What an options flow scanner should catch: we tested 5,785 breakouts
Most options flow scanners do one of two things: filter the tape for big or unusual prints, or flag stocks breaking a level. We tested the second kind on our own data to see which filters actually change the odds. The short answer: the breakout on its own is a coin flip, and the filters that help are the ones that confirm real money is on the same side.
The test
Every stock we cover, every session from August 24 to September 28, 2026: 296 symbols and 25 sessions. The rule is the classic opening-range breakout. Record the high and low from 9:30 to 9:45 ET, wait until 10:00, and take the first 5-minute close more than 0.15% beyond the range. The stop is the other side of the range, and the target is one range-width past the entry.
That produced 5,785 breakouts. A result counts as a hit if the target traded before the stop, with a bar touching both counted as a miss.
The breakout alone: a coin flip
48.6% of the 5,785 breakouts reached the target before the stop. After the stop and target, the average result was slightly below zero. A scanner that only flags breakouts is showing you a coin toss with a good-looking chart.
What changed the odds
Every filter was checked three ways before we believed it: it had to hold in both halves of the dates, in two separate groups of symbols, and beat 95% of 1,000 random filters that passed the same number of trades.
Options flow on the breakout's side passed. When at least 60% of the name's options premium, counted only up to 15 minutes before the break, was on the breakout's side, 52% reached the target against 45% for the rest. Trading with the morning gap also passed: 53% when the breakout went the same way as the gap, 45% when it went against it. Together, flow on the breakout's side plus the gap, 57% of 361 breakouts reached the target.
What did not matter
Time of day did not hold up: whether a breakout came before 11:30 or after 1:00 PM made no difference that survived the checks. SPY moving the same direction did not hold up either. VWAP could not be tested as a filter at all, because an opening-range breakout already sits on the right side of VWAP almost every time.
A tight opening range looked helpful on a small sample and reversed on the full one. That is exactly why every filter gets the three checks.
| Filter | Breakouts | Reached target first | Without the filter |
|---|---|---|---|
| None (every breakout) | 5,785 | 48.6% | — |
| Options flow ≥60% on the breakout's side | 723 | 52% | 45% (1,152) |
| Breakout in the gap's direction | 2,790 | 53% | 45% (2,995) |
| Flow on side AND with the gap | 361 | 57% | 46% (1,514) |
| Volume at least 1.5× normal | 825 | 52% | 48% (4,959) |
Common questions
Do opening-range breakouts work?
Not on their own in our test: 48.6% of 5,785 breakouts reached a one-range target before the stop. They improved when options flow and the morning gap agreed with the direction.
What should an options flow scanner filter for?
Premium on the same side as the move, measured only from what was knowable at the time, plus agreement with the gap. Size alone and time of day did not change the odds in our test.
Is this FlowMonkey's alert record?
No. This is a test of a generic breakout rule on stock prices. FlowMonkey's own alerts are listed individually, winners and losers, on the public receipts page.
The breakout is a coin flip. What moved the odds was real money on the same side (options flow in the breakout's direction) and trading with the gap, not the time of day or a tight range.
Keep reading
See it on a live board
FlowMonkey prices every level by whether your ticker can actually reach it today — measured on that symbol's own range, re-sorted as the session burns down.
See pricing