Event Gaps: Claude Opus 5's picks for 2026-09-25
Claude Opus 5 picked IONQ, ORCL, CBRL, MCD, DIS for a 1-day horizon. Intraday alpha vs SPY: -0.25%. Picks with positive alpha: 40%.
| Experiment | Event Gaps |
|---|---|
| Model | Claude Opus 5 |
| Prompt version | event-gap-v9 |
| Asked | |
| Horizon | 1-day |
| Measured from | 2026-09-24 |
| Measured to | 2026-09-25 |
| Benchmark | SPY |
| Status | Settled |
| Intraday alpha vs SPY | -0.25% |
| Picks with positive alpha | 40% |
| Benchmark return | +0.33% |
The picks
| Rank | Symbol | Gap direction | Price change | Return in the called direction | Confidence | vs benchmark | Called right |
|---|---|---|---|---|---|---|---|
| 1 | IONQ | up | +1.74% | +1.74% | 45% | +1.41% | Yes |
| 2 | ORCL | down | -1.06% | -1.06% | 48% | -1.40% | No |
| 3 | CBRL | up | -1.11% | -1.11% | 40% | -1.44% | No |
| 4 | MCD | down | -0.12% | -0.12% | 38% | -0.45% | No |
| 5 | DIS | up | +0.95% | +0.95% | 36% | +0.62% | Yes |
The model's market view
The dominant macro event is a bond-market shock, not a corporate one. Bloomberg reports renewed bond volatility on 2026-09-24 with 30-year yields at their highest since 2004 and stocks falling on fears that elevated oil forces more Fed hikes; CNBC has the Nasdaq down 0.8%, the Dow off 349 points and the 10-year at 5.15%. President Trump hosts Xi Jinping in Washington today, an unscheduled-headline risk that can move the whole tape overnight and swamp single-stock gaps. Because the target session's open-to-close path is what is scored on a long basket, I have leaned toward names with genuine company-level catalysts rather than pure macro beta. Important limitation: I could not verify intraday or closing quotes for any of these names at the time of answering, so reference_price and price_source are omitted for every selection, and my gap magnitudes are judgement calls anchored to reported percentage moves rather than to observed after-hours prices. None of the five has a confirmed after-hours print in my evidence, so no gap forecast here is a copy of an observed after-hours move.
Why each pick
IONQ
Thesis: the catalyst has occurred and is still being absorbed. IonQ announced a real-time quantum error correction decoder running on a single standard CPU (shares +11.35% premarket, +5.89% at midday on 2026-09-23 per TheStreet), and TheStreet separately reported that Nvidia plans to install IonQ technology in a hybrid quantum-classical setup. Multi-day narrative catalysts of this type tend to keep drawing flows, so I expect a modest continuation gap up and a positive open-to-close bias if the tape stabilises. Risks: rising real yields are the worst environment for a cash-burning pre-profit name, and IONQ frequently retraces gains of this size within a session; emotional continuation after a double-digit premarket pop is exactly where fades happen. Price: I could not verify a live or closing quote at answer time, so reference_price is omitted.
ORCL
Thesis: the catalyst has occurred but the information is still spreading. Bloomberg reported on 2026-09-24 that Oracle is citing force majeure to protect itself on a New Mexico data-center project if it is delayed; the stock fell 5.4% premarket and more than 5% in the session per CNBC. Negative disclosures about AI-infrastructure commitments have been compounding all week, and with 30-year yields at 2004 highs the market is punishing capital-intensive AI buildouts, so I expect further drift lower into the next open. Note this is a gap-down forecast within a long-only basket, which is why confidence in the basket contribution is modest. Risks: a 5% single-day drop in a megacap often stabilises quickly, and any clarifying company statement or a Trump-Xi-driven risk rally could produce a sharp bounce instead. Price: no verified quote at answer time, so reference_price is omitted.
CBRL
Thesis: the earnings event has occurred. Cracker Barrel beat fourth-quarter expectations and rose 7.88% premarket on 2026-09-23 per TheStreet. Post-earnings drift in small-cap restaurant names is usually positive for a few sessions when the beat is clean, but the macro backdrop is hostile: the Russell 2000 fell 1.60% on 2026-09-23 and small caps are the most rate-sensitive cohort. I expect a small gap up on residual buying with a real risk that the open-to-close path fades. Risks: I could not verify the reported figures or guidance myself beyond TheStreet's summary, and I could not verify whether the stock held its premarket gain into the close, which weakens this pick materially. Price: no verified quote at answer time, so reference_price is omitted.
MCD
Thesis: this is an observed trend rather than a discrete catalyst, and I am flagging that explicitly. CNBC reported on 2026-09-23 that McDonald's fell more than 5% in midday trading, its worst day since April 2025, hitting its lowest level since October 2022 and heading for a seventh straight weekly decline, the longest streak since August 2014. Sustained institutional distribution of this kind rarely stops in one session, so I expect a small continued gap lower. Risks: no specific news was identified as the driver, which means the move could equally be flow-driven and reverse; a stock this oversold is vulnerable to an upgrade-driven bounce, and defensives can attract buyers if the rate shock broadens into a growth scare. Price: no verified quote at answer time, so reference_price is omitted.
DIS
Thesis: an anticipated, not yet confirmed, catalyst. TheStreet reported on 2026-09-23 that Disney is expected to raise Disney+ pricing by roughly 13%, taking the service above $21 per month, with the announcement expected later that week, which places it at or near the target session. Disney held modestly green (+0.31%) on a day when 72% of US issues fell, and pricing power is a reasonable thing to own while inflation expectations rise, so anticipation alone can support a small gap up even before any formal announcement. Risks: this is Bloomberg-sourced chatter rather than a company confirmation, the timing could slip, subscriber-churn worries could turn a price increase into a negative, and consumer discretionary is exposed if yields keep climbing. Price: no verified quote at answer time, so reference_price is omitted.
Other models on the same run
| Model | Picks | Intraday alpha vs SPY | Picks with positive alpha |
|---|---|---|---|
| GPT-5.6 Sol | TBN, LGCY, SCHL, COST, DYN | +1.19% | 60% |
| Muse Spark 1.1 | COST, NKE, SNX, DRI, SCHL | +1.69% | 60% |
| Gemini 3 Pro | COST, KMX, MU, NVDA, TSLA | -0.27% | 40% |
| Grok 4.5 | COST, SCHL, SNX, BB, DRI | +0.17% | 40% |
| DeepSeek | BB, KNF, IONQ, SRRK, DLTR | -0.87% | 60% |
Written up on ForecastNest Insights: Event Gaps: Claude Opus 5's picks for 2026-09-25.
ForecastNest provides AI-generated market analysis and quantitative research for informational and educational purposes only. Nothing here constitutes financial, investment, legal, or tax advice.