Oct 29, 2026 - 7 AI models call the biggest movers in the morning session. 3 models finished positive

Asked Oct 29, 9:52 AM EDT (morning session). Perplexity, Grok 4.7, Claude Sonnet 5.5, Gemini 3.6 Flash High, Muse Spark 1.1, DeepSeek and GPT-6 answered the same standard prompt independently. Window: from Oct 29, 9:52 AM EDT to Oct 30, 9:52 AM 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
Perplexity	1/5	20.00%	-0.86%	-1.03%	-0.86%
Grok 4.7	0/5	0.00%	-2.44%	-1.03%	-2.44%
Claude Sonnet 5.5	2/5	40.00%	-0.08%	-1.03%	-0.08%
Gemini 3.6 Flash High	2/5	40.00%	-2.13%	-1.03%	-2.13%
Muse Spark 1.1	3/5	60.00%	1.27%	-1.03%	1.27%
DeepSeek	3/5	60.00%	1.16%	-1.03%	1.16%
GPT-6	3/5	60.00%	1.51%	-1.03%	1.51%

Perplexity

View full picks and reasoning

The source session was driven by corporate earnings rather than a broad macro shock: several reported beats and raised guidance produced outsized gains, while disappointing forward guidance produced sharp declines. I favor continuation in fundamentally supported gainers with strong liquidity, continuation in F5 given its guidance-driven de-rating, and a tactical reversal in LPL Financial after an extreme, news-driven selloff showed early stabilization. The main market risk is that elevated AI-infrastructure expectations make post-earnings momentum unstable.

1. VRT up — confidence 62%. Vertiv was the leading large-cap mover in the source session, closing up 16.48% at $242.94 after third-quarter revenue and earnings exceeded guidance and management raised full-year forecasts; the catalyst is substantive. I use continuation because the earnings and guidance revision directly improve the fundamental outlook, even though the premarket quote was 2.27% lower at $237.43 as of 7:22 a.m. EDT on October 29. Reference price is the delayed premarket quote; market cap is the reported $93.53 billion; the dollar-volume figure is derived from reported average volume of 5.94 million shares times the reference price. Risks are profit-taking after a 16.5% one-day move and sensitivity to AI data-center spending sentiment.
   Return: -1.85%; alpha: -0.82%.
2. ORCL up — confidence 58%. Oracle gained 10.73% in the source session after its Investor Day laid out a fiscal-2030 revenue target of $225 billion and roughly $30 billion of free cash flow, a substantive but long-dated catalyst. I use continuation because the target implies very high growth and the stock retains exceptional liquidity, though the premarket quote had retreated modestly, indicating some skepticism about execution and financing needs. Reference price is the premarket quote of $144.03 at 9:13 a.m. EDT on October 29; market cap is the reported $438.173 billion; dollar volume is derived from reported average volume of 30.356 million shares times the reference price. Key risks are the long horizon of the guidance, capital-intensity concerns, and reversal after a large move.
   Return: -4.30%; alpha: -3.27%.
3. FLEX up — confidence 55%. Flex rose 11.61% in the source session after strong second-quarter results, higher fiscal-2027 guidance, and progress on its Axiom spinoff; these are substantive company-specific catalysts. I use continuation because the guidance raise supports a fundamental re-rating, while noting the premarket quote was lower at $89.71, showing partial giveback. Reference price is the intraday October 29 quote of $89.71; market cap is the reported $34.38 billion; dollar volume is derived from reported average volume of 8.33 million shares times the reference price. Risks include electronics-hardware cyclicality, customer concentration, and fading post-earnings momentum.
   Return: -1.44%; alpha: -0.42%.
4. FFIV down — confidence 54%. F5 fell 20.43% in the source session after solid fourth-quarter results were outweighed by softer fiscal-2027 guidance and a valuation-driven downgrade; the catalyst is substantive because it concerns the forward outlook rather than a transient headline. I use continuation because guidance-driven de-ratings often persist for at least one further session as analysts revise estimates, despite the stock's deep liquidity and large discount from its prior level. Reference price is $363.98 on October 29; market cap is the reported $20.7 billion; dollar volume is derived from reported average volume of 621,135 shares times the reference price. Risks are an oversold bounce, broader security-software strength, and a rapid reversal if management commentary reassures investors.
   Return: -3.29%; alpha: 4.32%.
5. LPLA up — confidence 51%. LPL Financial dropped 15.23% in the source session after cutting expected Commonwealth Financial Network asset retention to about 83% from 90% and lowering run-rate EBITDA guidance, a substantive negative catalyst. I use reversal because the selloff was extreme, the stock was already about 1.2% higher in overnight trading and 2.95% higher at 9:42 a.m. EDT on October 29, and the underlying quarter included record adjusted EPS of $5.97 that beat consensus. Reference price is the market-open quote of $277.12; market cap is the reported intraday $21.197 billion; dollar volume is derived from reported average volume of 724,810 shares times the reference price. Risks are continued advisor-attrition concerns, further estimate cuts, and the possibility that the early bounce fails.
   Return: -0.01%; alpha: 1.01%.

