If you've ever opened an options-flow tool during market hours, you know what I'm talking about. Thousands of prints scrolling past every minute. A wall of data so dense you can't read it, can't act on it, can't tell what matters from what doesn't.
99% of that data is noise.
Market-maker spreads filling against each other. Dealers hedging long stock positions with put overlays. Retail FOMO calls bought thirty minutes before earnings. Adjustments. Roll-overs. Cabinet trades. Stale theta. None of it tells you anything about where institutional capital is actually moving.
The few prints that do matter — real conviction sweeps, accumulation across multiple strikes, $500K+ blocks lifting the offer with no offsetting flow — get buried in the firehose. By the time you spot one with your own eyes, the position is already baked in and the move is half over.
That's the problem I sat down to solve when I started building Profit Builders. Not "publish a flow tool." Not "compete on UI." Build a scanner that throws away the 99% so you can actually see the 1%.
What "actionable" means
When we say a signal is actionable, we mean something specific:
- It comes from informed capital, not market-maker plumbing or retail noise
- It's directional, not a hedge
- It's sized big enough to move price if the conviction is real
- It clears a documented set of filters we publish, not an "AI" black box
- It hits your phone fast enough that you can act on it before the move is over
Every one of those criteria is testable. Every one of them is enforced by the scanner before a signal ever reaches you. If a print fails a single one of them, you don't see it.
How the filtering works
The scanner watches every options print across a 220-symbol watchlist in real time. On a typical session that's tens of thousands of prints. Out of all of those, maybe 40 to 60 ever clear the gates.
Here's what each print runs through:
- Sweep detection — was the order broken across multiple exchanges, indicating urgency? A spray across venues at the offer means something different than a single block to one venue.
- Vol/OI ratio — is the volume meaningful relative to existing open interest? A 50× vol/OI print is positioning. A 0.3× print is rebalancing.
- Market-maker filter — was the trade between two MM accounts hedging each other? We screen those out at the database layer before grading.
- Direction inference — did the trade hit at the bid (selling pressure), the offer (buying pressure), or somewhere in the middle (uninformative)? Mid-market fills tell you nothing.
- DTE and premium gates — is the expiry far enough out and the premium large enough to indicate real conviction, not a lottery ticket?
- Position-action classification — opening or closing flow? A 5,000-contract opening sweep is a thesis. A 5,000-contract closing sweep is profit-taking.
- Deep-ITM and far-OTM screens — extreme strikes get scrutinized differently from at-the-money flow because the math behaves differently.
Each filter is a rule, not a model. We can show you the gate. We can show you why a given print passed or failed. There's no "the AI thought this was a good idea." Either it cleared the documented criteria or it didn't.
Prints that clear all of the gates get graded A. Prints that clear most but not all get graded B. The grade is a byproduct of the filtering, not the headline. The headline is: you see the 40 prints that mattered, not the 40,000 that didn't.
Why real-time matters
A signal that's right but slow is worth nothing. By the time you read about a sweep on Twitter, the strike is already 20% higher. By the time most flow tools batch and serve their output, the alpha has moved.
Median latency from print on the tape to alert in your Discord or Telegram: 1.4 seconds. The institutional desk and your phone learn about the same trade at roughly the same time. What you do with that 1.4-second window is up to you. The point is the window exists.
What you do with the signal
Profit Builders is not a trading bot. It doesn't execute trades. It doesn't manage positions. It doesn't predict where the underlying is going.
It surfaces what an informed actor just did — sized correctly, structured directionally, filtered for actionability — and gets out of your way. You read the signal, you read the rest of the chain, you check IV and gamma and whatever else feeds your process, and you trade your own book.
The tool's job is to put the right 40 prints in front of you instead of the wrong 40,000. That's it. That's the entire pitch.
The receipts
Because we believe in the filtering, we don't hide what comes out of it. Every signal that clears the gates is logged the moment it fires and published with its outcome at profitbuilders.io/methodology — OPRA-grade resolved trades, public, searchable.
You don't have to take our word for any of this. Spend twenty minutes on /methodology. Sort by ticker. Sort by P&L. Find the cold streaks. Find the hot streaks. Do the math yourself.
Grade A is the institutional-data-quality tier of our methodology. That's not a marketing number. That's whatever the data returns when you measure honestly across the live OPRA tape. With asymmetric directional options trades and disciplined sizing, a sub-50% hit rate produces strong positive expectancy when winners run +30% and losers are capped by time decay — but you don't need to take our framing of that, either. The math is on the page.
If the filtering ever stops working, you'll see it in the data before we do. That's intentional.
What we promise
The product is the filtering. As long as Profit Builders exists, it will:
- Surface only the prints that clear documented gates
- Publish those gates and the rules behind them
- Alert you fast enough to act on the signal
- Stay out of your way once the signal is delivered
- Show you every outcome, win or lose, so you can verify the work
We don't promise win rates. We don't promise returns. We don't promise the next big move. We promise to keep throwing away the noise — and to keep proving, in public, that the signal we surface is the kind of signal worth seeing.
The receipts are at profitbuilders.io/methodology. Spend twenty minutes there. See if the filtering holds.
— Ralph
Related Reading
See the prints we can't publish here.
Every strike, every expiration, every accumulation pattern — tracked in real time.
Start Free 7-Day Trial