SPY falls 1.20% as 37 failed breakouts outpace 32 winners across 118 setups

Risk got sold first, and nothing on the tape argued back.

The market spent Friday unwinding growth, small caps, and anything that needed clean sentiment to keep working. QQQ took the bigger hit than SPY, and IWM was worst of all, which tells you this was not a selective cleanup. It was broad risk reduction. The headline flow did not offer a real macro rescue either. Outside of a scattered series of halt notices and a soft miss from Shinhan Financial, there was no positive catalyst to absorb profit-taking, so sellers kept control into the close.

The more useful read came from rotation. Semis are still one of the few groups that remain strong week over week, but they still got hit on the day, which is what late-week de-risking looks like when leadership starts getting trimmed instead of bought. Housing and materials have now clearly rotated out versus SPY, while retail weakness kept getting worse. Tradeable implication: this was not a session to force upside continuation in beta. The better opportunities were in selective bearish breaks and defensive relative strength, especially where volume confirmed the move. For context and levels, review yesterday's debrief.

Indexes broke lower, breadth stayed ugly

The scanner found 118 total setups, split almost evenly between 61 bulls and 57 bears. That balance looks neutral on paper, but the resolution data was not. We got 32 breakouts against 37 failures and 3 expirations, which is a weak conversion rate for a market that needed cleaner follow-through. Only 7 setups graded A or better. see live setups in the scanner.

DELL BULL is still alive, but barely comfortable

The system picked AMD. The grade won, not the story.

AMD scanner chart – May 15, 2026 | UnxEdge

Where the volume was today

That is the kind of read traders routinely get wrong. High relative volume is not automatically bullish. In this case, it likely reflected active repositioning inside industrials rather than clean accumulation. When a setup carries almost 4x normal volume and still cannot convert, that is not strength. That is supply meeting demand and winning. In a better tape, this type of volume would have sponsored follow-through. On Friday it just marked a battleground that bulls lost.

Wex traded. Xcel stayed out.

All bot activity is published in real time, and today's tape did not offer much to celebrate. watch Wex and Xcel trade live in the Edge Lab.

Xcel closed no trades today. That restraint was probably the better decision given how sloppy the follow-through was outside a handful of isolated names.

The bots left money on the table, but not all misses are mistakes

The headline number was 14 missed breakouts and +10.1R in missed opportunity. That looks bad because it is bad. The biggest miss was XLB on the short side, where elevated volume and a clean downside move produced a full TP3 runner. CVNA was another real miss, not a trivial one, and should have been more actionable given the weak market backdrop and the stock's tendency to trend once it breaks.

That said, some of the smaller misses were acceptable filter rejects. A low-RVOL move like T does not need to be chased just because it happened to work. The problem is not that the filter was selective. The problem is that it may still be underweighting clean bearish continuation when the index tape is deteriorating. watch Wex and Xcel trade live in the Edge Lab.

More failures than breakouts, which fits the tape

Notable conversions included BMY BULL, C BEAR, CVNA BEAR, CVX BULL, DDOG BULL, and DECK BEAR. The list itself tells the story: mixed pockets worked, but there was no clean broad upside trend carrying lower-quality names with it.

The levels that matter next

These are the unresolved A and A- names still coiling. This is the section to actually trade from, not just read. see live setups in the scanner.

<