Event Gaps: Claude Opus 5.5's picks for 2026-10-30
Claude Opus 5.5 picked RDDT, INTC, MDLZ, RIVN, TWLO for a 1-day horizon. Intraday alpha vs SPY: +3.66%. Picks with positive alpha: 100%.
| Experiment | Event Gaps |
|---|---|
| Model | Claude Opus 5.5 |
| Prompt version | event-gap-v13 |
| Asked | |
| Horizon | 1-day |
| Measured from | 2026-10-30 |
| Measured to | 2026-11-02 |
| Benchmark | SPY |
| Status | Settled |
| Intraday alpha vs SPY | +3.66% |
| Picks with positive alpha | 100% |
| Benchmark return | +0.59% |
The picks
| Rank | Symbol | Gap direction | Price change | Return in the called direction | Confidence | vs benchmark | Called right |
|---|---|---|---|---|---|---|---|
| 1 | RDDT | -1.33% | +1.33% | 58% | +0.73% | Yes | |
| 2 | INTC | -3.18% | +3.18% | 50% | +2.59% | Yes | |
| 3 | MDLZ | +1.89% | +1.89% | 55% | +1.30% | Yes | |
| 4 | RIVN | +10.98% | +10.98% | 50% | +10.39% | Yes | |
| 5 | TWLO | +3.87% | +3.87% | 45% | +3.28% | Yes |
The model's market view
Risk-off, tech-led session. Alphabet's post-earnings drop on Thursday, Amazon's Q3 miss with negative free cash flow, and Intel's Q4 guide-down hit the AI/semis complex: Nasdaq Composite about -2.3% and Dow about -0.6% at 10:27 ET per Yahoo Finance's live blog; QQQ was about -2.5% and SPY about -1% in premarket per QSG. Megacaps showed dip-buying after the open (Amazon recovered from about -7% premarket to about -2% by 10:27 ET), so gap-continuation is not guaranteed. The book mixes continuation shorts in names where the business outlook itself changed (Reddit's search-traffic and DAU warning, Intel's guide reset) with longs in beat-and-raise names that are moving against the tape (Mondelez as a defensive staple, Rivian on a beat/raise plus upgrade, Twilio on an intraday reversal after a full-year raise). The score is net of SPY, so defensive longs and broken-guide shorts both benefit if the tech rotation keeps going. The window runs to 10:35 ET Monday Nov 2. US daylight saving ends Nov 1, so 15:35Z is 10:35 EST, and the window spans a weekend gap.
Why each pick
RDDT
THESIS (down): The Q3 beat (revenue $895M, +53% y/y) is backward-looking. What changed is the outlook: Q4 revenue guidance of $1.00-1.03B was below the ~$1.046-1.05B consensus, management said search-referral traffic fell faster late in the quarter, warned DAUq could fall by a low-double-digit % in Q4 from 130.3M, and plans to test a primarily logged-in experience. These are FACTS from the company release and letter. INFERENCE: a logged-in shift can protect monetization (>80% of revenue comes from logged-in users), but it shrinks the top-of-funnel. It also arrives while the shares are already at multi-year lows, so it is a fundamental reset, not just sentiment, and sell-side target cuts (B. Riley 270->190 today) usually keep coming for several days.
COMPLETED DAILY OHLCV (US/Eastern regular sessions, split-adjusted; sources: https://www.tipranks.com/stocks/rddt/historical-prices for Oct 23-28, https://sg.finance.yahoo.com/quote/RDDT/ for Oct 29):
2026-10-23 O148.01 H149.50 L143.92 C144.25 V4060729
2026-10-26 O144.25 H144.55 L133.33 C138.45 V7869811
2026-10-27 O138.39 H140.98 L135.07 C136.46 V4613246
2026-10-28 O135.25 H138.16 L133.95 C134.70 V5326334
2026-10-29 O129.95 H131.89 L126.37 C129.60 V16862051
return_5d_pct=-13.75 (129.60 vs 150.26 on Oct 22). high_20d=164.94 (Oct 16), low_20d=126.37 (Oct 29), avg_volume_20d=4880604 (Oct 2-Oct 29, 20 sessions). Note: Yahoo's quote page shows previous close 129.87 against 129.60 on its history header; I used 129.60.
