Unusual
Activity
Somebody just spent $2.6 million on a bet that expires in 32 days. This is the page where you find out who, on what, and how hard.
Every options print that clears the premium floor gets scored on four axes. Anything scoring 80 or better lands on the default view of this feed; the filter is yours to move. That is the whole product in one sentence. The rest of this page teaches you to actually read it.
THE 60-SECOND VERSION
If you read nothing else, read this.
Institutions can't hide size. When a fund buys 4,000 contracts, that trade prints on the public tape within seconds. It doesn't come with a name attached, but it comes with a price, a strike, an expiry and an urgency, and those four things tell you almost everything.
Size is the signal
A $40k print is somebody's opinion. A $2M print is somebody's job. Premium is the column that separates the two, and it's the first thing your eye should go to.
New money beats big money
Volume above open interest means the position didn't exist this morning. Somebody is opening, not unwinding. That's the difference between a signal and an exit.
Sort by score, not premium
A $2M print at score 86 is usually a hedge. A $400k print at score 110 is somebody with a strong opinion and a deadline. The score already did the maths for you.
THE ALPHA
What this actually buys you.
Not "signals." Three specific, unfair advantages that a retail trader without a flow feed simply does not have.
Positioning shows up before the headline does. When a name gets swept at 10:14am and the news breaks at 3:40pm, you were looking at it for five hours.
The hardest part of trading isn't the entry, it's deciding what to look at. This is a pre-filtered shortlist of names where real capital is already committed.
A stock tip gives you a ticker. A flow print gives you the exact contract, the price target implied by the strike, and the date by which the buyer needs to be right.
ANNOTATED · THE REAL SCREEN
Eleven columns. Four of them matter.
This is the feed as it appears in the product. The numbered pins mark the four columns that decide whether a row is worth your time. Everything else is context you check second.
| Time | Ticker | Flow Type 2 | Type | Sentiment | Strike | Expiry | Premium 1 | Vol / OI 3 | Score 4 |
|---|---|---|---|---|---|---|---|---|---|
| 10:14 AM | NVDASTOCK | GOLDEN_SWEEP | CALL | BULLISH | $195 | 10/16/26 32d | $2.61M | 5.2× NEW6,190 vol / 1,190 oi | 112 |
| 10:09 AM | XLEETF | BLOCK | PUT | NEUTRAL | $88 | 01/15/27 123d | $3.88M | 0.4× OLD9,200 vol / 24,100 oi | 87 |
| 09:58 AM | SMCISTOCK | SWEEP⚡ 3× hit | CALL | BULLISH | $52 | 09/18/26 4d | $412K | 11.2× NEW8,900 vol / 795 oi | 104 |
| 09:41 AM | SOFISTOCK | SPLIT | CALL | NEUTRALCalls Closing | $14 | 11/20/26 67d | $96K | 0.9× OLD3,100 vol / 3,440 oi | 86 |
Illustrative rows built from the real feed structure and scoring. Not live quotes, not recommendations.
Premium: how much they committed
Price × volume × 100. The headline conviction number. $2.61M means somebody wrote a real cheque; the money is already at risk before you clicked.
LOOK FOR → $500k+ on a single contract
Flow Type: how badly they wanted it
GOLDEN_SWEEP is the platform's highest-conviction bar: $2.5M+ premium, 5x or more Vol/OI, and within 15% of the money, on any trade type. BLOCK is negotiated size, planned not panicked. SPLIT is deliberately slow. Urgency is information.
LOOK FOR → GOLDEN_SWEEP, or SWEEP with a repeat-hit tag
Vol / OI: whether it's new money
6,190 contracts traded against 1,190 that existed = 5.2×, tagged NEW: the position was built today. Below 1.0× the tag reads OLD, which can be closing or rolling. A THIN OI tag means a high ratio on a tiny base: wide spreads, and getting out is the risk. This ratio is also the score's biggest axis (40 of 100 points).
LOOK FOR → 3.0× or higher, tagged NEW
Score: all of the above, in one number
Vol/OI against the ticker's own 30-day baseline, premium size, spread tightness and DTE, plus pattern bonuses. This is the column to sort by. 112 is a different animal from 87.
LOOK FOR → 110+ (institutional tier)
Where the score comes from
Four weighted components. Max 100 before bonuses; 125 is the ceiling, and the highest score on the feed in the last month was 115.
The three tiers, and what to do
Same feed, three very different levels of "pay attention."
WORKED EXAMPLE · ILLUSTRATIVE
One row, start to finish.
Here is the whole loop on the NVDA row above: what you saw, what you checked, what you bought, and what you did when it worked. Numbers are illustrative. Real tickers, made-up fills, so you can follow the arithmetic.
The print lands
NVDA $195C 10/16/26, GOLDEN_SWEEP, $2.61M premium, Vol/OI 5.2×, score 112. Spot is $181, so the buyer needs +7.7% in 32 days. Institutional tier: it goes to the top of the sorted feed on its own.
Three checks before you touch it
Never trade a single row. The whole point of having a platform is that you can confirm in under ten minutes.
You don't copy the trade. You copy the thesis.
The institution bought 32 DTE at $195 because it can absorb the theta. You take the same direction with more room: NVDA $190C 11/20/26 (67 DTE), 4 contracts at $6.85 = $2,740 risked. Longer expiry, closer strike, one third of the decay pressure. Stop: thesis breaks if NVDA closes below $174.
Sell into the move, not after it
NVDA trades $193 on an analyst raise. Contract marks $13.40. You sell half at $13.40 to take the risk off, and trail the rest with a stop at breakeven. That's the discipline the print can't give you. Flow tells you where to look, never when to leave.
Illustrative. This is what the arithmetic looks like when the thesis works, roughly one time in three. The other two are why the position was 4 contracts and not 40.
NVDA drifts to $176 instead. Nothing dramatic happens. No crash, no bad news, the print just doesn't play out. Eleven days of theta later the contract marks $3.90 and you're down $1,180. The institution is still fine; their book has a hundred other positions. Yours had one. That asymmetry, not the entry, is what you're managing.
THE FILTER
Green flags, red flags.
Screenshot this one. Run every row you like through both columns before you size anything.
✓Take it seriously when…
✕Walk away when…
HOW PEOPLE LOSE MONEY HERE
Four mistakes, in order of cost.
Every one of these is somebody blowing up an account with good data. The feed is not the problem. The reflex is.
Buying the exact contract they bought
A fund holding 6,000 contracts is indifferent to a 40% drawdown on one leg. You are not. Their 0DTE lottery ticket is your total loss.
Treating every big put as bearish
Institutions buy puts to protect a long book they've held for two years. A $4M put on an index ETF is usually insurance, and shorting alongside it is how you get run over.
Chasing the feed all day
Two hundred rows print a day. Trading twenty of them is not twenty edges, it's twenty commissions and no thesis. Volume of trades is the enemy here.
No exit written down before entry
Flow tells you where to look. It never tells you when to leave; the institution's exit doesn't print with a label on it. Traders give back winners waiting for a signal that never comes.
PAIRS WITH
This tool is step one of four.
Unusual Activity finds the name. These three tell you whether to believe it. Nobody good trades a single feed.
Go and watch it for a week without trading.
Seriously. Sort by score, read the top three rows each morning, and note what happened by Friday. The pattern recognition is the product. The feed is just the delivery mechanism.
Every number, contract, fill and P&L on this page is illustrative and built from the real structure of the feed. Nothing here is a recommendation, a live quote, or a promise of a result. Options carry substantial risk of loss, including total loss of premium paid.