Extreme Movers: Claude Opus 5's picks for 2026-09-10
Claude Opus 5 picked COO, AVAV, SIG, CASY, BKNG for a 1-day horizon. Average directional return: +3.04%. Directions called correctly: 80%.
| Experiment | Extreme Movers |
|---|---|
| Model | Claude Opus 5 |
| Prompt version | extreme-movers-v6 |
| Asked | |
| Horizon | 1-day |
| Measured from | 2026-09-09 |
| Measured to | 2026-09-10 |
| Benchmark | SPY |
| Status | Settled |
| Average directional return | +3.04% |
| Directions called correctly | 80% |
| Benchmark return | -0.60% |
The picks
| Rank | Symbol | Direction | Price change | Return in the called direction | Confidence | vs benchmark | Called right |
|---|---|---|---|---|---|---|---|
| 1 | COO | down | -14.67% | +14.67% | 68% | +15.27% | Yes |
| 2 | AVAV | up | +4.45% | +4.45% | 62% | +5.05% | Yes |
| 3 | SIG | up | -4.65% | -4.65% | 52% | -4.06% | No |
| 4 | CASY | down | -0.22% | +0.22% | 50% | +0.82% | Yes |
| 5 | BKNG | up | +0.52% | +0.52% | 47% | +1.12% | Yes |
The model's market view
A method note first, because it shapes the confidence numbers. Aggregator screener pages I reached were serving stale data, and news-derived mover lists proved unreliable when checked — one source reported Marvell up 6.4% and Meta up 5.4% on September 9 when the verified closes were +4.26% and +6.55% respectively. Every figure below was therefore verified individually against quote data for the September 9 close, and candidates I could not confirm were dropped rather than filled in. All five selections clear the exclusion rule comfortably on both price and market capitalisation. Two of the five calls rest on evidence that is already observable rather than on inference, and I want that disclosed plainly: Cooper Companies and AeroVironment both reported after Wednesday's close and their extended-hours marks sit far from their regular-session closes, which for a close-to-close scoring window is close to mechanical information rather than forecasting. The other three are genuine judgement calls. Session context: SPY closed at $762.40, down 0.46%, with the Russell 2000 down 1.24%, as Brent crossed $100 for the first time since late July at $100.80 and Treasury yields set fresh 52-week highs, following US strikes on five Iranian tankers and an Iranian ballistic missile attack on a US Navy warship. The overriding risk to all five is macro rather than idiosyncratic: August PPI and weekly jobless claims land at 8:30am ET on September 10, and with crude above $100 the producer-price print is exactly where an energy shock surfaces first. A hot reading would compress the up-calls and deepen the down-calls regardless of company specifics.
Why each pick
COO
Ranked first because it is the least speculative call available. Cooper Companies fell 6.22% during the September 9 regular session to close at $63.48, then reported fiscal third quarter results after the close — revenue of roughly $1.07 billion — and collapsed in extended hours to $53.35, down 15.96%, marked at 7:59pm EDT. Because the scoring window runs from the September 9 close to the September 10 close, that extended-hours mark is already-observable evidence rather than a forecast, and closing back above $63.48 would require recovering the entire 16% loss within one session. The catalyst is substantive and structural rather than sentimental: weak forward guidance combined with the formal conclusion of the company's strategic review, which permanently removes an acquisition premium that some holders were specifically paid to own. Risks: extended-hours prices trade on thin volume and deep after-hours dislocations routinely compress by 20-30% of their magnitude at the open; a heavily oversold medical device name can attract value and index-tracking flows during the session; and healthcare is a defensive sector that catches haven bids if a hot PPI print drives investors out of cyclicals.
