Event Gaps: Claude Opus 5's picks for 2026-09-10
Claude Opus 5 picked AVAV, NAVN, AEO, M, COO for a 1-day horizon. Intraday alpha vs SPY: -0.08%. Picks with positive alpha: 60%.
| Experiment | Event Gaps |
|---|---|
| Model | Claude Opus 5 |
| Prompt version | event-gap-v7 |
| Asked | |
| Horizon | 1-day |
| Measured from | 2026-09-09 |
| Measured to | 2026-09-10 |
| Benchmark | SPY |
| Status | Settled |
| Intraday alpha vs SPY | -0.08% |
| Picks with positive alpha | 60% |
| Benchmark return | -0.03% |
The picks
| Rank | Symbol | Gap direction | Price change | Return in the called direction | Confidence | vs benchmark | Called right |
|---|---|---|---|---|---|---|---|
| 1 | AVAV | up | +0.20% | +0.20% | 56% | +0.23% | Yes |
| 2 | NAVN | down | -5.28% | -5.28% | 50% | -5.26% | No |
| 3 | AEO | down | -1.09% | -1.09% | 47% | -1.06% | No |
| 4 | M | down | +1.23% | +1.23% | 43% | +1.26% | Yes |
| 5 | COO | down | +4.41% | +4.41% | 42% | +4.44% | Yes |
The model's market view
Wednesday September 9 produced a dense after-close earnings slate, and four names carry verified extended-hours dislocations into the September 10 open, with a fifth reporting before that open. Every after-hours figure below was checked individually against quote data captured near the 8:00pm EDT close of extended trading rather than taken from an aggregated mover list. The backdrop matters more than usual: SPY closed at $762.40, down 0.46%, Brent crossed $100 for the first time since late July at $100.80, and Treasury yields set fresh 52-week highs after US forces struck five Iranian tankers and Iran fired ballistic missiles at a US Navy warship. Critically, August PPI and weekly jobless claims are officially scheduled for 8:30am ET on September 10, an hour before the open — meaning the opening prints will already embed the macro reaction, so my gap forecasts are deliberately not straight copies of the after-hours marks. Deep after-hours dislocations usually compress somewhat as liquidity returns at the open, and a hot producer-price reading with crude above $100 would compress the up-gaps and widen the down-gaps. On the scoring: this is an equal-weight long portfolio measured open-to-close, so the four gap-down selections are bets on intraday stabilisation after the opening print, not short positions. That distinction drives the ranking — I rank highest the names whose reported fundamentals were good, because a gap driven by valuation or positioning rather than by deteriorating results is the one that tends to find a bid during the session, while guidance cuts and removed strategic optionality carry genuine emotional continuation risk into a second and third day. Oracle and Adobe report after Thursday's close, so they cannot gap at the open and are excluded.
Why each pick
AVAV
Thesis: this is the only clean up-gap in the group and the only one where the reported facts exceeded expectations without a caveat. AeroVironment beat fiscal Q1 2027 estimates on revenue of $480.5 million with backlog reaching a record $1.5 billion and reported directed-energy programme wins. The stock had fallen 5.36% during the regular session to close at $140.80, so the after-hours move to $146.40 is a recovery from a depressed base rather than an extension of an already-strong day, which leaves more room for continuation than a stock gapping up from a high. It is also the one name here with macro tailwind rather than headwind: US forces struck five Iranian tankers on September 9 and Iran fired ballistic missiles at a US Navy warship, and unmanned systems and loitering munitions are exactly the procurement category such an environment accelerates. Risks: coverage noted the record backlog masks a margin split between segments, and margin questions have been punished severely in this tape; defence stocks did not respond to Wednesday's escalation at all, with RTX actually falling 0.63%, so the geopolitical tailwind is a hypothesis rather than an observed sector behaviour; and at a $7.10 billion market capitalisation the name is volatile enough to give back a 4% gap within an hour.
NAVN
Thesis: the cleanest fundamentals-intact dislocation available. Navan reported second quarter results that were stronger than expected on both the quarter and the guidance, and the stock fell anyway — Investing.com recorded an 11% decline shortly after the print, and by 5:54pm EDT the extended-hours mark had widened to $21.58, down 16.64% from the $25.89 close. When a company beats on results and raises the bar and the stock still sells off, the cause is positioning, valuation or float dynamics rather than business deterioration, and that is the profile most likely to attract buyers once the opening liquidation clears. There is no negative fundamental datapoint for the market to keep discounting through the session. Risks: this is a recent IPO with a $6.58 billion market capitalisation and a limited float, and newly listed names frequently bleed for multiple sessions on lockup and index-inclusion mechanics that have nothing to do with the quarter; the after-hours decline widened rather than stabilised through the evening, from roughly 11% to nearly 17%, which is the opposite of the pattern I am relying on; and a high-multiple travel-software name is exposed to both the rate backdrop and the fuel-cost implications of crude above $100.