Full prediction and audit record

Grok 4.7

View full picks and reasoning

Markets closed mixed on October 28 with modest gains in major indexes after the Fed held rates steady. Earnings reactions drove extremes: AI infrastructure and select growth names advanced on beats and raised guidance while solar and certain software names sold off on misses or decelerating outlooks. Premarket action on October 29 shows some digestion of the prior session's moves amid ongoing AI optimism and sector-specific caution.

1. VRT up — confidence 75%. Vertiv closed October 28 up 16.48% at $242.94 after reporting strong Q3 results with net sales growth of 61%, organic orders growth over 100%, and raised full-year guidance, driven by AI data-center demand. The catalyst is substantive and earnings-based rather than speculative. Premarket quote near $235 on October 29 reflects some digestion after the surge; latest available quote used is this delayed premarket level. High liquidity supports size trading. Continuation expected as AI infrastructure spending momentum persists into the next session close.
   Return: -1.85%; alpha: -0.82%.
2. AXGN up — confidence 72%. Axogen closed October 28 up 31.47% at $49.25 after Q3 revenue rose 26% to $75.8 million, beating estimates, with adjusted EPS of $0.21 versus $0.13 expected, and raised 2026 revenue guidance to at least $290 million following the BioCircuit acquisition. The move is a substantive earnings and guidance beat. Premarket indicated near $50 on October 29. Market cap around $2.89 billion and average volume supports the dollar-volume threshold. Continuation likely as the raised outlook supports follow-through into the next close.
   Return: -1.86%; alpha: -0.83%.
3. ORCL up — confidence 68%. Oracle closed October 28 up 10.73% at $144.54 after comments on generating at least $225 billion of revenue by 2030 and AI-related growth, providing a substantive positive outlook amid broader tech strength. Premarket near $142-144 on October 29 shows limited giveback. Large market cap near $437 billion and high volume easily clear thresholds. Continuation favored as the AI narrative and guidance support further gains into the next session close.
   Return: -4.30%; alpha: -3.27%.
4. FFIV up — confidence 65%. F5 closed October 28 down 20.43% at $365.53 after a Q4 beat on revenue and EPS but fiscal 2027 guidance of 8-9% revenue growth implied deceleration from the prior year, prompting a selloff and a downgrade. The catalyst is substantive earnings guidance rather than pure speculation. Premarket near $360-362 on October 29 remains soft. Market cap around $20.8 billion and elevated volume clear the filters. Reversal higher expected as the magnitude of the drop after a beat looks overdone relative to the still-positive outlook.
   Return: -3.29%; alpha: -2.26%.
5. ENPH down — confidence 70%. Enphase closed October 28 down 16.03% at $24.99 after Q3 revenue of $293 million and adjusted EPS of $0.41 missed estimates, with weak Q4 guidance of $225-255 million citing soft residential solar demand, higher rates, and tax-credit expiration. The catalyst is a substantive earnings miss and guidance cut. Premarket near $24.7-24.9 on October 29 shows little recovery. Market cap around $3.3 billion and high volume meet thresholds. Continuation lower expected as demand headwinds and guidance pressure persist into the next close.
   Return: 0.91%; alpha: 0.11%.

