk actually has; the other $215,000 net is an opinion, current as of Friday's close, about two positions that are still exposed to Monday's market. If B gaps down 15% and C recovers fully on Monday, the realised number stays at exactly +$180,000 — nothing changes it — while the unrealised number could easily flip to a net loss on the two open positions.

This matters concretely for a decision like a year-end bonus pool: paying out against unrealised gains rewards a mark that has not yet survived contact with an actual sale, and a position that reverses after the payout has already happened cannot claw the compensation back.

Why the distinction changes behavior, not just accounting

The practical failure mode is anchoring to a high unrealised number as if it were already banked — refusing to trim a position because "I don't want to give back the gain," when the gain was never given in the first place, only marked. Scaling out and taking partial profits exists precisely to convert some unrealised P&L into realised P&L on purpose, at a size tied to the thesis, rather than either holding everything on hope or selling everything out of fear that the mark will move.