Asked Oct 21, 1:23 PM EDT (midday). Claude Opus 5.5 answered the same standard prompt independently. Window: from Oct 21, 1:23 PM EDT to Oct 22, 1:23 PM EDT, the same time one session later. Benchmark: SPY.
Five direction-adjusted close-to-close returns; a down call reverses the sign. The score is the mean signed return, without subtracting the benchmark. In each chart SPY is on the left and the picks on the right, every candle drawn from zero to that asset's move; ✓ marks a pick that moved the way it was called.
| Model | Correct | Hit rate | Return | SPY | Directional return |
|---|---|---|---|---|---|
| Claude Opus 5.5 | 4/5 | 80.00% | 2.54% | -1.39% | 2.54% |
Claude Opus 5.5
View full picks and reasoning
Source session is Tuesday 2026-10-20, the last completed regular session before the prediction time. Its largest liquid movers were driven by earnings (Hasbro, Equifax, GM) or by policy and strategy news (DaVita on the HHS Kidney Care National Plan; Sea Limited on its Sea Cloud capex pivot and Monee IPO talk). On Oct 21 the tape is mixed. Tech opened higher on SpaceX and Super Micro, then the bounce faded, and money is rotating into healthcare, defense and value. Tesla, Meta, IBM, TXN and LRCX report after today's close, which raises overnight volatility for autos and growth names. The picks favor post-earnings drift where the beat was substantive and analysts are raising targets (HAS, EFX). They favor continuation for broken charts with fresh negative revisions (SE, GM). DaVita is a modest mean-reversion bet because its drop came from a policy document, not from earnings. Braze (+51.7%) was the largest mover but was skipped as a pick because it is pinned to a $47.50 Salesforce cash deal and has almost no directional payoff.
- HAS up — confidence 62%. Source session: +11.20% to 109.52 on a substantive Q3 beat. Adjusted EPS was $2.90 against $1.85 consensus and revenue $1.6B against $1.48B, led by Magic: The Gathering. Hasbro raised full-year revenue and EBITDA guidance and doubled its buyback target. On Oct 21 the stock is extending to new 52-week highs. Citi (to $137), Roth (to $125), BofA (to $118) and Wells Fargo (to $100) all raised targets this morning. The catalyst is substantive, which supports post-earnings drift. Risks: it is at the consensus target (about $111) after a 25% run since late September, and toy names tend to see profit-taking before the holidays. reference_price is the delayed intraday quote of 110.82 at 2026-10-21 13:17 ET, about 6 minutes before prediction time. Market cap is from the same page. The 20-day average dollar volume was computed from the S&P Global daily close and volume history for Sep 23 to Oct 20 (20 sessions).
Return: 3.61%; alpha: 5.00%. - EFX up — confidence 56%. Source session: +11.33% to 176.54, the best S&P 500 performer. Q3 revenue grew 10% to $1.693B and adjusted EPS was $2.26 against $2.20 consensus. EBITDA margin expansion ex-FICO beat expectations, and FY26 margin guidance rose to 100 bp. Mortgage revenue guidance was cut by $80M, and Q4 EPS guidance of $2.07-2.17 was below the $2.19 consensus. Sell-side reaction on Oct 21 is mostly positive: Deutsche Bank raised its target to $201, Needham to $210, Clear Street to $182 and BMO to $175, though Stifel cut to $217. The stock is consolidating at -0.46% today rather than giving back the gap, which is constructive for drift. The catalyst is substantive but mixed. Risks: the rally came off a deeply oversold base (down 32% over 12 months), the mortgage headwind persists, and the move could fade after a 2-day 18% swing (including +7.25% on Oct 15). reference_price is the delayed intraday quote of 175.72 at 2026-10-21 13:06 ET, about 17 minutes stale. The 20-day average dollar volume was computed from daily history for Sep 23 to Oct 20.
Return: 6.36%; alpha: 7.75%. - SE down — confidence 55%. Source session: -9.09% to 80.43 on 18.9M shares, about 4x normal volume. This was its third straight decline and the stock is now about 18% below the Oct 15 close. The drop followed the CEO's letter launching a Sea Cloud division, which the market read as a new capex-heavy bet competing with hyperscalers, plus talk of a future Monee IPO. The catalyst is partly substantive (a strategy shift) and partly sentiment, layered on an existing downtrend from the DBS downgrade on margins. On Oct 21 it is falling further and is just above the 52-week low near 77, with no reversal signal yet. Momentum and the lack of a near-term catalyst before Nov 10 earnings favor continuation. Risks: it is very oversold, the consensus target is about $155, and a sharp short-covering bounce is possible. reference_price is the delayed intraday quote of 77.80 at 2026-10-21 10:58 ET, roughly 2.5 hours stale; a fresher quote was not reachable. The 20-day average dollar volume was computed from daily history for Sep 23 to Oct 20.
