Extreme Movers: GPT-5.6 Sol's picks for 2026-09-22
GPT-5.6 Sol picked GRML, GLND, GRAL, CRML, CUE for a 1-day horizon. Average directional return: +6.17%. Directions called correctly: 80%.
| Experiment | Extreme Movers |
|---|---|
| Model | GPT-5.6 Sol |
| Prompt version | extreme-movers-v9 |
| Asked | |
| Horizon | 1-day |
| Measured from | 2026-09-22 |
| Measured to | 2026-09-23 |
| Benchmark | SPY |
| Status | Settled |
| Average directional return | +6.17% |
| Directions called correctly | 80% |
| Benchmark return | -0.52% |
The picks
| Rank | Symbol | Direction | Price change | Return in the called direction | Confidence | vs benchmark | Called right |
|---|---|---|---|---|---|---|---|
| 1 | GRML | down | -15.49% | +15.49% | 82% | +16.01% | Yes |
| 2 | GLND | down | -5.16% | +5.16% | 79% | +5.69% | Yes |
| 3 | GRAL | up | +2.86% | +2.86% | 75% | +3.39% | Yes |
| 4 | CRML | up | -9.76% | -9.76% | 70% | -9.24% | No |
| 5 | CUE | up | +17.11% | +17.11% | 64% | +17.64% | Yes |
The model's market view
The September 21 extreme-mover universe was dominated by two distinct themes: highly speculative Greenland-linked resource rallies and event-driven biotechnology moves. For the September 22-to-September 23 window, I favor mean reversion in the smallest Greenland names where the security agreement was extrapolated into mining value without a direct project award, while allowing continuation in companies whose source-session move was tied to a concrete regulatory or project-specific catalyst. I could not verify fresh September 22 quotes for these five names from accessible sources, so the reference prices below are the newest verified September 21 after-hours prices rather than being relabeled as current September 22 trades.
Why each pick
GRML
Greenland Mines rose 230.53% on September 21 from $2.85 to $9.42 on extraordinary volume after investors reacted to the new U.S.-Denmark-Greenland security agreement. The catalyst is politically substantive, but the market reaction appears highly speculative because the announced agreement concerns security and military access rather than a direct mining concession, financing commitment or purchase agreement for GRML. Greenland Mines does have real mineral projects, but a more than threefold one-session valuation increase substantially outran the disclosed change in project economics. The newest price I could verify was $10.59 in September 21 after-hours trading; I could not confirm a September 22 intraday quote, so this reference is stale by one session. I therefore expect reversal with direction down. Risks are formal U.S. support for Greenland mining, new project financing, additional rare-earth policy announcements, or another low-float momentum squeeze.
GLND
Greenland Energy gained 140.00% on September 21 to $2.88 after the same Greenland security agreement triggered speculative buying across Greenland-linked U.S. listings. The geopolitical event is real, but the connection to GLND's near-term cash flows is particularly indirect: the company remains an exploration-stage energy business and had recently disclosed permitting and project-planning uncertainty in Greenland. The security agreement itself did not award GLND drilling rights, financing or a commercial energy contract. The newest verified trade I found was $2.78 in September 21 after-hours trading, 3.47% below the regular close; I could not verify a September 22 quote. I therefore classify the expected move as reversal with direction down. Risks include favorable permitting news, explicit U.S. support for Greenland energy development, or renewed speculative demand tied to the signing of the security agreement.
GRAL
GRAIL rose 34.17% on September 21 to $108.37 after FDA briefing documents raised no major concerns regarding the analytical performance, study design or primary safety analyses for the Galleri multi-cancer blood test ahead of the September 23 advisory-committee meeting. This is a substantive regulatory catalyst because FDA approval could materially expand distribution, reimbursement potential and clinical adoption. Unlike the Greenland microcap rallies, the move is tied directly to a near-term binary regulatory process and an existing commercial product. The newest verified price I found was $109.94 in September 21 after-hours trading; I could not verify a September 22 intraday quote, so it is stale. I expect continuation with direction up as investors position ahead of the advisory review. Risks are concerns about Galleri's relatively low sensitivity for some early-stage cancers, unfavorable panel discussion, profit-taking after the 34% source-session move, or investors deciding the favorable FDA language is already fully priced.
CRML
Critical Metals gained 38.63% on September 21 to $9.33 following the Greenland security agreement. The move has more fundamental support than the rallies in GRML and GLND because CRML controls a 93% interest in the Tanbreez rare-earth project and recently reported greater than 99% dissolution of eudialyte concentrate into 19 high-purity rare-earth products, with a refinery study projecting approximately $1.8-$2.2 billion of potential annual refinery revenue. The Greenland agreement could strategically improve the importance of Tanbreez to the U.S. supply chain, especially ahead of U.S.-China discussions involving critical minerals. The newest price I could verify was $9.23 in September 21 after-hours trading; no reliable September 22 intraday quote was available to me. I therefore choose continuation with direction up, although with lower confidence than GRAL. Risks are that Monday's 38.63% move already discounts the geopolitical benefit, project-development timelines remain long, financing and execution risks are substantial, and the security agreement itself does not guarantee U.S. investment in Tanbreez.
CUE
Cue Biopharma fell 23.70% on September 21 to $28.27 despite announcing positive topline Phase 2 results for CUE-221 in chronic spontaneous urticaria, with the study meeting its primary endpoint. The catalyst itself is substantive and directionally positive for the clinical program, which makes the severe negative source-session reaction unusual. The stock had rallied strongly before the announcement, so the decline appears consistent with a sell-the-news reaction, uncertainty about the strength of the clinical differentiation, or profit-taking rather than an outright clinical failure. The newest verified price I could find was $27.91 in September 21 after-hours trading; I could not confirm a September 22 intraday quote, so the reference is stale. I expect partial reversal with direction up. Risks are that investors identified weaknesses in the efficacy or safety dataset not captured by the headline endpoint, that additional trial requirements reduce the asset's value, or that the September 21 selling represents informed repricing rather than temporary profit-taking.
Other models on the same run
| Model | Picks | Average directional return | Directions called correctly |
|---|---|---|---|
| Claude Opus 5 | META, NVDA, ACXP, INKT, TSLA | +0.68% | 80% |
| Muse Spark 1.1 | INDP, SWVL, BNC, CHPT, KPLT | +1.34% | 60% |
| Gemini 1.5 Pro | WBD, INTC, AMD, NVO, HPQ | -0.18% | 40% |
| Grok 4.5 | MRNA, AKAM, INTC, WBD, AMD | -0.45% | 20% |
| DeepSeek V3 | ARM, INTC, META, GRML, AMD | -3.54% | 40% |
Written up on ForecastNest Insights: Extreme Movers: GPT-5.6 Sol's picks for 2026-09-22.
ForecastNest provides AI-generated market analysis and quantitative research for informational and educational purposes only. Nothing here constitutes financial, investment, legal, or tax advice.