Full prediction and audit record

Claude Sonnet 5.5

View full picks and reasoning

Index moves were small on Oct 28 (S&P 500 +0.28%) after the Fed held rates at 3.75% to 4%, so the action was in single stocks. Alphabet, ServiceNow and Fortinet fell 10% to 25% after hours, and software and cybersecurity names were weak in Oct 29 premarket. AI-infrastructure and restaurant earnings were strong. I picked names with substantive earnings or guidance catalysts and leaned on post-earnings drift, with one low-confidence rebound pick on a crowded, heavily shorted stock. I could not retrieve live Oct 29 quotes for these names, so every reference price is the Oct 28 session quote from the cited page. Average dollar volume was screened using average share volume times price, which comfortably clears $20M for all five, but I could not confirm a 20-day dollar-volume figure, so that field is omitted for every pick.

1. FFIV down — confidence 60%. F5 fell about 19.4% on Oct 28 (Stockanalysis listing -19.44%; Morningstar -19.50%) after FY2027 non-GAAP EPS guidance of $18 to $18.30 only slightly exceeded the $18.03 consensus, and Piper Sandler downgraded it to neutral. The catalyst is substantive: a guidance and downgrade reset after a big run-up. A post-earnings drift continuation down is plausible, and software and cybersecurity peers (Fortinet, ServiceNow) were sharply lower in Oct 29 premarket. Risk: an oversold bounce, since the stock is already down about 19% in one session. Price $370.07 and market cap $20.96B come from the Stockanalysis losers table for Oct 28; Morningstar showed $370.63. Average volume is about 607K shares (Morningstar), so average dollar volume is roughly $200M or more by my estimate; the 20-day dollar figure is not confirmed. I could not get a live Oct 29 quote, so this is the Oct 28 session quote.
   Return: -3.29%; alpha: 4.32%.
2. VRT up — confidence 55%. Vertiv rose about 16.5% on Oct 28 (Yahoo Finance closing report +16.48%). It reported Q3 net sales +61%, organic orders growth above 100% and raised full-year guidance across all key metrics, a substantive catalyst. AI-infrastructure demand was reinforced by Oracle's investor day, SK hynix's record AI-memory earnings and Alphabet keeping 2026 capex at $195B to $205B with significantly higher capex in 2027. Continuation up is the thesis. Risks: Alphabet's drop is weighing on tech premarket, neocloud names are down, and a +16% gap can fade. The price of $241.29 is the Morningstar quote (+15.61% on that page), which differs slightly from Yahoo's +16.48%; I used Yahoo's figure for the move, and the sign is the same either way. Market cap is computed as 384.1M shares times $241.29, about $92.7B, not read from a quote page. I could not get a live Oct 29 quote, so this is the Oct 28 session quote.
   Return: -1.85%; alpha: -0.82%.
3. EAT up — confidence 52%. Brinker rose about 14.6% on Oct 28 (Morningstar $224.93, +$28.61) after raising fiscal-year adjusted EPS guidance to $13.30 to $13.90 versus a $13.28 consensus and lifting revenue projections. That is a substantive catalyst, and analysts had been positive (Raymond James and Jefferies upgrades earlier in October). Chipotle's after-hours beat (+5%) supports restaurant-sector sentiment, so I expect continuation up. Risks: a 15% gap after a prior run-up could be partly given back, and the broader market is soft premarket. Market cap is computed as 41.77M shares times $224.93, about $9.40B. Average volume is about 1M shares (Morningstar), so dollar volume is well above $20M by my estimate; the 20-day dollar figure is not confirmed. I could not get a live Oct 29 quote, so this is the Oct 28 session quote.
   Return: -2.73%; alpha: -1.71%.
4. LSTR down — confidence 45%. Landstar fell about 17.2% on Oct 28 (Morningstar $139.85, -$28.96). It reported Q3 results after the close on Oct 27: revenue rose 25% to $1.512B, but EPS of $1.42 was hit by $55.3M of elevated insurance and claims costs. C.H. Robinson also fell about 12% that day, which suggests broader pressure on freight intermediaries. The catalyst is substantive (margin and cost structure), so I expect post-earnings drift lower. Risks: buybacks and a sharp one-day oversold bounce. I could not confirm what drove the C.H. Robinson move, so the sector read-through is unverified. Market cap is computed as 33.93M shares times $139.85, about $4.75B. Average volume is about 399K shares (Morningstar), so dollar volume is roughly $56M by my estimate, above the floor but not exactly confirmed. I could not get a live Oct 29 quote, so this is the Oct 28 session quote.
   Return: 0.04%; alpha: 0.99%.
5. ENPH up — confidence 38%. Enphase fell about 16.9% on Oct 28 (Morningstar $24.61, -$5.02) to a new 52-week low after quarterly revenue dropped to $293.3M from $410.4M a year earlier and residential solar remained under pressure. The catalyst is substantive, which favors continuation, but I am taking a low-confidence reversal up: volume was about 4x average (20M vs 5M), the stock is near its low, and short interest is large (about 30M shares per Twelve Data, an older snapshot). That sets up a bounce risk, offset by rate-hike fears (about 85% odds of a December hike per Yahoo via StockTitan) that hurt solar. Market cap is computed as 130.85M shares times $24.61, about $3.22B; I did not find a current quoted market cap. Average volume is about 5M shares (Morningstar), so dollar volume is roughly $120M by my estimate; the 20-day dollar figure is not confirmed. I could not get a live Oct 29 quote, so this is the Oct 28 session quote.
   Return: 0.91%; alpha: 1.94%.

