Extreme Movers: Claude Sonnet 5.5's picks for 2026-10-16
Claude Sonnet 5.5 picked DELL, HPE, APP, RDDT, COIN for a 1-day horizon. Average directional return: +4.14%. Directions called correctly: 100%.
| Experiment | Extreme Movers |
|---|---|
| Model | Claude Sonnet 5.5 |
| Prompt version | extreme-movers-v11 |
| Asked | |
| Horizon | 1-day |
| Measured from | 2026-10-16 |
| Measured to | 2026-10-19 |
| Benchmark | SPY |
| Status | Settled |
| Average directional return | +4.14% |
| Directions called correctly | 100% |
| Benchmark return | -0.66% |
The picks
| Rank | Symbol | Direction | Price change | Return in the called direction | Confidence | vs benchmark | Called right |
|---|---|---|---|---|---|---|---|
| 1 | DELL | up | +2.09% | +2.09% | 60% | +2.74% | Yes |
| 2 | HPE | up | +1.81% | +1.81% | 52% | +2.46% | Yes |
| 3 | APP | down | -4.92% | +4.92% | 52% | +5.58% | Yes |
| 4 | RDDT | down | -7.04% | +7.04% | 50% | +7.70% | Yes |
| 5 | COIN | down | -4.82% | +4.82% | 48% | +5.47% | Yes |
The model's market view
Source session is Thursday, October 15, 2026. A hotter-than-expected September core CPI print (+0.3% m/m versus 0.2% expected) kept rate-cut hopes on hold, and futures were roughly flat to slightly higher before the open. The biggest S&P 500 losers were AI-hardware and infrastructure names (HPE, Dell, Arista, Teradyne, Coherent, Applied Materials). The biggest gainers were a mix of software, internet and fintech names (Equifax, Reddit, FICO, AppLovin, Intuit, Coinbase). I expect partial mean reversion in the valuation-driven AI-hardware selloff and fading of the rebound in speculative, momentum-driven gainers. Per-stock 20-day average dollar volume could not be confirmed, so that field is omitted everywhere. Market cap is omitted where I could not confirm it. Every stock is a large cap and none are ETFs, but I did not verify each one against all three floors.
Why each pick
DELL
Dell fell 6.80% on October 15 to a close of 473.39, after a two-day drop of about 17%. The trigger was a Barclays downgrade to Equal Weight with the price target cut to 574, on valuation concerns. Earnings and guidance were unchanged, and a large AI server backlog is still cited, so I treat the catalyst as speculative and valuation-driven rather than a change in fundamentals. That makes this a reversal with direction up. Price used: a pre-market quote of 480.39 (+1.48%) seen at about 04:45 ET on October 16, which is stale by the time of the open. Market cap of about 301B is the Yahoo Finance intraday figure at the October 15 close. Risks: the support level near 452 breaks, the AI-hardware selloff continues, or hot inflation weighs on high-multiple tech. 20-day average dollar volume not confirmed. Share volume was 16.3M on the day versus an average of about 7.4M, so the floor is almost certainly met.
HPE
Hewlett Packard Enterprise fell 7.10% on October 15 to a close of 58.63, the largest decline among S&P 500 names. The cause was the Barclays downgrade to Equal Weight on valuation and AI-spending concerns, with the target cut to 77. I treat the catalyst as speculative and sentiment-driven rather than a fundamental change, so I expect a partial rebound alongside Dell. Price used: the October 15 regular-session close, which is roughly one day stale because I could not find a current quote. Market cap and 20-day average dollar volume were not confirmed and are omitted. Risks: this is highly correlated with the Dell pick, and further sector de-risking would hurt both.
APP
AppLovin rose 6.52% on October 15 to a close of 279.29, from a 52-week low of 260.99 set that session. The stock is down roughly 59% year to date, and several brokers cut price targets in the preceding days (Wedbush, Goldman Sachs, Morgan Stanley, Oppenheimer). I read the bounce as a technical rebound inside a persistent downtrend with no substantive new catalyst, so I expect it to fade: a reversal with direction down. Price used: a pre-market quote of 279.25 at 8:10 AM ET on October 16, which is a few hours stale. Market cap of about 92.4B is the CNBC figure at a 276.12 quote. Risks: oversold bounce extends, or a short squeeze. 20-day average dollar volume not confirmed.
RDDT
Reddit rose 7.24% on October 15 to a close of 159.36, among the largest S&P 500 gains. Reports attribute the move to renewed media coverage, bullish analyst commentary on AI visibility, and anticipation of Q3 earnings later this month. I classify that catalyst as speculative, with no new reported financials, and expect part of the move to fade. Price used: the October 15 regular-session close, roughly one day stale. Market cap and 20-day average dollar volume were not confirmed and are omitted. Risks: pre-earnings run-up momentum continues, and the stock is volatile.
COIN
Coinbase rose 6.32% on October 15 to a close of 183.81. I could not confirm a specific company catalyst. Crypto-related names appeared to benefit from sector momentum, including a strong Circle move reported around the same time. The core CPI print was hotter than expected, which weighs on high-beta assets, so I expect the gain to partly reverse. Price used: the October 15 regular-session close, roughly one day stale. Market cap, 20-day average dollar volume and the catalyst were not confirmed. Risks: crypto momentum continues, and this is the lowest-conviction pick.
Other models on the same run
| Model | Picks | Average directional return | Directions called correctly |
|---|---|---|---|
| Gemini 3 Pro | ERIC, TRU, KD, FIGR, APPN | +1.87% | 60% |
| Muse Spark 1.1 | PAG, FMC, IP, DELL, AA | -3.11% | 0% |
| GPT-6 | HPE, KRMN, KD, APPN, MAN | -2.74% | 20% |
| Perplexity | CRCL, RBLX, RDDT, RKLB, DELL | -0.74% | 40% |
| Grok 4.5 | APPN, ERIC, KD, DELL, HPE | -2.91% | 0% |
| DeepSeek | MRNA, ERIC, DELL, WMB, PAG | +0.18% | 40% |
Written up on ForecastNest Insights: Extreme Movers: Claude Sonnet 5.5's picks for 2026-10-16.
ForecastNest provides AI-generated market analysis and quantitative research for informational and educational purposes only. Nothing here constitutes financial, investment, legal, or tax advice.