Event Gaps: Claude Sonnet 5.5's picks for 2026-09-30
Claude Sonnet 5.5 picked MU, ACN, MKC, AYI, NKE for a 1-day horizon. Intraday alpha vs SPY: -6.06%. Picks with positive alpha: 20%.
| Experiment | Event Gaps |
|---|---|
| Model | Claude Sonnet 5.5 |
| Prompt version | event-gap-v12 |
| Asked | |
| Horizon | 1-day |
| Measured from | 2026-09-30 |
| Measured to | 2026-10-01 |
| Benchmark | SPY |
| Status | Settled |
| Intraday alpha vs SPY | -6.06% |
| Picks with positive alpha | 20% |
| Benchmark return | -0.96% |
The picks
| Rank | Symbol | Gap direction | Price change | Return in the called direction | Confidence | vs benchmark | Called right |
|---|---|---|---|---|---|---|---|
| 1 | MU | -3.83% | -3.83% | 55% | -2.87% | No | |
| 2 | ACN | +24.31% | -24.31% | 54% | -23.35% | No | |
| 3 | MKC | -3.22% | -3.22% | 52% | -2.26% | No | |
| 4 | AYI | -3.74% | -3.74% | 52% | -2.78% | No | |
| 5 | NKE | -0.01% | -0.01% | 45% | +0.95% | Yes |
The model's market view
Cautious, event-driven tape. Yields are elevated (30-year at its highest since 2002 per one market brief), oil is firm, and Friday's payrolls are still ahead. Micron's print after the close on Sep 30 is the main event and sets the read-through for AI and memory stocks. Accenture, McCormick and Acuity report before the open on Oct 1, all inside the window. Nike reports after the window ends, so only anticipation counts for it. Direction calls are low-conviction because each stock is already priced for a wide range of outcomes. Where I could not verify a figure, the rationale says so.
Why each pick
MU
Micron reports fiscal Q4 after the close on Sep 30 (call at 4:30 p.m. ET), so the full reaction falls inside the window. Consensus is roughly $51.2-51.5B revenue and about $31.5-31.8 EPS, against company guidance of $50B +/- $1B revenue, about 86% gross margin and $31 +/- $1 EPS. Micron has beaten EPS estimates for many straight quarters, HBM is described as booked through 2027, and free cash flow and buyback commentary could add upside. The stock trades near 6.7x next-year earnings. Guidance for the next quarter matters more than the Q4 beat. Risks: the stock has rallied about 17% off its Sep 16 low, expectations are very high, and it has sold off after prior record prints. Options implied about +/-11% last quarter, and recent reactions have ranged from -3.8% to +15.7%. This is a scheduled event (occurred: no; officially scheduled: yes). Anticipation is already lifting memory names, so the pre-report drift is partly priced. I lean up because next-quarter guidance likely exceeds the Street. Reference price is an intraday snapshot from Sep 30 at $1,072.00. Market cap of about $1.19T comes from Investing.com. I did not find a verified 20-day average dollar volume; the figure given is a rough unverified estimate, and MU is far above the $20M floor at any plausible volume.
ACN
Accenture reports fiscal Q4 and full-year 2026 before the open on Oct 1 (about 6:39 a.m. ET), inside the window. Consensus is about $3.18-3.19 EPS on roughly $18.0B revenue. Options imply about an 8% move. The key items are the first fiscal 2027 outlook, bookings, and whether federal business returns to growth. Last quarter brought an earnings miss, lower revenue guidance, weaker bookings and large deals pushed into fiscal 2027. About $9B of acquisitions also make organic growth look thin (roughly 1.5-2.5% by one estimate). The stock closed at $177.12 on Sep 29 after rebounding about 38% from its June low, so a soft or acquisition-driven fiscal 2027 guide could reverse part of that. I lean down modestly on that basis. Upside risk: expectations are depressed, the stock is down about 39% from its January high, and margin and AI-services commentary could squeeze shorts. Scheduled and not yet observed. Reference price is the Sep 29 regular-session close; I found no Sep 30 quote. Market cap is about 611.94M shares times $177.12, which is my own calculation. The 20-day average dollar volume is an unverified estimate from older short-interest and days-to-cover data.
