Options Flow and Order Internalization
Volume and open interest tell you what is held. Flow tells you what is being bought and sold right now. The two are different stories.
What flow means
Options flow is the stream of orders hitting the market in real time. Each trade is tagged by strike, expiry, type (call or put), size, and (most importantly) whether it executed at the bid or the ask.
A trade at the ask is an aggressive buy: someone wanted the option enough to pay the seller's price. A trade at the bid is an aggressive sell: someone wanted out enough to accept the buyer's price. The aggregate of these tags, across all strikes and expiries, paints a picture of who is doing what.
Volume vs flow vs open interest
Three related but distinct measures:
- Volume: total contracts traded today. A single trade contributes one count regardless of who initiated it.
- Flow: volume tagged by initiator. Bid-side volume vs ask-side volume.
- Open interest: contracts outstanding overnight. Today's volume that opens new positions adds to OI; volume that closes existing positions reduces it.
Flow shows you what is being added or removed today. Open interest shows you the cumulative state of positioning. They can disagree: a day of heavy ask-side flow that all closes existing short positions actually reduces open interest even as it shows aggressive buying.
What flow signals can tell you
Flow analysis tries to separate informed flow (institutions positioning ahead of moves) from uninformed flow (retail noise, random hedging). Several heuristics get applied:
- Block trades. Large size (1000+ contracts) executed in single prints. Often institutional. Worth noting individually.
- Sweeps. Single buyers crossing multiple exchanges to hit liquidity simultaneously. Suggests urgency, often informed.
- Repeated flow at a strike. The same direction of aggressive flow building over hours or days. Suggests an intentional position.
- Unusual options activity (UOA). Volume that vastly exceeds historical norms. Sometimes informational, often noise.
None of these are reliable signals on their own. The hit rate on flow-based prediction is uncomfortable for anyone who has actually backtested it. Flow is one input among many, and it is most useful as a corroborating data point rather than a primary signal.
Order internalization
Most retail options orders never reach the public exchanges. They are routed to wholesalers (Citadel Securities, Susquehanna, others) who pay the broker for the order flow (PFOF) and execute the order internally against their own book.
This is called order internalization. The wholesaler offers the customer a fill at or better than the National Best Bid and Offer (NBBO), and pockets the spread. For the retail customer, the execution is usually fine. For the public market, the consequence is that retail flow is mostly invisible to the visible volume tape.
This matters for flow analysis. The volume you see on public exchanges is dominated by institutional and market maker activity. Retail flow exists but is largely off-tape. When flow analysis platforms claim to track retail vs institutional flow, they are using inferences (size, time of day, contract characteristics) rather than direct identification.
Flow vs price action
When flow and price diverge, traders pay attention. Heavy call buying with the stock falling (bullish positioning into weakness) can signal an attempted bottom. Heavy put buying with the stock rising (defensive positioning into strength) can signal an attempted top.
Both patterns have weak statistical edges and high false-positive rates. They are most useful as confirmation rather than initiation. A trader who has a bullish setup from technical or systematic analysis can use call flow as a corroborating data point. A trader who buys solely because flow looks bullish will be wrong often.
Dark pools and SIP
Equity markets have dark pools (private exchanges where orders are not displayed publicly) that handle a meaningful fraction of total volume. Options markets have less dark trading because the OCC requires all options trades to be cleared centrally and reported. But there is still a lag between execution and public reporting, and large institutional trades can be partially obscured.
The Securities Information Processor (SIP) consolidates options trade data and publishes it with a small delay. Most retail flow tools work from the SIP. High-frequency proprietary flow tools work from direct exchange feeds and see trades milliseconds earlier.
0DTE flow
The rise of zero-day-to-expiry (0DTE) options on SPX has changed the flow landscape dramatically. 0DTE volume now exceeds longer-dated options by a wide margin most days. The flow patterns are different: shorter holding periods, more retail participation, more momentum-driven activity.
This has shifted some of the daily microstructure of SPX. The afternoon flow patterns covered in L.20 (charm flows, vanna flows, end-of-day hedging) interact differently with 0DTE volume. The space is still being studied. The patterns are not fully stable.
Where to actually use flow
Practical applications for systematic and discretionary traders:
- Confirmation. If your other signals say long, and call flow is heavy, that is a small additional confidence point.
- Risk awareness. If your position is short premium at a strike and you see heavy buying flow at that strike, your gamma risk just got worse.
- Macro positioning gauge. Aggregate index flow can signal regime shifts (large protective put buying ahead of a known event window).
- Single-name idea generation. Unusual flow can flag stocks worth investigating further. It is a starting point, not an answer.
What you carry forward
- Flow is volume tagged by initiator (bid-side vs ask-side).
- Volume, flow, and open interest answer different questions about positioning.
- Most retail options flow is internalized by wholesalers and invisible to the public tape.
- Flow analysis is corroborating information, not a primary trading signal.
- 0DTE volume has reshaped intraday flow dynamics in the index complex.