Solar Cluster Lit, Phase 3 Cracked, A Quiet Beat Ran 28% — Week of May 11, 2026
*Your weekly market recap, brought to you by [SignalDeck.live](https://signaldeck.live).*
The week’s marquee win wasn’t a marquee call — OUST (Ouster) ran +28.49% from a Tuesday “highest-upside speculative” reversion play after its earnings-miss gap got bought back over four sessions. The week’s worst pain came from a stock we told you NOT to chase: POET printed -22.07% by Friday close, validating the “elevated bar / sell-the-news” caution. The regime read: Phase 3 mania finally cracked — CRDO, MRAM, and a handful of parabolic semis ate double-digit losses on the week — while a fresh solar/AI cluster set up Friday with SEDG, ENPH, and SHLS all printing double-digit gaps on volume.
## The Top Calls
- **OUST (Ouster)** · Electronic Components · Picked 2026-05-12 as highest-upside speculative · Friday relative strength 98 · **+28.49% in 3 sessions** — the contrarian R10.2 reversion thesis nailed it. After OUST’s 5/7 earnings-miss gap-down of -16.69%, our screener flagged the post-distribution recovery starting to bid. Tuesday’s volume confirmation closed the case; the rest of the week was institutional bid layering. This is the textbook “buy the bounce when the bid actually shows up” trade, not “buy the dip and hope.”
- **BW (Babcock & Wilcox)** · Specialty Industrial · Picked 2026-05-11 as a CANDIDATE (only 60 score) · Friday relative strength 94 · **+15.46% on the week** — the under-the-radar surprise. Earnings printed a clean revenue blowout (+43% surprise), $2.5B in bookings, $14B pipeline. Cold industrials sector mostly killed the conviction tier but the institutional bid showed up anyway. The lesson: don’t sleep on CANDIDATE-tier picks when the fundamentals are this loud, even in a cold cluster.
- **NXT (Nextracker)** · Solar · Picked 2026-05-12 after hours · Friday relative strength 97 · **+15.04% in 3 sessions** — A6.4 mega-cap structural-raiser archetype firing exactly like AMD did two weeks ago. Q4 record, FY26 guide raised, Q2 raised. The forward EPS curve re-prices and institutional accumulation runs days, not minutes. Solar’s becoming the new AI-adjacent leadership pole alongside semis.
- **MDB (MongoDB)** · Software-Infrastructure · Picked 2026-05-11 · Friday relative strength 91 · **+5.95% on the week** — database-modernization compounding thesis intact. The boring trades are the ones that pay rent.
- **NBIX (Neurocrine)** · Drug Manufacturers · Picked 2026-05-11 with perfect quality score · Friday relative strength 90 · **+5.32% on the week** — drug-launch pipeline grinding higher in a market that mostly punished biotech.
> Pattern that fired hardest: the A6.4 mega-cap structural-raiser archetype (raised forward guide on the call with structural narrative). NXT was this week’s instance. AMD was the proof-of-concept two weeks ago. When mega-caps raise both the current quarter and the multi-year forward guide on the same call, the trade is days-long bid, not a same-day pop.
## The Quiet Beats
Friday’s earnings docket printed five separate STRONG signals — too early to formally score them, but they’re the multi-week setups going into next week.
- **SEDG (SolarEdge)** · Solar · Picked 2026-05-15 · Relative strength 99 (elite) · **+22.85% same-day** — Q1 print + Q2 raise + AI data center power pivot. The “6th consecutive quarter of margin expansion” language is the kind of thing that re-rates the multiple. The most asymmetric setup going into Monday.
- **FIG (Figma)** · Software-Application · Picked 2026-05-15 · Relative strength 89 · **+13.17% same-day** — fresh IPO first major print as a public company. FY26 guide raised to $1.42B (+35%), net-dollar-retention 139%, customer +54%. Analyst models still being calibrated → multi-session re-rating runway.
- **ENPH (Enphase)** · Solar · Picked 2026-05-15 · Relative strength 99 (elite) · **+10.16% same-day** — IQ9S preorders open + 1.25MW solid-state transformer for AI data centers announced. Section 25D residential tax credit deadline pulling forward demand.
- **SHLS (Shoals)** · Solar · Picked 2026-05-15 · Relative strength 98 · **+11.15% same-day** — Q1 revenue +75% YoY, record $758M backlog. Solar cluster sympathy.
- **CSCO (Cisco)** · Communication Equipment · Picked 2026-05-14 · Relative strength 98 (elite) · **+2.35% in 1 session** — A6.4 mega-cap archetype, currently in the slow-accumulation phase. Mirrors how AMD looked at Day 1 before extending +29% by Day 4. Patience pays here.
> The Friday solar trio (SEDG, ENPH, SHLS) all sit at relative strength 97–99 on a single-day catalyst stack: residential pull-forward, AI data center power demand, structural margin inflection. When three names in one cluster print double-digits on the same day with elite relative strength, the cluster is rotating, not a one-off.