PRICE ACTION: five straight lower closes into the print, with Oct 29 volume 3.5x the 20-day average. That is distribution before the event, not capitulation. Oct 26 left a long lower wick but closed near the middle of its range, and every bar made lower highs. The stock had already broken below the 52-week low of 119.27 premarket.
INTRADAY (PARTIAL SESSION, not in the stats above): latest_price=97.24, quote_time=2026-10-30T09:51:36-04:00, live/real-time regular session, source https://finance.yahoo.com/quote/RDDT/. Open 100.25, high_so_far 101.95, low_so_far 95.04, cumulative_volume 7583540 as of that time. Pre-event price 129.60 (Oct 29 close) -> observed event-to-quote move -24.97%. Relative intraday volume cannot be established (no same-time baseline).
FORECAST: the ~25% gap is already realised and not part of the forecast. I expect a further ~4% decline from the quote over the remaining window, as more target cuts land and holders exit after a 62% drawdown from the high.
COUNTER-EVIDENCE: the stock is very oversold, trades at ~22x trailing EPS after the drop, has a $1B buyback authorization, an average target of ~$213, and a 10% short float that could squeeze. The weekend gap adds two-way risk.
INVALIDATION: a sustained move back above today's 101.95 high, i.e. holding above the 100.25 open (the gap-and-go failing), would invalidate the continuation call.
INTC
THESIS (down): FACTS: Q3 beat (adj EPS $0.52 vs $0.40; revenue $16.51B vs $16.47B). But Q4 guidance was revenue $15-16B vs ~$17.05B consensus, adj EPS $0.25 vs ~$0.43-0.44, and gross margin ~38% vs 43% in Q3. Intel also flagged PC market declines of 12% in 2026 and 10% in 2027. Cantor estimates the guide implies PC units ~45% lower y/y, blaming memory, wafer and substrate shortages. This is a change to business expectations, not just sentiment. Sell-side targets were cut sharply today (Truist 108->72, RBC 100->75, JPMorgan 85->80, Northland 120->93), and Morningstar cut fair value to $85. INFERENCE: consensus EPS for Q4 and 2027 still has to come down, and AI/semis are de-risking as a group (SOXX ~-6% premarket), which favours more downside in the first sessions.
COMPLETED DAILY OHLCV (US/Eastern regular sessions, split-adjusted; source https://www.tipranks.com/stocks/intc/historical-prices for Oct 23-28; Oct 29 close from Yahoo's previous close, https://finance.yahoo.com/quote/INTC/):
2026-10-23 O85.92 H87.87 L84.74 C86.96 V71747883
2026-10-26 O87.94 H92.09 L86.46 C90.47 V98099266
2026-10-27 O92.47 H94.98 L91.86 C93.74 V89980609
2026-10-28 O95.09 H95.68 L93.00 C95.53 V81517047
2026-10-29 O null H null L null C92.07 V null (OHLV unverified)
return_5d_pct=7.70 (92.07 vs 85.49 on Oct 22). high_20d=null and low_20d=null because the Oct 29 range is unverified; the 19 verified sessions (Oct 2-28) give a high of 126.00 (Oct 2) and a low of 84.65 (Oct 22). avg_volume_20d=null (Oct 29 volume missing); the Oct 1-28 20-session mean was 96517076 shares, which is also the basis of the dollar-volume field.
PRICE ACTION: the stock rallied ~12% off the Oct 22 low into the print on below-average volume, closing near the high on Oct 28. That is a crowded long into a guide-down. Today it gapped under the summer ~$80 floor that held from July to October, and the 20-session context is a breakdown, not a retest.
