Quant Memo
Foundational

Blotter Hygiene: Keeping The Order Record Clean

The order blotter is the running record of every order sent, filled, cancelled or rejected during the day, and keeping it clean and current is what makes it possible to reconstruct exactly what happened when something goes wrong.

Every order a desk sends — filled, partially filled, cancelled, replaced, rejected — leaves a record on the blotter: a running log of order activity through the day. In a quiet moment it looks like bookkeeping. When a fill looks wrong, a client asks why an order took three minutes to execute, or a regulator asks for a reconstruction of trading around a specific event, the blotter is the only record that can answer the question precisely. Blotter hygiene is the discipline of keeping that record accurate and complete as it's being generated, not cleaning it up after the fact.

What "clean" means

  • Every order has a clear status — working, filled, partially filled, cancelled, rejected — and that status is current, not stale from an earlier state.
  • Cancel-and-replace chains are traceable. If an order was modified five times before filling, each modification should be identifiable, not collapsed into a single ambiguous entry.
  • Rejected and unfilled orders stay visible, not silently dropped — a rejected order that isn't investigated can represent risk the trader thinks is on and isn't.
  • Manual annotations are timestamped and attributed. If a trader adds a note explaining an unusual fill, it needs to be clear who added it and when.

Why it decays without attention

Under normal trading volume, blotter entries mostly take care of themselves through the order management system. The problems show up around the edges: a stale working order that was meant to be cancelled but wasn't, a fill that got booked to the wrong strategy tag, a manual trade entered outside the normal flow that never got the same level of detail as an electronic one. None of these are dramatic individually, but they accumulate, and a blotter with enough small inconsistencies stops being trustworthy as a record.

Common blotter problemConsequence if unresolved
Stale working ordersRisk the trader thinks is off is actually still live
Orphaned partial fillsP&L and position don't tie out
Missing rejection reasonsCan't diagnose recurring rejects
Untagged manual tradesBreaks strategy-level attribution

What this means in practice

A quick blotter review at the end of the day — confirming every order is in a terminal, understood state — catches these problems while they're cheap to fix, rather than surfacing weeks later during a reconciliation break or a compliance inquiry. Desks that treat the blotter as a living record needing daily attention avoid the far more painful process of reconstructing what happened after the fact from fragments.

The order blotter is the authoritative, timestamped record of everything a desk tried to do during the day, not just what filled. Keeping it clean in real time — resolving stale orders, tagging manual entries, tracking rejects — is what makes it possible to answer "what actually happened" precisely, whenever that question comes up.

Related concepts

Further reading

  • Harris, Trading and Exchanges (ch. 4)
ShareTwitterLinkedIn