AEO
Thesis: American Eagle reported second quarter fiscal 2026 results after the close with an EPS beat and growth driven by the Aerie brand, but the shares fell on the outlook, marking $15.14 in extended hours at 7:59pm EDT against a $16.89 close. The open-to-close case is that a roughly 10% repricing of a $2.83 billion specialty retailer on guidance tone, when the quarter itself beat and the core growth brand is performing, tends to overshoot and partially retrace once the transcript circulates. Risks: this is a materially lower-quality beat than it first appears, because coverage attributed the EPS upside to tariff refunds rather than operating performance — a one-time item that does not recur and that analysts will strip out of forward models overnight. The consumer backdrop is also actively deteriorating with Brent above $100 feeding into gasoline prices, which is a direct tax on the mall-based teen retail customer, and Macy's fell 4.19% on Wednesday in sympathy positioning. Overnight price-target cuts landing before the open would deepen rather than repair this gap.
M
Thesis and disclosure: this is the one selection whose catalyst has NOT occurred at prediction time. Macy's is officially scheduled to report before the September 10 open, so a sharp gap is near-certain by construction, but the direction is an uncertain expectation reasoned from evidence rather than an observed fact, and I am stating that plainly rather than implying I know the result. The down-gap reasoning: American Eagle fell over 10% in extended hours on its outlook hours earlier, establishing a negative read-through for mall and department-store retail; Macy's itself fell 4.19% to $21.51 during Wednesday's regular session, which reads as pre-positioning ahead of the print; and Brent above $100 is a direct squeeze on discretionary retail spending. The open-to-close case, which is what is actually scored, is that Macy's has a long history of extreme opening reactions that mean-revert substantially during the session as the initial print is digested. Risks: the direction call is a coin flip dressed in a read-through, and if Macy's beats the gap goes the other way entirely; at a $5.66 billion market capitalisation with high short interest the name can squeeze violently in either direction; and PPI at 8:30am ET lands between now and the report's absorption, adding an unrelated macro layer to the opening print.
COO
Thesis: Cooper Companies reported fiscal third quarter 2026 revenue of $1.07 billion and fell hard on two things at once — weak forward guidance and the conclusion of its strategic review. Extended hours marked it at $53.35, down 15.96%, at 7:59pm EDT against a $63.48 close, on top of a 6.22% regular-session loss, so the stock has now lost roughly a fifth of its value across the two moves. The long case for the session is purely mechanical: a 16% repricing of a $12.38 billion medical device company is large enough that value and index-tracking flows typically appear at the open, and healthcare has been an area of forced rotation as investors flee rate-sensitive and oil-exposed sectors. Risks: this is ranked last because the catalyst is the most structurally negative of the five. Concluding a strategic review without a transaction removes an M&A premium that some holders were explicitly paid to own, and that overhang does not return — it is permanent, not sentiment. Guidance cuts in medtech also invite multi-session downgrade cycles, and the sector already has open wounds this week from Stryker's supply-chain disclosure and Boston Scientific's cyberattack-driven guidance miss, so sector sponsorship is weak precisely when this name needs a bid.
Other models on the same run
| Model | Picks | Intraday alpha vs SPY | Picks with positive alpha |
|---|---|---|---|
| GPT-5.6 Sol | COO, AVAV, SKIL, M, DBI | -1.97% | 80% |
| Muse Spark 1.1 | AVAV, COO, ORCL, ADBE, CAL | -0.04% | 40% |
| Gemini 1.5 Pro | COO, IQV, BAX, ADBE, CPRT | +0.18% | 60% |
| Grok 4.5 | COO, AEO, NAVN, AVAV, CHWY | +0.33% | 60% |
| Gemini 3 Pro | ASO, NET, COO, NAVN, AVAV | +1.38% | 60% |
Written up on ForecastNest Insights: Event Gaps: Claude Opus 5's picks for 2026-09-10.
ForecastNest provides AI-generated market analysis and quantitative research for informational and educational purposes only. Nothing here constitutes financial, investment, legal, or tax advice.