INTRADAY (PARTIAL SESSION): latest_price=65.54, quote_time=2026-10-30T10:18:02-04:00, live/real-time regular session, source https://finance.yahoo.com/quote/INTC/. Open 68.35, high_so_far 68.77, low_so_far 64.12, cumulative_volume 147687394 as of that time (Robinhood showed a 69.80 high, unreconciled). Pre-event price 92.07 -> observed event-to-quote move -28.82%. Yahoo's 10:27 ET blog showed -27.28%, so there was a small bounce. Relative intraday volume vs a same-time baseline cannot be established.
FORECAST: the ~29% drop has already happened. I expect a further ~3% from the quote as target cuts and index/ETF de-risking continue. Confidence is modest.
COUNTER-EVIDENCE: this is one of the worst days since 1974; even the slashed targets ($72-93) sit above the current price; retail dip-buying, any bounce in AI peers, and the weekend gap all create real reversal risk.
INVALIDATION: a sustained move above today's 68.77 high, i.e. above the opening print (the gap starting to fill), would invalidate the call.
MDLZ
THESIS (up): FACTS: Q3 adj EPS $0.78 vs ~$0.71 consensus (~10% beat). Revenue was $9.97B, roughly in line; organic growth was +2.1% with volume/mix +1.8 pp, a volume inflection. Adjusted gross margin rose 380 bp to 34.2% and adjusted operating margin rose 190 bp to 13.9%. FY26 guidance was reaffirmed (organic revenue >=2%, adj EPS flat to +5% cc, FCF ~$3B). Targets were raised (TD Cowen $75, Jefferies $73, UBS $68; Goldman reiterated Buy at $75). This is a fundamental improvement in margin and volume quality. INFERENCE: on a day when the AI/tech trade is unwinding, a low-beta staple with an earnings catalyst is a natural rotation destination, which matters because the score is net of SPY.
COMPLETED DAILY OHLCV (US/Eastern regular sessions, split-adjusted; sources: https://stockanalysis.com/stocks/mdlz/history/ for Oct 23-26; https://stockscan.io/stocks/MDLZ/price-history for Oct 27-28 high/low/volume; Oct 27 close derived from its -0.98% change; Oct 28 close from Seeking Alpha and open from the CNBC quote page; Oct 29 open/high/low/close from https://www.fool.com/quote/nasdaq/mdlz/):
2026-10-23 O65.03 H65.20 L63.40 C63.80 V7420353
2026-10-26 O63.65 H63.78 L62.67 C62.97 V5595653
2026-10-27 O null H62.60 L62.00 C62.35 V7125789
2026-10-28 O61.62 H62.23 L61.51 C61.87 V5797827
2026-10-29 O62.19 H62.66 L61.91 C62.53 V null
return_5d_pct=-4.08 (62.53 vs 65.19 on Oct 22). high_20d=65.22 (Oct 22), low_20d=57.32 (Oct 5), Oct 2-Oct 29 window. avg_volume_20d=null (Oct 29 volume unverified); the Oct 1-28 mean was 7334400, which is also the basis of the dollar-volume field.
PRICE ACTION: a pullback of about -4% from the Oct 22 52-week-high close into the print, on lighter volume, then a close near the top of the range on Oct 29. The earnings gap is taking it through the prior 52-week high of 66.65.
INTRADAY (PARTIAL SESSION): latest_price=65.018 (last trade cited by MarketBeat; quote_time=null, exact time not stated, article published 2026-10-30T13:40:54Z), regular-session real-time per MarketBeat. Source https://www.marketbeat.com/instant-alerts/price-mondelez-international-nasdaq-mdlz-sets-new-52-week-high-on-better-than-expected-earnings-2026-10-30/. high_so_far 67.00, cumulative_volume 205921 as of that article. A Motley Fool quote of 65.45 (+4.67%) had no timestamp. Pre-event price 62.53 -> observed event-to-quote move +3.98%. Freshness cannot be fully verified.