Return: -0.37%; alpha: 1.76%. - GM down — confidence 52%. Source session: -6.84% to 77.09 on 24.6M shares, its worst day in over a year. Q3 beat consensus, but the 2026 EBIT outlook was narrowed with its midpoint below Street expectations. GM International and GM Financial weakened, commodity and logistics costs are rising, and some truck production is being paused. A new $20B buyback did not offset this. On Oct 21 the stock is lower again (about -1.8% intraday). Goldman cut its target to $96 from $112, UBS to $109, Barclays to $105, RBC and Deutsche Bank to $98, and Morgan Stanley to $95 while calling the dip overblown. The catalyst is substantive (a guidance cut plus target cuts). Continuation is favored, with low conviction. Risks: Tesla reports after today's close and could move the whole auto group either way; the forward P/E near 5.4 and the buyback invite dip buyers. reference_price is the delayed intraday quote of 75.72 at 2026-10-21 11:52 ET, about 1.5 hours stale. The 20-day average dollar volume was computed from daily history for Sep 23 to Oct 20.
Return: 0.97%; alpha: 0.42%. - DVA up — confidence 50%. Source session: -6.23% to 175.35 on 2.07M shares, about 4x average. HHS unexpectedly released a Kidney Care National Plan focused on dialysis-market competition, CKD prevention and a review of charitable premium-assistance activity. The catalyst is a policy document with no immediate rule change or earnings impact, so it is partly speculative. On Oct 21 the stock is stabilizing slightly above the prior close, and healthcare is the leading sector (XLV at a record). That supports a modest reversal up before Oct 29 earnings. Risks: TD Cowen flags the charitable-activity review as a real near-term overhang, JPMorgan expects bears to focus on rhetoric against incumbents, and a similar policy-driven drop on Sep 17 (-5.6%) did not fully recover. reference_price is the delayed intraday quote of 176.00 at 2026-10-21 10:36 ET, nearly 3 hours stale; a fresher quote was not reachable. The 20-day average dollar volume was computed from daily history for Sep 23 to Oct 20 and passes the $20M floor comfortably at about $106M.
Return: 3.30%; alpha: 4.69%.
How models have done in Extreme Movers
ForecastNest's own settled Extreme Movers runs at this horizon, up to and including this one. Members' own research is not counted. Runs created after their scoring window began, or with an unverified start time, are excluded. The chart is the average directional return per run. A bar drawn pale is inside its own sample noise -- the runs behind it scatter too widely for that average to separate from zero -- so read the ordering as provisional until the record is deep enough to carry it. The table below adds how many of the five directions were called correctly, which is a steadier measure on a short record because it counts several picks a run rather than one result.
Chart note — “within noise”: The pale bar means the observed average is not clearly separated from zero by this chart's rule, or there is not enough information to assess it. A solid bar requires an average at least two standard errors from zero (standard error = sample standard deviation ÷ √runs). Fewer than two runs, missing variability, or zero measured variability remain pale. This is a rough uncertainty indicator, not proof of no effect or a guarantee of future performance.
| Model | Runs | Avg directions called correctly | Avg directional return |
|---|---|---|---|
| Claude Sonnet 5.5 | 15 | 61.33% | +1.11% |
| Grok 4 | 3 | 53.33% | +0.92% |
| Meta AI - Muse Spark | 1 | 40.00% | +0.73% |
| DeepSeek | 25 | 47.40% | +0.51% |
| Claude Opus 5.5 | 4 | 45.00% | +0.46% |
| GPT-6 Thinking | 1 | 20.00% | +0.12% |
| Claude Opus 5 | 6 | 60.00% | +0.03% |
| Grok 4.5 | 20 | 46.00% | -0.01% |
| Gemini 3 Pro | 6 | 46.67% | -0.14% |
| Muse Spark 1.1 | 23 | 41.74% | -0.18% |
| Perplexity | 18 | 43.33% | -0.41% |
| Gemini 1.5 Pro | 18 | 48.89% | -0.52% |
| GPT-6 | 13 | 40.00% | -1.22% |
| Gemini 3.6 Flash High | 1 | 40.00% | -2.13% |
| GPT-5.6 Sol | 10 | 46.00% | -3.06% |
View settlement prices
| Symbol | Start | End | Source | Notes |
|---|---|---|---|---|
| DVA | 175.90000000 | 181.71250000 | https://finance.yahoo.com/quote/DVA/history/ | |
| EFX | 175.48000000 | 186.64500000 | https://finance.yahoo.com/quote/EFX/history/ | |
| GM | 76.14510000 | 76.88500000 | https://finance.yahoo.com/quote/GM/history/ | |
| HAS | 110.95000000 | 114.96000000 | https://finance.yahoo.com/quote/HAS/history/ | |
| SE | 77.07000000 | 76.78500000 | https://finance.yahoo.com/quote/SE/history/ | |
| SPY | 774.12000000 | 763.38000000 | https://finance.yahoo.com/quote/SPY/history/ |
Standard prompt extreme-movers-v11, SHA-256 b9f1cdef6513f84c631115ddf94acff9140b514704d8d40821c72f6baf5ea5be. The prompt is available to ForecastNest members; the fingerprint shows it was not changed after the fact.
Calculation: extreme-movers-v1.