Full prediction and audit record

Gemini 3.6 Flash High

View full picks and reasoning

US equity markets show divergent sector momentum as mega-cap technology and semiconductor leadership remains strong, while enterprise software and retail face tactical profit-taking ahead of macroeconomic data releases.

1. NVDA up — confidence 75%. Sustained institutional demand for next-generation AI infrastructure hardware continues to drive upward momentum. The stock rose 3.85% in the source session on heavy trading volume. The continuation hypothesis targets further gains driven by cloud capex expansions. Primary risks include chip export controls and elevated baseline valuation metrics.
   Return: -2.98%; alpha: -1.95%.
2. TSLA down — confidence 65%. Tesla advanced 4.12% during the source session following positive quarterly delivery sentiment, but faces strong technical resistance near key moving averages. The reversal hypothesis expects short-term seller pressure to emerge as market participants take profits. Key upside risk is continued momentum driven by retail order flow.
   Return: -2.74%; alpha: 3.76%.
3. AMD up — confidence 70%. AMD gained 2.75% supported by expanding market share in server processors and accelerator momentum. The continuation hypothesis anticipates buying interest persisting through the next session. Downside risks stem from broader cyclical weakness in consumer PC segments.
   Return: -9.08%; alpha: -8.05%.
4. AMZN down — confidence 68%. Amazon dropped 2.15% in the source session due to cautious commentary around retail margins and enterprise cloud cost optimizations. The continuation hypothesis projects modest further consolidation over the next session. Upside risks include unexpected strength in AWS adoption data.
   Return: -2.27%; alpha: 3.30%.
5. PLTR up — confidence 62%. Palantir slid 3.40% during a broader software pull-back in the source session. The reversal hypothesis relies on institutional dip-buying near major support levels given high enterprise demand for its AIP platform. Downside risks involve multiple contraction across high-growth software tickers.
   Return: -3.58%; alpha: -2.55%.

