Reading Unusual Activity — Smart-Money Options Flow
Every option trade leaves a fingerprint: trade size, price relative to bid/ask, prior open interest, condition codes. The Unusual Activity scanner walks the tape across a 25-ticker watchlist and surfaces only the prints that pass three independent bars — large premium, clear aggressor side, and a size that dwarfs the contract's prior-day OI. The result is a short list of trades where size and conviction both line up.
The three-bar filter
The Unusual Activity scanner is conservative on purpose. The options tape is loud — millions of contracts trade every day, most carrying no directional signal. The three-bar filter (premium, OI multiplier, aggressor side) cuts the noise by roughly 99.9% before anything reaches your screen. What survives is a short list of trades where SIZE and CONVICTION both line up.
Reading a print card
Top row. Ticker · strike + call/put + DTE · BOT or SLD chip (aggressor side) · classification chip (bullish call buy / bearish put buy / call sell / put sell) · Sweep badge if applicable · total premium.
Detail row.Size · Price · OI multiplier · Strike vs spot · Tape time (ET). The OI multiplier in particular is the single most useful number — anything 5×+ is meaningfully bigger than the contract's prior position base and almost certainly opening flow.
How signal stacks
- Size + sweep + OTM = strongest signal. A $200k+ sweep on a +5-10% OTM strike, especially with multiple prints in a 30-minute window, is the institutional pattern most retail flow services charge to alert on.
- Repeated direction = real signal. Same ticker, same direction, multiple prints over 2-3 days. The first print could be a hedge; the third is a thesis.
- Conflicting flow = no signal. If today shows $300k bullish call buys AND $250k bearish put buys on the same ticker, two different desks have opposite theses — neutral.
What this is NOT
Following options flow is NOT a get-rich strategy. Well-filtered sweeps have a 55-60% directional hit rate over the next 5-20 trading days. That's a real edge, but it requires position sizing (winners pay for losers) and patience (the thesis often takes 1-3 weeks to play out). The cards on this page are the SHORT LIST — what to investigate, not what to copy-trade.
Frequently asked questions
What's the filter for an 'unusual' print?
Three conditions must all hold: (1) Premium ≥ $50,000 — total notional (size × price × 100). (2) OI multiplier ≥ 3× — trade size is at least 3x the prior day's open interest on that exact strike/expiry. (3) Clear aggressor side — fill price within 1¢ of the bid (aggressive seller) or ask (aggressive buyer); midmarket fills are dropped because they don't carry signal. Sweeps (Polygon condition code 41) are flagged separately but don't gate the filter.
What do the four classification labels mean?
Bullish call buy = aggressive buyer of calls (lifting offers). Bearish put buy = aggressive buyer of puts. Call sell = aggressive seller of calls (short call — either covered or directional bearish). Put sell = aggressive seller of puts (short put — often cash-secured or directional bullish). 'Ambiguous' (midmarket fills) is dropped from the scan entirely.
What's a 'Sweep' and why does it matter?
A sweep is one order broken across multiple exchanges to grab all available liquidity simultaneously — Polygon flags this as condition code 41 (Intermarket Sweep Order). It's conventionally read as urgency: the trader didn't want to wait, suggesting either short-dated thesis or scale. Sweeps are weighted slightly heavier than equivalent-sized block trades in most institutional scanners; the violet 'Sweep' chip on the card surfaces them so you can spot the difference.
Why a 3x OI multiplier?
It's the de-facto institutional standard. Below 3x, you're often watching market-makers shuffle inventory or vol traders adjusting hedges — not opening directional bets. At 3x+, the trade is meaningfully BIGGER than the existing open interest, which strongly implies a new position rather than a roll or close. Some scanners use 2x or 5x; we settled on 3x as a balance between signal and quantity. You can adjust in lib/uoa.ts if you want it tighter.
Daily summary vs Latest intraday — what's the difference?
The daily summary (top of page when populated) is published at 4:15 PM ET after the EOD chain settles. It's the official 'flow of the day' record — top 25 prints across the watchlist with classification breakdown. The Latest intraday section (amber-pulse banner, appears during RTH) refreshes every 5 minutes with whatever has cleared the filter in the last hour. Both use the same uoa_prints table — the daily scan is a SUMMARY of the day's prints, the Latest section is a live RECENT-prints feed.
How do I act on this in practice?
Treat it as a SIGNAL, not a TRADE. A single $200k bullish call sweep on AAPL doesn't mean buy AAPL calls — it means there's a 50-55% historical edge on the underlying moving in that direction over the next 5-20 trading days. Best practice: cross-reference with Options Edge (is the IV cheap?), check the catalyst calendar (earnings within 21 days?), and size as a directional bet, not a 'follow the whales' all-in. Hit rates on well-filtered sweeps are 55-60% — meaningful, but you need position sizing to capture it.
Why aren't 0DTE prints in the list?
0DTE flow is dominated by dealer hedging, not directional bets — so it doesn't carry the same signal. The scanner doesn't explicitly exclude 0DTE, but the OI multiplier filter naturally trims most of it: 0DTE OI is huge by close so it's hard for a $50k print to be 3x the existing OI. Real institutional bets show up in 1-4 week and longer-dated chains, which is what surfaces here.
What does 'Strike vs spot' mean?
The strike's distance from spot, as a percent. Positive = OTM call or ITM put; negative = ITM call or OTM put. For directional reads, very-OTM prints (+10% or more) are higher-conviction directional bets — the buyer wants leverage and is willing to pay the premium. Slightly-OTM prints (+2-5%) often pair with delta hedging or trend continuation.
How is 'aggressor side' determined?
We compare the trade's reported fill price to the contract's NBBO bid and ask at trade time. Price ≥ ask − 1¢ → aggressive buyer (lifted the offer). Price ≤ bid + 1¢ → aggressive seller (hit the bid). Anything in between is midmarket — dropped from the scan. This is the textbook microstructure read of aggressor side, used by every institutional flow scanner.
Can a small trader use this data?
Yes — but with realistic expectations. You can't lift a $200k contract block at the price the whale paid; by the time you read the page, the premium may have moved 10-20%. Where retail can win: (1) recognizing PATTERNS — same ticker getting hit with bullish call sweeps three days running is a real signal; (2) directional spreads — buy a $5-wide call spread on the underlying instead of trying to match the whale's exact contract; (3) Multi-week holds — the edge plays out over 5-20 days, not minutes.