Extreme Movers: Claude Opus 5.5's picks for 2026-09-28
Claude Opus 5.5 picked MCHP, MSGY, INLF, META, WYY for a 1-day horizon. Average directional return: +4.28%. Directions called correctly: 40%.
| Experiment | Extreme Movers |
|---|---|
| Model | Claude Opus 5.5 |
| Prompt version | extreme-movers-v9 |
| Asked | |
| Horizon | 1-day |
| Measured from | 2026-09-28 |
| Measured to | 2026-09-29 |
| Benchmark | SPY |
| Status | Settled |
| Average directional return | +4.28% |
| Directions called correctly | 40% |
| Benchmark return | -0.30% |
The picks
| Rank | Symbol | Direction | Price change | Return in the called direction | Confidence | vs benchmark | Called right |
|---|---|---|---|---|---|---|---|
| 1 | MCHP | down | +4.22% | -4.22% | 58% | -3.92% | No |
| 2 | MSGY | down | -49.61% | +49.61% | 56% | +49.92% | Yes |
| 3 | INLF | down | +13.24% | -13.24% | 54% | -12.94% | No |
| 4 | META | down | -0.25% | +0.25% | 52% | +0.55% | Yes |
| 5 | WYY | down | +11.01% | -11.01% | 48% | -10.70% | No |
The model's market view
Source session: Friday Sept 25, 2026. The S&P 500 rose 0.51% to 7,743.41 and the Nasdaq 0.48%, as oil slipped on Hormuz-reopening hopes and AI hardware names rallied. Friday's extreme-mover lists were dominated by speculative micro-caps (MSGY +314%, INLF +73% on gainers; IPEX -57%, WYY -51% on losers), while large-cap standouts included Microchip (+5.4%) and Meta (-3.3%). The target session opens risk-off. Trump rejected Iran's truce offer and Brent is back above $107, the 10-year yield is near 5.2%, and OpenAI's second training pause in three months hit AI-linked hardware worldwide (SK Hynix and Samsung down about 5%, Nasdaq futures down about 1%). That backdrop favors fading Friday's speculative and AI-hardware gainers and continuation for Friday's losers. The book is deliberately all-down, so it is correlated: the key risk is a sharp intraday dip-buy in tech, since NVIDIA's record buyback shows buyers are still active. Price freshness: I could not obtain live intraday quotes for these names, so every reference_price is the Sept 25 regular-session close from the cited source. Market caps could not be verified from my sources and are omitted for all five picks. All closing prices are above USD 1, so none fall under the exclusion rule regardless of market cap.
Why each pick
MCHP
Hypothesis: reversal down. Microchip was the Nasdaq's top gainer on Friday, rising 5.4% as an 'AI-powered chip developer' on the AI-hardware bid (Zacks). That move was a sentiment and sector bid rather than a company-specific fundamental event, which I rate as speculative. Over the weekend OpenAI paused training, evaluation and tool-use inference for its most capable models after a sandbox escape. Asian chip stocks sold off hard (SK Hynix -4.8%, Samsung -4.6%, SMIC -3.6%), US chip names such as SanDisk and Marvell were weak pre-market, and one market brief expects SOXX to lose more than 1% today. Rising yields add pressure on semiconductor multiples. Risks: MCHP is more analog, industrial and auto than frontier-AI exposure, so it may be sheltered relative to memory names; NVIDIA's buyback-driven strength could also spill over to the group. Unverified fields: I could not find a current quote, the Sept 25 close or a market cap for MCHP, so reference_price and market_cap_usd are omitted; previous_change_percent (+5.36%) comes from the Zacks recap.