MKC
McCormick reports fiscal Q3 before the open on Oct 1 (call at 8:00 a.m. ET), inside the window. Consensus is about $0.75-0.76 EPS, down from $0.85, on about $1.98B revenue. Estimates have drifted down roughly 9.5% over three months, and TD Cowen downgraded to Hold on Sep 14. The stock is down about 26% over 52 weeks and trades around $48-49, so a lot of caution is priced in. McCormick beat EPS in three of the last four quarters, and the Unilever food-deal review adds noise but is unlikely to resolve within the window. I lean modestly up on a relief reaction to a beatable bar. Risks: Zacks shows a negative earnings-surprise indicator, margin and tariff pressure could drive a guidance cut, and the tape is sensitive to yields. Scheduled and not yet observed. Reference price of $48.15 comes from a Sep 29-30 preview quote, and I could not confirm its exact timestamp. Market cap of $12.88B comes from a Sep 26 data provider report. The 20-day average dollar volume is an unverified estimate.
AYI
Acuity reports fiscal Q4 before the open on Oct 1, inside the window (the exact release time is not confirmed; the June report came at 6:00 a.m. ET). Consensus is about $5.66 EPS on $1.25B revenue, versus $5.20 a year ago, and the company has beaten EPS estimates in recent quarters. The stock gapped down on Sep 30 after William Blair moved to market perform and Baird cut its target to $360, which lowers the bar into the print. That gap is already in the price and is not part of my forecast. I lean up on a beat plus fiscal 2027 commentary on lighting demand, margins and the intelligent spaces segment. Risks: demand and backlog commentary could disappoint, downgrades signal reduced conviction, and lighting sales were guided flat to down. Scheduled and not yet observed. The $319.42 reference price comes from a pre-earnings preview dated about Sep 29 and does not reflect the Sep 30 gap down; a market cap of $9.23B implies roughly $310, so the true current price is likely lower. Market cap comes from a Sep 30 article. The 20-day average dollar volume is an unverified estimate from an older average volume of roughly 500,000 shares.
NKE
Nike reports fiscal Q1 2027 after the close on Oct 1 (call at 5 p.m. ET). That is after this window ends at 10:28 a.m. ET on Oct 1, so this pick rests on anticipation only. The stock is near a roughly 12-year low, down about 42% year to date, and short interest is reported at a record of about 87M shares. Bank of America downgraded it to Sell on a $30 target, and consensus is about $0.44 EPS on roughly $11.3B revenue, down year over year. The options market implies about an 8% post-earnings move, but that reaction happens after the window closes. I lean slightly up because heavy short positioning and depressed expectations can produce a modest run-up into a heavily watched print. Risks: continued negative analyst revisions, reports of retailers cutting spring 2027 orders, and a weak macro tape could push it lower before the report. The catalyst is scheduled and not yet observed, and its outcome is uncertain. The $35.85 reference price is an undated Yahoo Finance snapshot, so its date is unconfirmed. Market cap is my rough estimate of about $53B from an assumed share count of roughly 1.48B, not a verified figure. The 20-day average dollar volume is an unverified estimate.
Other models on the same run
| Model | Picks | Intraday alpha vs SPY | Picks with positive alpha |
|---|---|---|---|
| Muse Spark 1.1 | MU, NKE, CAG, STZ, JEF | +1.67% | 80% |
| Gemini 1.5 Pro | MU, JBL, FDS, CAG, CCL | -1.26% | 20% |
| Perplexity | MU, NKE, ACN, CAG, JBL | -4.12% | 40% |
| Grok 4.5 | MU, BA, ACN, NKE, JBL | +5.10% | 60% |
| DeepSeek | MU, ACN, AYI, CTVA, MKC | -5.90% | 25% |
| GPT-6 | MU, CALM, HPE, JBL, BA | -0.84% | 20% |
Written up on ForecastNest Insights: Event Gaps: Claude Sonnet 5.5's picks for 2026-09-30.
ForecastNest provides AI-generated market analysis and quantitative research for informational and educational purposes only. Nothing here constitutes financial, investment, legal, or tax advice.