SPY falls 1.40% as 37 failed breakouts outnumber 32 winners; AMD picked

Risk got sold first, questions came second.

The tape leaned defensive all session and there was no mystery behind it: soft breadth, small caps getting hit hardest, and yet another reminder that when credit headlines start creeping into the feed, traders stop paying growth multiples and start paying attention to liquidity. The DOJ probe headline around BlackRock private credit valuations was not a direct equity market bomb, but it was the kind of macro-adjacent headline that keeps pressure on anything reliant on easy financing and rosy asset marks. That showed up where it usually does first, in IWM, housing, retail, and the lower-quality end of the market.

At the same time, the market did not get a clean bullish tech offset. The Nvidia disclosure headlines around CoreWeave, Intel, Synopsys, Coherent, and others kept the AI infrastructure conversation alive, but they did not translate into fresh broad-based semi leadership today. That matters because when QQQ is red and semis are not bailing it out intraday, dip buyers lose their best excuse. Energy stayed the relative winner on the week while housing and materials have clearly rotated out week over week, which is not the profile of a tape that wants to reward late long entries.

Tradeable implication: if you were leaning long, you needed isolated strength and clean trigger discipline. If you were leaning short, the better risk was in weak cyclicals, regional financials, and names already failing relative support. The broad market was not impossible today, just unforgiving. For context, here is yesterday's debrief.

Index damage was real, especially under the surface

The scanner stayed active despite the weakness: 61 bulls, 57 bears, with 7 A-grade setups total. Resolution quality was mediocre. We got 32 breakouts against 37 failures and 3 expirations, which is not disaster tape, but it is not efficient trend tape either. Near any live grade or breakout level, see live setups in the scanner.

DELL is still open, but it did not resolve today

AMD got the nod because grade beat noise

Today's Pick

AMD scanner chart – May 15, 2026 | UnxEdge

<div class="pick-symbol"><a href="/ticker/AMD" style="color:#00C851;text-decoration:none;">AMD</a></div>
<p>Verdict: trade it.</p>
<p><a href="/ticker/AMD" style="color:#00C851;text-decoration:none;">AMD</a> was the only A++ setup in the stack, and that matters more than distance or pressure when the rules force a clean hierarchy. The alternatives at A+ simply did not have enough sample size to earn trust. <a href="/ticker/AMD" style="color:#00C851;text-decoration:none;">AMD</a>'s backtest is not massive, but it is credible enough to matter: 7 trades, 71.43% win rate, 2.51 profit factor, and 0.4307 average R.</p>
<p>The external backdrop was neutral rather than supportive. News flow around AI hardware and supply chain commentary kept the name in the conversation, but there was no direct bearish company-specific hit behind the setup. In a weak tape, that is often enough. The pick was not about a flashy catalyst. It was about having the best statistical profile on the board. For current grades and trigger proximity, <a href="/proximity">see live setups in the scanner</a>.</p>

Where the volume was today

This is the kind of stat traders misuse. High RVOL does not automatically mean valid breakout. In this case, the industrials volume spike looked more like noisy rotation and failed participation than durable institutional sponsorship. When an A- setup with nearly 4x relative volume cannot convert, it tells you the market is demanding much cleaner follow-through than the scanner alone can guarantee.

Execution was messy, and the logs say so

For live bot activity and real-time monitoring, watch Wex and Xcel trade live in the Edge Lab.

Wex took three trades, but all three came back as orphaned execution records with no Alpaca position found. That means there is nothing useful to celebrate or analyze from P&L. It is an execution integrity issue first, strategy issue second.

Xcel did not execute today. Given the tape, that is not automatically a miss. Sometimes no trade is cleaner than forcing bad ones. But when the market still produced several valid breakouts elsewhere, zero execution also raises the question of whether filters are too tight or the trigger stack is lagging. Again, watch Wex and Xcel trade live in the Edge Lab.

The system left +10.1R on the table

The headline miss is XLB. A nearly 4R bear runner on a day when cyclicals were under pressure is the exact type of move a breakout engine should at least be sniffing. CVNA was another meaningful miss, and that one hurts more because high-beta consumer risk was clearly in trouble all day.

Some misses are acceptable filter discipline. V and CVX were cleaner counterexamples, but they only reached TP1 and did not define the session. The real blind spot was bearish participation in weak groups, not isolated longs.

More failure than follow-through

Notable confirmed names included BMY BULL B, C BEAR B+, CVNA BEAR A, CVX BULL A-, DDOG BULL A-, and DECK BEAR A-. That mix says the same thing the indices did: this was not a unified market, it was a selective market with a defensive tilt.

The levels that actually matter next

These are the unresolved A and A- setups still coiling. Entries stay conditional. No guessing, no front-running. For live changes in distance and grade, see live setups in the scanner.

ABBV A- BULL. Current distance 1.086%, pressure 68, RVOL 1.0x. Entry trigger above today's breakout level once confirmed in the scanner. Stop below the reclaimed trigger candle low. TP1 should be set at 1R from entry.

ABNB A- BEAR. Current distance 0.671%, pressure 66, RVOL 1.0x. Entry trigger on a break of support confirmation. Stop above the trigger candle high. TP1 at 1R.

ANET A- BEAR. Current distance 0.109%, pressure 53, RVOL 1.0x. This is the closest coil in the group. Entry triggers only on clean downside confirmation. Stop above the immediate failed pivot. TP1 at 1R.

APP A- BULL. Current distance 0.788%, pressure 78, RVOL 1.0x. Strong pressure score, but the market needs to cooperate. Entry only through breakout confirmation. Stop below local structure. TP1 at 1R.

AVGO A- BULL. Current distance 1.585