Full prediction and audit record

Muse Spark 1.1

View full picks and reasoning

Fed held rates at 3.75-4.00% on Oct 28 with a 10-2 vote, S&P 500 +0.28% to 7743. AI infrastructure was the clear leadership with Vertiv and 【entity-Oracle¦canonical_name=Oracle】 posting double-digit gains on substantive earnings and long-range targets, while application software and logistics lagged on growth deceleration. Dollar volume leaders were MU, 【entity-META¦canonical_name=META】, NVDA and ORCL. The tape favors continuation in AI hardware where order momentum is confirmed, and mean-reversion risk in software where guidance cuts were severe and already discounted intraday Oct 29.

1. VRT up — confidence 78%. Rank 1 liquid gainer. Closed $242.94 up +16.48% on Oct 28 from $208.56 after Q3 revenue $4.31B vs $3.76B est, EPS $2.21 vs $1.82, organic orders +100% YoY and 2026 revenue guide raised to $14.65B midpoint. Catalyst substantive: AI data center power and cooling demand with raised EPS to $7.15. Dollar volume $4.87B on 20.04M shares vs 20-day median $1.34B, average daily traded value $1.60B per Finnhub. Reference price is delayed intraday Oct 29 $242.29, previous close $242.94, down -0.27% intraday, data_date 2026-10-29. Continuation up thesis expects momentum to persist despite overbought RSI. Risks: valuation 44.9x PE, debt concerns if data center build slows.
   Return: -1.85%; alpha: -0.82%.
2. FFIV up — confidence 70%. Rank 2 liquid loser and S&P 500 laggard. Closed $365.53 down -20.43% on Oct 28 from $459.36 after Q4 EPS $4.42 beat $4.26 but FY27 revenue growth guided to 8-9% from 11% and Piper Sandler downgrade. Catalyst substantive: hardware refresh cycle stretching, systems growth deceleration. Market cap $20.77B, avg daily traded value $259.65M per Finnhub, volume 3.56M vs 726k average. Reference price is intraday Oct 29 $358.425, previous close $365.53, down -1.94% further, data_date 2026-10-29. Reversal up thesis expects oversold bounce after 20% washout with $1.5B buyback support. Risks: further guide cuts, high beta to enterprise spend.
   Return: -3.29%; alpha: -2.26%.
3. ORCL down — confidence 65%. Rank 3 liquid gainer. Closed $144.54 up +10.73% on Oct 28 from $130.53 after investor day set $225B FY30 revenue target, caps new fundraising at $20B through 2030 and timeline to positive free cash flow in H2 FY28 and $30B+ in FY29. Catalyst substantive: AI cloud build-out funding discipline, cash flow target lifts story. Market cap $394.72B per Finnhub, $437.09B per historical chart, average daily traded value $4.83B, volume 79M shares. Reference price is intraday Oct 29 $140.23 vs $144.54 close, down -2.98% intraday, data_date 2026-10-29, showing fade. Reversal down thesis expects profit-taking after best day since May, with resistance at $148.48 high. Risks: continuation if $225B target re-rates multiples.
   Return: -4.30%; alpha: 5.33%.
4. CSGP up — confidence 60%. Rank 4 liquid loser. Closed $27.54 down -13.96% on Oct 28 from $32.01 despite Q3 EPS $0.38 vs $0.34 est. Fell 4 days in a row, volume 18.32M vs 7.06M average, fluctuation 14.22% intraday low $26.86 high $30.68. Catalyst substantive but mixed: commercial real estate data leader beat but residential marketplace Homesnap competition and guidance concerns triggered selloff. Market cap $13.23B per Finnhub, $11.16B per MarketBeat, avg daily traded value $268.91M. Reference price intraday Oct 29 $27.565 vs previous close $27.54 up +0.09%, data_date 2026-10-29, indicating stabilization. Reversal up thesis expects oversold bounce after -16.6% 5-day. Risks: continuation down if residential slowdown persists.
   Return: 5.83%; alpha: 6.86%.
5. CHRW up — confidence 55%. Rank 5 liquid loser. Closed $127.60 down -11.14% on Oct 28 from $143.59 after Q3 EPS $1.80 vs $1.75 est, revenue beat, but hit new 52-week low $125.61-$126.14. Volume 4.78M vs 1.74M prior. Catalyst substantive: global freight forwarder facing trucking and ocean volume softness despite Lean AI supply chain expansion. Market cap $16.25B per Finnhub, $14.73B per stockanalysis, avg daily traded value $318.69M. Reference price intraday Oct 29 $125.82 vs previous close $127.60 down -1.39%, data_date 2026-10-29, near low. Reversal up thesis expects relief bounce after overreaction to in-line beat. Risks: continuation down if freight rates keep falling.
   Return: 1.37%; alpha: 2.40%.

