SPY slips 0.25% as 10 of 106 setups broke out; GDX was the pick
Market recap, bot performance, and scanner analysis for Tuesday, August 11.
The UnxEdge breakout scanner tracked 106 wedge pattern setups today, and only 10 actually broke clean while the indices went nowhere fast. That is the whole story: plenty of formations, not much follow-through, and the wedge breakout names that worked were selective rather than broad. For context, yesterday's debrief showed a tape that already looked fragile.
Macro was the real governor on risk. With CPI due before the bell tomorrow and PPI right behind it, traders spent the session cutting conviction and avoiding extension. That helps explain why SPY and QQQ drifted lower while IWM managed a green close. Small caps caught relative bids, but not enough to improve the quality of breakout participation across the board.
The stock-specific catalysts were there, but they were narrow. CoreWeave's strong Q2 and backlog commentary reinforced the AI infrastructure trade, which helped keep semi and compute-adjacent appetite from fully rolling over, even if it did not translate into a broad momentum day. Meanwhile, Realty Income's proposed $750 million convertible notes deal was a reminder that financing headlines still matter in a market heading into inflation data. Sector-wise, biotech is rotating in week over week while gold and metals remain the cleanest relative-strength pockets. That backdrop matters because today's best statistical swing pick was not some hot beta chase, it was GDX.
Market Snapshot
Flat index tape, weak breakout conversion
The scanner found quantity, not quality. A-grade inventory was thin, and only 10 of 43 resolved setups became breakouts. That 23.3% conversion rate is not the kind of tape where you spray entries and hope. It is the kind where level discipline matters and anything below top-tier structure gets punished. For setup grades and live levels, see live setups in the scanner.
Yesterday's Pick
DIA BULL got stopped
No spin needed. Yesterday's DIA bull idea failed, and it failed before any real trend extension showed up. That is exactly the problem with leaning long into a hesitant tape just ahead of CPI: your stop gets tested before your thesis gets validated.
Arxe Pick of the Day
GDX had the cleanest statistical edge
Verdict: trade_it
Why this one: highest backtest win rate at 95.45%, profit factor of 21.01, and the strongest pressure reading at 75 among the comparable candidates.
Context: all three candidate setups carried the same A- grade, similar condition, RVOL, and bar count, so GDX won on pure edge. With gold continuing to outperform SPY by a wide margin and inflation data due tomorrow, the macro backdrop actually supports the selection rather than fighting it.
GDX scanner chart – August 11, 2026 | UnxEdge
This is the kind of pick that makes sense in a messy tape. Instead of forcing growth beta, the model leaned into the strongest relative-strength pocket on the board. Gold has not just been outperforming, it has been persistently outperforming, and that matters more than a cute intraday bounce elsewhere. For current grades and breakout levels, see live setups in the scanner.
Separately, today's Arxe swing pick was PYPL BULL (HIGH), but neither bot took it. The clean explanation is that the setup did not convert into a valid automated trade path under current pre-filters while both systems remain paused for refinement.
Highest RVOL Setup
Where the volume was today
High relative volume did not mean high-quality downside follow-through. PENN printed the biggest RVOL on the board at 3.04x and still failed. That usually tells you the flow was event-driven or reactive rather than structurally clean. In a better tape, big volume confirms. In a choppy tape ahead of macro, big volume can just as easily trap late entries.
Bot Performance
Nothing live until the strategy is ready
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 three, and none of them were serious damage
Total missed upside was +0R in any meaningful sense because none of these reached TP1, TP2, or TP3 territory. These were minor moves, not painful omissions. CVS and UNH had weak RVOL, which is exactly the kind of thing a decent filter should hesitate on. SNAP was the only one with A- structure, but even there the follow-through was modest.
Filter insight: today was not about missing runners, it was about avoiding mediocre participation in a low-conviction tape, and the data says the filters were mostly right to stay selective.
Breakout Scorecard
Too many failures, not enough expansion
Notable names that did break included ANET BULL A-, BROS BEAR B+, CAVA BEAR B+, CELH BULL B, CVS BEAR B, and JPM BULL A-. That list tells you the market was not trading one clean theme. It was fragmented, alternating between defensive cracks, selective financial strength, and isolated momentum pockets.
A 23% breakout rate says the tape was tradable only if you demanded high-quality structure and accepted that most patterns were noise.
Watch Tomorrow
CPI morning, so levels matter more than opinions
These are the unresolved A and A- names still coiling into tomorrow. Use the trigger, not the story. For live updates, see live setups in the scanner.