## This Week’s Lessons
- **POET pattern** — explicitly flagged “DO NOT CHASE” pre-earnings on 5/14 because it had already run from $7 to $15 into the print. Result: **-22.07% by Friday close**. The “elevated bar / sell-the-news binary” caution is now graduated from a tentative rule to a confirmed one. Any stock that doubles or triples into earnings should be skipped through the print, regardless of how clean the fundamentals look.
- **CRDO pattern (Phase 3 mania crack)** — was the Monday “Quiet Beat sympathy” call that printed +11.56% on Day 1. By Friday: -18.24% cumulative, relative strength collapsed from 96 to 52. Phase 3 mania names finally cracked this week (CRDO, MRAM -14.69%, MNDY -6.39%). The “don’t chase Phase 3” framework held — but the lesson is also “take profits when relative strength is climactic” instead of riding sympathy plays into the crack.
- **EVER pattern** — was the Monday “best risk-adjusted” call with perfect quality score and price-to-earnings of 7. Result: -11.69% on the week, relative strength crashed from 94 to 47. The volume-on-entry-day signal was 0.81x (below the 1.0x threshold we’d just banked from the prior week). Perfect fundamentals don’t override a missing volume signal — when the institutional bid isn’t there on the entry day, sit out.
- **VRDN pattern (fakeout)** — Tuesday morning’s “Best risk-adjusted today” promotion at relative strength 86. Same-day close: collapsed -1.98%, relative strength dropped 34 percentile points in one session. By Friday: flat. The positive volume slope across the lookback was a lagging indicator — the leading signal was the missing same-day volume confirmation. New rule banked: Phase 2 promotions require same-day volume confirmation, not just a positive vol slope across the prior two weeks.
- **KLAR pattern (fresh IPO caveat)** — Wednesday’s top-N #4 pick at score 89 (highest in the alert). Result: -7.80% by Friday with relative strength holding 76. Fresh-IPO stocks have unstable volume baselines — what looks like a 4x volume signal in premarket can be thin-air after the open. The “wait for first 15-minute volume confirmation” gate stays in place.
- **Hit-rate by category this week:**
- **STRONG-tier earnings signals (scoreable):** 17 picks → 24% hit rate, 18% miss rate. Median NEUTRAL — the week’s regime favored slow accumulation over instant pops.
- **CANDIDATE-tier earnings signals (scoreable):** 17 picks → 35% hit rate, 41% miss rate. The CANDIDATE bucket includes more sector-headwind names; the misses cluster there for a reason.
- **Trend-continuation entries from Mon/Tue scans (scoreable):** 13 picks → 23% hit rate, 38% miss rate. The single OUST hit (+28.49%) carried the whole bucket.
- **Quiet Beats from Mon scan:** 0 hits / 5 picks → the cohort that lived in software-application and biotech got eaten by the Phase 3 crack.
> **FLAG:** SP “Best risk-adjusted” calls this week scored 0-for-2 (EVER, VRDN). Both had perfect quality scores but failed the same-day volume gate. The “best risk-adjusted” label is currently mis-calibrated — needs tighter criteria. Working on it.
## The Read
**Best call of the week:** OUST +28.49%. The reversion-after-distribution-gap thesis is one of the highest-conviction setups in the framework when the volume confirmation actually shows up. We waited four sessions for the bid to print, then leaned in.
**Worst call of the week:** EVER -11.69% (with CRDO -18.24% as the runner-up). EVER was actually called as “best risk-adjusted” Monday — that label is now under review. Two of two same-day misses for that bucket this week says the calibration is off.
**Best two-week compounding setup going into next week:** SEDG. Relative strength 99, +22.85% Friday on raised Q2 guide, AI data center pivot, sixth consecutive quarter of margin expansion. The mega-cap structural-raiser archetype is the highest-conviction multi-day trade in our system right now, and SEDG fits it exactly. Cross-check on Wednesday for follow-through.
**Don’t chase into Monday:** POET (-22% confirmed sell-the-news), CRDO (-18% Phase 3 crack), MRAM (-15%). Wait for a 3-day relative strength index hook below 80 before considering re-entry. Phase 3 names that just cracked rarely bounce immediately — they consolidate or keep bleeding for a week first.
**Lesson banked:** The “elevated bar / sell-the-news” caution is now a confirmed rule. Any stock that doubles or triples into an earnings print should be skipped through the print, even if our screener loves the fundamentals. POET’s $7→$15 pre-print run was the entire fuel — the actual print delivered the burn-off. We had the caution flagged correctly; the win is in the discipline of NOT taking the trade, not in the postmortem.
---
**Not financial advice.** This is independent market analysis and momentum research, not a recommendation to buy or sell anything. Markets are volatile, individual securities can lose 100% of their value, and past performance does not predict future results. Do your own due diligence, size positions you can afford to lose, and consult a licensed financial advisor before making investment decisions. Your money, your risk, your call.