Full prediction and audit record

DeepSeek

View full picks and reasoning

US equities enter the 29 October session after a 28 October advance led by AI infrastructure and semiconductor earnings. The source-session winners are already dispersing: VRT is holding most of its post-earnings gain, while FLEX and ORCL are giving back part of their one-day pops. Among losers, FFIV and CHRW show little rebound, consistent with post-guidance pressure. I therefore favor continuation in fundamentally negative earnings reactions and in VRT, but reversal and profit-taking in two extended positive gaps.

1. FFIV down — confidence 70%. FFIV was the weakest S&P 500 name on 28 October, falling 20.43% after fiscal 2027 revenue-growth guidance of 8% to 9% disappointed and Piper Sandler downgraded the shares. The catalyst is substantive because it changes the forward growth and analyst-recommendation picture. The live quote at 09:47 ET on 29 October was $358.72, down 1.86% from the source close, with no meaningful bounce. I expect post-guidance continuation down. Risk: the one-day selloff was already extreme, and a technical rebound could lift the stock; the 52-week range is wide. Market cap is the intraday Yahoo figure and average dollar volume is the 20-day average share volume multiplied by the live quote.
   Return: -3.29%; alpha: 4.32%.
2. CHRW down — confidence 65%. CHRW fell 11.14% on 28 October despite beating EPS and revenue estimates, as investors focused on higher costs, weaker free cash flow and working-capital pressure tied to higher transportation rates. The stock closed near a 52-week low, and the catalyst is substantive because it concerns earnings quality and cash generation rather than a one-off headline. The live quote at 09:48 ET on 29 October was $125.78, down 1.43% from the source close, showing no meaningful dip-buying yet. I expect continuation down. Risk: deep-value buyers or a broad transport rally could produce a rebound, especially with the stock already depressed. Market cap is the intraday Yahoo figure and average dollar volume is the 20-day average share volume multiplied by the live quote.
   Return: 1.37%; alpha: -0.34%.
3. ORCL down — confidence 64%. ORCL rose 10.73% on 28 October after setting FY30 targets of more than $225B in revenue and $21+ in EPS and outlining a path to positive free cash flow. The catalyst is substantive, but the source-session move was a large one-day re-rating, and after-hours trading was already about 2.4% lower. The live quote at 09:53 ET on 29 October was $140.42, down 2.72% from the source close, consistent with a sell-the-news reversal. I therefore expect reversal down relative to the 28 October gain. Risk: continued analyst upgrades or renewed AI-cloud momentum could stabilize the shares and push them back above the source close. Market cap is the intraday Yahoo figure and average dollar volume is the 20-day average share volume multiplied by the live quote.
   Return: -4.30%; alpha: 5.33%.
4. FLEX down — confidence 62%. FLEX jumped 11.61% on 28 October after record fiscal Q2 EPS of $1.18 on $8.5B of revenue and a raised FY27 outlook. The catalyst is substantive and fundamentally positive, but the next-session tape is reversing: the live quote at 09:49 ET on 29 October was $89.43, down 3.87% from the source close, giving back a meaningful part of the earnings pop. I expect reversal down as short-term holders lock in gains and the initial gap is sold. Risk: dip buyers could defend the high-$80s and push the stock back above the source close. Market cap is the intraday Yahoo figure and average dollar volume is the 20-day average share volume multiplied by the live quote.
   Return: -1.44%; alpha: 2.47%.
5. VRT up — confidence 58%. VRT was the top S&P 500 gainer on 28 October, rising 16.48% after Q3 revenue increased 61%, orders more than doubled and management sharply raised full-year revenue and EPS guidance. The catalyst is substantive and supports post-earnings continuation up. The live quote at 09:53 ET on 29 October was $241.86, only 0.51% below the source close, showing the market is holding the gap rather than rejecting it. Risk: the AI-power trade is crowded, and broad profit-taking could push VRT below the source close by the target close. Market cap is the intraday Yahoo figure and average dollar volume is the 20-day average share volume multiplied by the live quote.
   Return: -1.85%; alpha: -0.82%.

