SPY slips 0.64% as 11 of 60 setups broke out; AVGO was the pick
Market recap, bot performance, and scanner analysis for Thursday, July 16.
The UnxEdge breakout scanner only found 60 setups today, and that lighter menu mattered because the wedge pattern breaks that did trigger had almost no follow-through. Semis rolled over hard while SPY stayed relatively contained, and that split was the whole session in one line.
This was not a broad liquidation tape. It was a growth-heavy unwind, with QQQ down 1.76% against just a 0.64% drop in SPY, while IWM barely flinched. That lines up with the week-over-week rotation data: semis are now ROTATING OUT after modest leadership last week, while energy keeps absorbing relative-strength flows. When leadership narrows like this, wedge breakout quality usually degrades fast in tech and improves in defensive or commodity-linked pockets.
There was no major macro print to force positioning, so the market traded off micro catalysts and relative strength instead. Broker target hikes in financials and cybersecurity helped keep parts of the tape orderly, but they did nothing to save semiconductor risk appetite. With the macro calendar clear, traders defaulted to sector selection, and the implication was simple: if you were leaning long index beta through tech, you were on the wrong side of the split. For context, here is yesterday's debrief.
Market Snapshot
The index damage was real, but it was concentrated
No A-grade setups showed up. That matters. The scanner was basically telling you this was a B to A- tape with selective opportunity, not a session to spray breakout entries. Grade distribution came in at 27 A-, 21 B, and 12 B+. You can see live setups in the scanner when these levels tighten into trigger zones.
Yesterday's Pick
CVS never earned the benefit of the doubt
This is the kind of loss you take and move on from. The setup failed before it could build momentum, and in a tape where tech was bleeding and clean leadership was narrow, there was no reason to widen risk or invent patience that price action did not deserve.
Arxe Pick of the Day
AVGO won the model, but the live swing pick stayed with OKTA
Verdict: trade_it
Why it screened best: strongest balance of backtest quality and sample size, with a 71.43% win rate, 2.5 profit factor, and 14-trade sample that carries more weight than a tiny lookback edge.
Why that mattered today: even with QQQ weak, the system still preferred a higher-quality name with proven breakout behavior over lower-confidence alternatives.
AVGO scanner chart – July 16, 2026 | UnxEdge
The editorial swing pick published for today was OKTA BULL (HIGH), but neither bot took it. The reason was straightforward: the setup failed bot conversion, either on contract/liquidity constraints or a pre-filter rejection before a valid executable trigger materialized. That is not a bug. Sometimes the discretionary read and the execution stack do not align, and forcing the trade is how mediocre systems become bad ones. You can see live setups in the scanner and compare grade, pressure, and trigger quality directly.
Highest RVOL Setup
Where the volume was today
ORCL BEAR B posted the highest RVOL on the board at 2.14x and resolved as a breakout.
This was one of the cleaner tells of the day. Elevated relative volume plus a bearish resolution in a weak growth tape usually means institutions were actually involved, not just retail noise. There was no giant macro shock driving it. This looked more like sector pressure and stock-specific urgency inside enterprise tech, and that is exactly the kind of move worth respecting when the rest of the market is only modestly red.
Bot Performance
No live executions while the strategy gets rebuilt
Wex and Xcel are in training mode while strategy is being refined. The scanner and Arxe intelligence remain fully active. When trading resumes, every trade will be published here. You can watch Wex and Xcel trade live in the Edge Lab.
Missed Trades
The bots missed breakouts, but none of them were painful misses
Missed count was six breakouts, but total missed opportunity was effectively noise. No TP1 hits, no TP2s, no TP3 full runners, and aggregate missed R came in at +0R. That tells you the filters were not asleep at the wheel. They were mostly keeping the bots out of low-payoff breaks in a tape that lacked extension.
Calibration insight: today did not punish caution, but ORCL shows the system may still be too slow on high-RVOL bearish tech when sector momentum is clearly breaking down.
Breakout Scorecard
The scanner found movement, not quality trend extension
Notable names that did break included APP BEAR A-, CELH BEAR B, CRM BULL A-, NFLX BEAR A-, ORCL BEAR B, and QCOM BEAR B.
That lineup says a lot. The better downside action came in tech and growth, while the upside breaks were sparse and underpowered. A 28.9% breakout rate on a day with no A grades is a market telling you to trade smaller, hold winners less aggressively, and stop pretending every trigger deserves swing treatment.
The breakout rate says tape quality was mediocre and selective, not broken across the board but nowhere near good enough for loose risk discipline.
Watch Tomorrow
Specific levels that matter if the market gives clean triggers
These are the unresolved A- setups still coiling into tomorrow. No vague watchlist filler here. These are the names worth stalking if price confirms. You can see live setups in the scanner as these levels update.