The Pre-Open Checklist
The hour before the bell is where most avoidable losses are prevented. A written pre-open checklist turns "did I remember?" into "did I tick it?", and its job is to catch the problems that are already sitting in your book before you place a single order.
Prerequisites: The Shape Of A Trading Day
The losses that hurt most are rarely the ones you chose. They are the ones already sitting in the book when you sat down: a short whose borrow got recalled, a position that exists at the broker but not in your system, a name that goes ex-dividend today, a stop you left resting on a stock that split overnight. None of these are trading mistakes. They are things nobody looked at.
A checklist exists because human attention is unreliable under time pressure and completely unreliable on a busy morning. Airline crews, surgeons and traders all converged on the same solution for the same reason: the failure mode is not ignorance, it is omission. You know that shorts get recalled. You just didn't check today.
The point of writing it down is that a checklist runs the same on a quiet Tuesday as on the morning of a payrolls print. That is exactly when you need it, because the busy mornings are the ones where you would otherwise skip steps.
The four questions
Everything on a good pre-open list answers one of four questions, in this order.
Does my book agree with reality? Positions, cash and financing, in your system versus the prime broker's. This comes first because every later number is computed from it. A wrong position makes your risk wrong, your PnL wrong and your trade list wrong.
What happened while I was away? Overnight moves, earnings that printed after the bell, corporate actions effective today, and the economic calendar for the session ahead.
Can I actually do what I plan to do? Borrow on the shorts, margin headroom for the gross you want, limit utilisation, and whether anything you hold is halted or restricted.
What am I doing today, and what could go wrong? The target book, the trade list to reach it, an estimate of what that will cost, and the one or two scenarios that would make today painful.
The list itself
| Check | What you are really asking | A failed check looks like |
|---|---|---|
| Position reconciliation | Do my positions match the broker's, name by name? | Broker shows 12,000 shares, you show 10,000 |
| Cash and margin | Did yesterday settle, and do I have headroom? | Overnight margin call you did not know about |
| Corporate actions | Splits, dividends, spin-offs, mergers effective today | A resting order priced for the pre-split share |
| Borrow status | Are my shorts still borrowable, and at what rate? | A recall notice sitting in an unread email |
| Overnight moves | Which of my names gapped, and by how much? | Book already down two days of expected PnL at the open |
| Earnings and news | Did anything I hold print after the bell? | You are carrying an event you meant to be flat into |
| Macro calendar | What is scheduled, and at what time? | Sizing up at 08:29 before a 08:30 number |
| Model output | Did the model run, on clean data, to completion? | A stale signal file from two days ago |
| Risk versus limits | Where does the book sit against gross, net and VaR limits? | Starting the day at 96% of a limit |
| Today's trade list | What am I trading, in what size, by what method? | No plan, so everything becomes a market order at 15:58 |
| Leftover orders | Is anything still working or resting from yesterday? | A GTC order you forgot, filling into a gap |
A concrete morning
You arrive at 07:15 on a Thursday. Reconciliation is clean. Overnight moves are unremarkable. Then the borrow check fails: you are short 80,000 shares of a mid-cap, and the securities-lending desk has flagged 50,000 of them as recalled, with a buy-in if you are not flat by the close.
Because you found it at 07:20 rather than at 15:30, you have options. You can call two other lenders and try to re-borrow. You can buy back 50,000 shares in the morning session, when the stock trades $18m a day and 50,000 shares is a fifth of an hour's volume. Or you can replace the short with a listed put and keep the exposure while losing the borrow problem.
If instead you found it at 15:30, you have one option: buy 50,000 shares into the close, in a hurry, against a market that can see you coming. On a stock like that the difference between the calm version and the panicked version is easily 40 to 60 basis points — around $15,000 to $20,000 on a $3m short. The entire value of the checklist that morning was eight hours of warning.
A checklist is not about knowledge, it is about omission under time pressure. Its value is almost entirely in the number of hours of warning it buys you: the same problem found at 07:20 and at 15:30 costs completely different amounts of money.
Write the list so every item has a binary answer and a named owner. "Review risk" is not a check. "Gross exposure below $60m — yes or no" is. Anything that cannot fail cleanly will quietly stop being read.
The dangerous failure is a checklist that gets ticked without being run. Once it becomes ritual it is worse than nothing, because it manufactures confidence in checks nobody performed. Keep it short enough that running it honestly takes less time than faking it.
Related concepts
Practice in interviews
Further reading
- Gawande, The Checklist Manifesto
- Harris, Trading and Exchanges (ch. 5)