FORECAST: the +4% gap is already realised. I expect a further ~1.5% from the quote, as staples rotation continues and the post-earnings drift typical of quality-beat names plays out.
COUNTER-EVIDENCE: the 67.00 print followed by a retreat to ~65 is an upper wick, i.e. rejection at the highs. Revenue was only in line, Europe declined 3.3%, the FY guide was not raised, and a broad tech rebound could pull money back out of defensives.
INVALIDATION: a move back below the 62.53 prior close (a full gap fill) would invalidate the thesis.
RIVN
THESIS (up): FACTS: Q3 revenue $2.155B (+38% y/y) vs ~$2.0B consensus; adj loss/share $0.48 vs ~-$0.61 to -$0.65 expected; consolidated gross profit $176M; R2 became the best-selling model. 2026 guidance improved across metrics: deliveries 67-70k (low end raised from 65k), adj EBITDA loss narrowed to $1.75-1.85B, capex cut to $1.40-1.50B. Pro forma liquidity is $7.1B. Jefferies upgraded the stock to Buy ($17 target) on Oct 30, calling the ~30% pullback an entry point. This is a fundamental improvement (margin trajectory, funding visibility), not only sentiment. INFERENCE: the stock entered the print at a fresh 52-week low (11.38), heavily de-rated (down ~42% YTD), so the positioning reset leaves room for follow-through buying over the next session.
COMPLETED DAILY OHLCV (US/Eastern regular sessions, split-adjusted; sources: https://stockanalysis.com/stocks/rivn/history/ for Oct 23-26; https://stockscan.io/stocks/RIVN/price-history for Oct 27 high/low/volume and change; https://sg.finance.yahoo.com/quote/RIVN/ for Oct 28; https://finance.yahoo.com/quote/RIVN/ for Oct 29):
2026-10-23 O11.93 H12.02 L11.60 C11.83 V23554380
2026-10-26 O11.80 H11.81 L11.42 C11.43 V25075327
2026-10-27 O null H12.09 L11.50 C11.82 V35110865
2026-10-28 O11.74 H11.78 L11.44 C11.67 V27508555
2026-10-29 O11.69 H11.87 L11.38 C11.45 V41537048
return_5d_pct=-5.06 (11.45 vs 12.06 on Oct 22). high_20d=15.36 (Oct 2), low_20d=11.38 (Oct 29), avg_volume_20d=26306106 (Oct 2-Oct 29, 20 sessions).
PRICE ACTION: a steady downtrend with lower highs, and Oct 29 closed at the bottom of its range on 1.6x average volume, a breakdown into the event. That favours a short-covering/relief move after a positive surprise, but a weak trend can resume.
INTRADAY (PARTIAL SESSION): latest_price=12.03, quote_time=2026-10-30T09:59:00-04:00, regular session (Motley Fool quote module, likely delayed). Source https://fool.com/investing/2026/10/29/rivian-just-made-a-move-that-could-help-unlock-950. Range so far 11.60-12.21, cumulative_volume ~12.3M as of that time vs a 23.6M full-day average; relative intraday volume vs a same-time baseline cannot be established. Pre-event price 11.45 -> observed event-to-quote move +5.07%. The premarket high was ~12.53 (+9.4%), so part of the gap has already faded.
FORECAST: the ~5% gain is already realised. I expect a further ~3% from the quote.
COUNTER-EVIDENCE: the fade from premarket highs and a $11.60 dip after the open; high-beta growth is being sold today; the interim CFO change; Mizuho cut its target to $10; the long-term downtrend.
INVALIDATION: a move back below today's 11.60 low and the 11.45 prior close (gap fully filled) would invalidate the thesis.
