Options flow

What you are actually looking at

Every options trade prints publicly with the contract, size, price, and the prevailing bid and ask at execution. Flow scanners filter for large or unusual prints and present them as a feed. Three shapes recur:

• Sweeps. One order filled across multiple exchanges simultaneously. The trader prioritised speed over price, which is at least weak evidence of urgency.
• Blocks. A single large fill, often privately negotiated. Frequently institutional, and frequently a hedge.
• Splits. A large order broken into pieces to reduce market impact. Deliberately quiet.

Why most flow tells you nothing

This is the part flow-alert services tend to skip.

What you see	What it might actually be
$2M in calls bought	A hedge against an existing short stock position
Huge put volume	Portfolio insurance from a fund that stays long
Aggressive call selling	Covered calls written against a large share holding
Two prints, opposite directions	A single spread reported as separate legs
Massive size at the ask	A market maker's own hedging print

You cannot see who traded, whether they were opening or closing, or what else sits in the account. A print showing "bought at the ask" is inference, not fact — and on spreads the legs are routinely reported in ways that invert the apparent direction. Treating a feed of large prints as a feed of institutional convictions is the central error of the whole category.

What raises the signal

Four filters do most of the work:

1. Volume relative to open interest. Volume substantially exceeding existing open interest at that strike means new positioning rather than adjustment of something already there.
2. Aggression. Repeated fills at or above the ask indicate a buyer willing to pay up. Passive fills at the bid are far weaker evidence.
3. Repetition. The same strike and expiry accumulating over hours or days is worth more than one dramatic print. Single prints are noise more often than not.
4. Expiry that makes sense. A short-dated out-of-the-money strike bought aggressively is a directional bet with a deadline; two-year LEAPS are usually something structural.

Flow and dealer positioning are the same story

The reason the desk tracks flow is not to copy trades. It is that today's flow becomes tomorrow's dealer positioning. Every contract someone buys is a contract a dealer is short, and that short has to be hedged. Heavy call buying at a strike builds the gamma concentration that later behaves as a call wall.

Read that way, flow is a leading indicator for where the levels will be — not a set of tickers to follow. It is an input to the map, not a signal on its own.

Mistakes to avoid

• Copying a print blind. You do not know the entry, the thesis, the hedge on the other side, or the exit. You are copying one visible leg of something you cannot see.
• Chasing after the move. By the time a print is on your screen and you have decided, the premium has usually already repriced.
• Assuming size means smart. Large institutions hedge badly, get stopped out, and lose money like everyone else.
• Ignoring implied volatility. Buying the same strike after a print has bid up IV means paying a materially worse price for an identical thesis.

How this shows up in alerts

Flow feeds the pre-market note: where positioning is building, which strikes are accumulating, what the map looks like at the open. It rarely triggers a trade directly. Trades come from price arriving at a mapped level with structure behind it — the process is the same regardless of how interesting the flow looked.