AVAV
AeroVironment fell 5.36% during the regular session to close at $140.80, then beat fiscal first quarter 2027 estimates after the close on revenue of $480.5 million with backlog reaching a record $1.5 billion and directed-energy programme wins, lifting it to $146.40 in extended hours at 7:53pm EDT. Like the Cooper call, this rests on already-observable evidence rather than pure inference — the reversal is against the September 9 decline and the after-hours mark already sits about 4% above the reference close. The catalyst is substantive: record backlog is contracted future revenue, not a narrative. The setup is also favourable because the stock sold off into the print, meaning the recovery starts from a depressed base rather than extending an already-strong day, and unmanned systems procurement is precisely what an environment featuring US strikes on Iranian tankers and an Iranian missile attack on a US Navy warship accelerates. Risks: coverage noted the record backlog masks a margin split between segments, and this tape has punished margin questions severely; the geopolitical tailwind is a hypothesis rather than an observed behaviour, since defence did not respond to Wednesday's escalation at all, with RTX actually falling 0.63%; and a $7.10 billion name can surrender a 4% after-hours gain within an hour of the open.
SIG
Signet Jewelers was the largest verified gainer of the session, rising 23.96% to close at $102.48 after second quarter results that exceeded expectations alongside an upward revision to full-year guidance. The catalyst is substantive on both counts — a beat establishes the quarter, and a raised outlook establishes the trajectory, which is what analysts must rebuild models around. Post-earnings announcement drift is one of the better-documented short-horizon effects, and it is strongest exactly where a beat is accompanied by a guidance raise, because price-target increases arrive over several days rather than in one afternoon. At a $4.03 billion market capitalisation the name is small enough for that revision flow to move price meaningfully. Risks: a 24% single-session gain is an enormous move that invites immediate profit-taking, and the second day after such a spike is frequently negative; specialty retail faces a genuine macro headwind with Brent above $100 squeezing discretionary spending, and American Eagle fell over 10% in extended hours on its own outlook hours later; and a hot PPI print would hit consumer discretionary names hardest.
CASY
Casey's General Stores fell 14.24% to close at $629.03 on September 9, and the important detail is that this was the second leg of a three-stage decline rather than a single shock: the stock fell 2.99% during the September 8 session, dropped a further 8.31% in extended hours after its first quarter report, and then lost another 14.24% on Wednesday as multiple analysts lowered price targets despite the company beating earnings expectations. That pattern — a beat that the sell-side nonetheless downgrades on outlook grounds — typically produces a multi-session downgrade cycle, because not every covering analyst publishes on the same day. The catalyst is substantive rather than speculative, since price-target reductions permanently reset the valuation anchor. Risks: after cumulative losses exceeding 23% across three sessions, the name is deeply oversold and prone to a sharp technical bounce; convenience-store operators actually benefit from higher fuel margins when crude spikes, which is a live tailwind with Brent above $100 and cuts directly against this call; and a defensive retailer can attract rotation flows if a hot PPI print drives money out of cyclicals.
BKNG
Booking Holdings fell 3.81% to close at $173.43 after an EU court upheld the antitrust veto of its eTraveli acquisition, blocking a roughly 1.63 billion euro deal that regulators had already rejected three years earlier. The reversal case rests on the nature of the catalyst: this is a resolved regulatory event rather than an ongoing operational problem. The deal is definitively dead, the uncertainty premium is removed, and the purchase consideration stays on the balance sheet — the company is not losing revenue or customers, it is losing an option it had already been denied once. A 3.81% single-day repricing of a $130 billion company for confirming an outcome the market had largely assumed looks like an overreaction, and analyst consensus remains Buy with an average target of $238.68 against the current close. Risks: this is the lowest-conviction call. Travel is squarely exposed to Brent above $100 through fuel surcharges and discretionary spending compression, which is a fundamental headwind that has nothing to do with the court ruling and may be the real reason for the decline; a Buy consensus with a target 38% above spot may simply be stale; and the reversal thesis rests on my judgement about market overreaction rather than on any observable post-close evidence.
Other models on the same run
| Model | Picks | Average directional return | Directions called correctly |
|---|---|---|---|
| GPT-5.6 Sol | TTAN, BRZE, CASY, META, VRT | -1.39% | 20% |
| Muse Spark 1.1 | AVAV, COO, CAL, ADBE, ORCL | -2.20% | 60% |
| Grok 4.5 | NET, META, TTAN, CASY, SIG | +0.02% | 20% |
Written up on ForecastNest Insights: Extreme Movers: 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.