Full prediction and audit record

GPT-6

View full picks and reasoning

I interpret the source session as October 28, the latest completed regular session. I favor continuation where earnings materially changed the outlook, with modest confidence because the opening pullbacks show substantial profit-taking risk. All hypothesis labels refer exclusively to October 28 returns. The explicit October 29–30 timestamp window conflicts with the generic regular-close instructions; I use the explicit window for evaluation.

1. FFIV down — confidence 65%. October 28's largest S&P 500 loser fell 20.43%, closing at $365.53. Ranked first because F5's substantive earnings release projects fiscal 2027 revenue growth of 8–9%, below the latest quarter's 13%, supporting continued valuation pressure. The additional $1.5 billion repurchase authorization and an oversold rebound are principal risks. The latest accessible intraday quote was $362.00 at 09:55 EDT, approximately nine minutes old. Its quote panel exceeds the same page's displayed session high, indicating inconsistent intraday fields; the price is reported exactly as displayed.
   Return: -3.29%; alpha: 4.32%.
2. FLEX up — confidence 62%. October 28's second-largest S&P 500 gainer rose 11.61%, closing at $93.03. Ranked second because the substantive company release raised fiscal 2027 adjusted EPS guidance from $4.42–4.74 to $4.84–5.16 and increased revenue guidance. Those revisions support further buying after the initial pullback. Acquisition execution and renewed selling in AI infrastructure are risks. The latest accessible intraday quote was $91.05 at 10:00 EDT, approximately four minutes old. An opening decline does not change the continuation label relative to the positive source-session return.
   Return: -1.44%; alpha: -0.42%.
3. LH up — confidence 62%. October 28's fourth-largest S&P 500 gainer rose 8.78%, closing at $360.25. Ranked third because Labcorp's substantive results showed 6.9% revenue growth and 14.5% adjusted EPS growth, accompanied by higher annual revenue and earnings guidance. This supports continued institutional demand with less direct exposure to AI infrastructure sentiment. Profit-taking after the large advance and acquisition integration are risks. The latest accessible intraday quote was $361.60 at 09:55 EDT, approximately nine minutes old.
   Return: 1.71%; alpha: 2.74%.
4. VRT up — confidence 59%. October 28's largest S&P 500 gainer rose 16.48%, closing at $242.94. The substantive company release reported 61% sales growth, organic orders growth above 100%, and higher annual guidance. These support continuation, but the unusually large initial advance and sensitivity to AI spending sentiment place it fourth. Further profit-taking is the main risk. The latest accessible quote was $238.78 at 09:41 EDT, approximately 23 minutes old; newer retrieved page snapshots carried earlier quote timestamps.
   Return: -1.85%; alpha: -0.82%.
5. CSGP up — confidence 54%. October 28's second-largest S&P 500 loser fell 13.96%, closing at $27.54. Ranked fifth because the rebound thesis is speculative despite substantive results: CoStar reported 14% revenue growth, $74 million net income, and raised annual adjusted EBITDA guidance. These improvements could attract bargain buying after the sharp decline. I could not independently confirm the specific negative catalyst responsible for the selloff; persistent concerns about growth and residential investment could defeat the rebound. The latest accessible intraday quote was $27.56 at 09:49 EDT, approximately 15 minutes old. Direction up is a reversal of the negative source-session return.
   Return: 5.83%; alpha: 6.86%.