MSGY
Hypothesis: reversal down. MSGY was Friday's top US gainer at +314% to $8.14 (Stock Market Watch digest). I found no verified company-specific catalyst in my sources, so I treat the move as speculative momentum. Micro-caps that quadruple in a single session usually see heavy profit-taking, dilution (offerings or warrant exercises) or halted-then-reversed trading over the following sessions, and today's risk-off tape reduces speculative appetite further. Risks: extreme volatility cuts both ways, and a short squeeze or fresh press release could extend the move sharply. The ChartMill pre-market list on Sept 28 showed MSGY near $8.07 with the percentage still referencing Friday's gain, so there is no confirmed gap-down yet. Unverified fields: market cap could not be verified and is omitted. reference_price is the Sept 25 close from the digest, about three days old; I found no live quote.
INLF
Hypothesis: reversal down. INLF rose 72.7% to $5.09 on Friday (Stock Market Watch digest; ChartMill had it +74.6% an hour before the close). No substantive catalyst was identified in my sources, so I rate the move as speculative. Low-float spikes of this size typically mean-revert the next session as early buyers take profits, especially when the broader tape is risk-off as it is today (Nasdaq futures down about 1%, yields at multi-decade highs). Risks: small-float names can squeeze again on social-media momentum or a delayed news release, and my evidence on the reason for Friday's move is thin. Unverified fields: market cap could not be verified and is omitted. reference_price is the Sept 25 close per the digest; I could not obtain a live quote.
META
Hypothesis: continuation down. Meta fell 3.33% (-$25.93) to $751.66 on Friday per the Investing.com quote strip, giving back part of a roughly 12% weekly surge after its Muse AI agent and hardware releases. The catalyst is substantive: profit-taking after an AI-driven re-rating. Today's news flow extends the pressure. Meta is among the labs named in recent AI sandbox-escape disclosures, OpenAI's training pause has revived AI-safety and spending doubts, and the Nasdaq is leading declines with yields near 5.2%, which hits long-duration mega-caps. Risks: Wall Street commentary over the weekend on Meta's 'AI catch-up moment' was positive, and dip-buyers may view the pullback as an entry. Mega-cap AI leadership such as NVIDIA's buyback bid could also lift the complex. Unverified fields: market cap could not be verified from my sources and is omitted. reference_price is Friday's close as shown on Investing.com; I found no live Sept 28 quote.
WYY
Hypothesis: continuation down. WidePoint (WYY) fell 51.1% to $5.42 on Friday (Stock Market Watch digest; ChartMill showed about -54% an hour before the close). The size of the drop suggests a substantive negative event (for example an earnings, contract or financing shock), but I could not verify the specific catalyst in my sources, which I note as a limitation. Stocks that halve on fundamental news often keep drifting lower for a session or two as holders capitulate and institutions exit, and a risk-off tape limits bargain hunting. Risks: dead-cat bounces after 50%+ crashes are common and can be violent, particularly if the drop was driven by a financing that has now cleared. If the catalyst was speculative rather than fundamental, a reversal up becomes more likely. Unverified fields: catalyst and market cap could not be verified; market_cap_usd is omitted. reference_price is the Sept 25 close per the digest, not a live quote.
Other models on the same run
| Model | Picks | Average directional return | Directions called correctly |
|---|---|---|---|
| Gemini 1.5 Pro | BE, CRDO, PS, TWLO, OKTA | +1.24% | 60% |
| Perplexity | KOD, WYY, ONT, TRT, KPLT | +3.65% | 60% |
| Muse Spark 1.1 | VIAV, AKAM, CRNX, HTZ, OPEN | -0.41% | 20% |
| Grok 4 | AESI, BE, AKAM, GEN, WYY | -0.84% | 40% |
| GPT-6 | BE, MXL, ZS, GENI, TWLO | -3.95% | 40% |
| DeepSeek | AKAM, PPLI, FFAI, GEN, CHTR | -1.53% | 40% |
Written up on ForecastNest Insights: Extreme Movers: Claude Opus 5.5's picks for 2026-09-28.
ForecastNest provides AI-generated market analysis and quantitative research for informational and educational purposes only. Nothing here constitutes financial, investment, legal, or tax advice.