Explaining A Bad Day
A bad P&L day needs a real explanation traced to specific positions and causes — not a story, and not silence — because the explanation is what tells you whether to change anything tomorrow.
Prerequisites: Where Did Today's P&L Come From?, Telling P&L Noise From P&L Signal
The book lost $1.1m today. Your risk manager is going to ask why before the market opens tomorrow, and "the market was down" is not an answer — the market being down is exactly what the hedges and the risk model were supposed to already account for. A real explanation says which positions lost money, whether that loss matches what those positions were supposed to do in a down market, and whether anything about today changes what you believe going forward.
The three explanations, in order of how comfortable they are
"This is what the position is for." A short-vol position loses money when volatility spikes; that is not a surprise, it's the position doing its job on the day its job is unpleasant. If the loss size roughly matches what the position's risk profile predicted, this is the easiest and most honest explanation, even though it's still a loss.
"The hedge didn't track." You were long a stock and short an index future as a hedge, the stock fell in line with the sector, but the specific stock fell more than the sector because of company-specific news the hedge was never designed to catch. This is basis risk showing up — not a broken thesis, but a real gap between what you thought you were protected against and what actually happened.
"We were wrong." The thesis itself failed. The catalyst you were positioned for didn't happen, or happened and the market didn't care, or a competing effect you underweighted turned out to dominate. This is the explanation that should change tomorrow's positioning, not just today's mood.
Most bad days are a mix, and the discomfort of admitting which fraction is "we were wrong" versus "this is what the position is for" is exactly why traders reach for vaguer explanations first.
Worked example
A pairs desk is down $650,000. First pass: long Retailer A, short Retailer B, on a thesis that A was taking share from B. Today both stocks fell — A by 2 percent, B by 1 percent, both on a broad consumer-discretionary sell-off tied to a rate scare. The pair itself only cost $40,000 (A underperforming B by 1 percent on a $4m gross pair). That is not where the $650,000 came from.
Digging further: a separate, larger single-stock short in the same sector — not part of any pair, an outright directional bet that the sector was overvalued — is down $610,000, because the position was three times the size of the pairs book and had no hedge against a broad sector rally, which is exactly what happened for reasons unrelated to your specific short's fundamentals. That is squarely "we were wrong" — wrong about sizing an outright short against a systematic risk you had no offsetting position for, not wrong about the company. Tomorrow's action is concrete: either hedge the sector exposure on that short or cut the size, because the position as constructed is exposed to a risk that has nothing to do with the reason you put it on.
A real explanation for a bad day traces the loss to specific positions and sorts each into "the position did what it's for," "the hedge had a gap," or "the thesis was wrong" — only the last one should change what you do tomorrow.
"The market was volatile" is not an explanation, it's a description of the environment every position was already sitting in. If volatility alone explained the loss, the risk model — which already prices in volatility — should have predicted a loss of roughly that size. When it didn't, volatility isn't the actual explanation; something more specific is.
Writing it down changes the answer
An explanation that only ever gets said out loud in a meeting tends to stay comfortably vague — "a mix of factors," "some sector weakness" — because nobody has to commit to a specific, checkable claim. Writing the explanation down, in a form specific enough that someone could check it against tomorrow's data, is what forces the vague version into something real: not "sector weakness" but "the sector ETF was down 1.8 percent and our beta-adjusted exposure explains $210,000 of the $650,000, leaving $440,000 attributable to the outright short specifically." The written version is uncomfortable in a useful way — it's much harder to hide behind a sentence you know will be reread next week than one that only had to survive a single conversation.
The same discipline on good days
It's tempting to only apply this rigor to losses, since gains don't prompt the same uncomfortable questions from a risk manager. But a gain deserves the identical breakdown, because "we were right" and "we got lucky" produce the same green number and very different lessons — a gain that traces back to a position sized far outside normal risk limits, that happened to work, teaches exactly the wrong lesson if it goes unexamined simply because nobody was upset about it.
Related concepts
Practice in interviews
Further reading
- Green, Managing a Trading Desk