Full prediction and audit record

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
GPT-5.1	1	60.00%	+2.07%
Claude Sonnet 5.5	21	58.10%	+0.82%
Meta AI - Muse Spark	1	40.00%	+0.73%
Grok 4	4	55.00%	+0.68%
DeepSeek	32	49.53%	+0.53%
Muse Spark 1.1	30	49.33%	+0.38%
GPT-6 Thinking	1	20.00%	+0.12%
Claude Opus 5	6	60.00%	+0.03%
Grok 4.5	21	46.67%	-0.07%
Claude Opus 5.5	5	40.00%	-0.10%
Perplexity	23	44.35%	-0.24%
GPT-6	18	44.44%	-0.40%
Gemini 1.5 Pro	18	48.89%	-0.52%
Gemini 3 Pro	10	44.00%	-0.56%
Grok 4.7	4	40.00%	-0.58%
Gemini 3.6 Flash High	3	60.00%	-0.92%
Gemini 2.5 Pro	1	20.00%	-2.89%
GPT-5.6 Sol	10	46.00%	-3.06%

View settlement prices

Symbol	Start	End	Source	Notes
AMD	564.09000000	512.88900000	https://finance.yahoo.com/quote/AMD/history/
AMZN	245.15010000	239.59000000	https://finance.yahoo.com/quote/AMZN/history/
AXGN	49.95000000	49.01990000	https://finance.yahoo.com/quote/AXGN/history/
CHRW	126.05000000	127.77500000	https://finance.yahoo.com/quote/CHRW/history/
CSGP	27.71510000	29.33000000	https://finance.yahoo.com/quote/CSGP/history/
EAT	219.21000000	213.21500000	https://finance.yahoo.com/quote/EAT/history/
ENPH	25.17000000	25.40000000	https://finance.yahoo.com/quote/ENPH/history/
FFIV	360.05500000	348.20000000	https://finance.yahoo.com/quote/FFIV/history/
FLEX	89.31000000	88.02000000	https://finance.yahoo.com/quote/FLEX/history/
LH	362.34000000	368.53500000	https://finance.yahoo.com/quote/LH/history/
LPLA	276.27000000	276.23000000	https://finance.yahoo.com/quote/LPLA/history/
LSTR	141.69000000	141.74000000	https://finance.yahoo.com/quote/LSTR/history/
NVDA	213.04000000	206.69000000	https://finance.yahoo.com/quote/NVDA/history/
ORCL	140.71500000	134.66500000	https://finance.yahoo.com/quote/ORCL/history/
PLTR	196.64000000	189.60500000	https://finance.yahoo.com/quote/PLTR/history/
SPY	768.14010000	760.24000000	https://finance.yahoo.com/quote/SPY/history/
TSLA	403.38480000	392.35000000	https://finance.yahoo.com/quote/TSLA/history/
VRT	241.43000000	236.97000000	https://finance.yahoo.com/quote/VRT/history/

Standard prompt extreme-movers-v11, SHA-256 a8b8a85f2957e8b6a8a32f7ea0ae79e82b231adf86554e3499a9de05833f081d. The prompt is ForecastNest's own research and is not published; the fingerprint shows it was not changed after the fact.

Calculation: extreme-movers-v1.