TWLO
THESIS (up, reversal of an overshoot): FACTS: Q3 revenue was a record $1.56B (+20% reported, +16% organic) and adj EPS $1.65 vs $1.47; non-GAAP operating income was a record $326M (+39%) and FCF a record $370M. FY26 guidance was RAISED (organic growth 15% from 13-13.5%, reported 20%), and the preliminary 2027 view of >=10% organic growth and FCF >=$1.35B sits above the prior 7-8% framework. The selloff rests on Q4 organic guidance of 11-12% (vs 16% in Q3) and Messaging (~60% of revenue) slowing to 17% from 18%. INFERENCE: an ~18-23% drawdown for a raise-the-year print with a Q4 decel looks more like de-grossing in a tech rout than a broken thesis. Needham cut its target to $240 but kept Buy; Wells Fargo is at $300 Overweight.
COMPLETED DAILY OHLCV (US/Eastern regular sessions, split-adjusted; sources: https://stockanalysis.com/stocks/twlo/history/ for Oct 23-26; Oct 27 close from https://finance.yahoo.com/quote/TWLO/news/; Oct 28 close from Yahoo's previous close; Oct 29 from https://finance.yahoo.com/quote/TWLO/):
2026-10-23 O256.00 H263.70 L253.11 C261.79 V2081474
2026-10-26 O262.90 H264.62 L257.60 C263.65 V1549109
2026-10-27 O null H null L null C254.56 V null
2026-10-28 O null H null L null C253.77 V null
2026-10-29 O243.06 H247.16 L232.73 C241.90 V4847236
return_5d_pct=-4.62 (241.90 vs 253.63 on Oct 22). high_20d=null and low_20d=null (Oct 27-28 ranges unverified); the 18 verified sessions in Oct 2-29 give a high of 308.40 (Oct 6) and a low of 232.73 (Oct 29). avg_volume_20d=null; the Sep 29-Oct 26 20-session mean was 4038900, inflated by the 31.4M S&P inclusion day on Oct 5 (~2.6M without it), and is the basis of the dollar-volume field.
PRICE ACTION: the stock was already down ~22% from the Oct 6 high of 308.40 before the print and fell 4.7% on Oct 29 ahead of the report. Today it opened at 186.59 and rebounded to ~199, a strong close-to-open bounce off the gap low. The 20-session context is a deep pullback, not a fresh breakdown from highs.
INTRADAY (PARTIAL SESSION): latest_price=199.117 (last trade cited by MarketBeat; quote_time=null, exact time not stated), regular session. Source https://www.marketbeat.com/instant-alerts/price-twilio-nyse-twlo-shares-gap-down-following-analyst-downgrade-2026-10-30/. open 186.59, cumulative_volume 1904987 as of that article. TokenPost showed 186.73 at 09:36 ET. Pre-event price 241.90 -> observed event-to-quote move -17.69%. Freshness cannot be fully verified, which is reflected in the lower confidence.
FORECAST: the -18% drop is already realised. I expect ~+2.5% from the quote as the reversal extends.
COUNTER-EVIDENCE: messaging deceleration is a real fundamental negative; Q4 operating income steps down sequentially; insiders sold ~$135M in 90 days; UBS and Needham cut targets; a continued software/AI selloff could easily push it back to the open.
INVALIDATION: a break below the 186.59 opening low would invalidate the reversal thesis.
Other models on the same run
| Model | Picks | Intraday alpha vs SPY | Picks with positive alpha |
|---|---|---|---|
| Perplexity | INTC, AMZN, PLTR, AAPL, LLY | -1.02% | 40% |
| Muse Spark 1.1 | INTC, AMZN, REGN, RIVN, GOOGL | +1.72% | 60% |
| Grok 4.7 | REGN, INTC, AMZN, XOM, LLY | -0.61% | 40% |
| Gemini 3.6 Flash High | AAPL, AMZN, INTC, MSFT, GOOGL | -0.16% | 40% |
| DeepSeek | INTC, COIN, REGN, AMZN, AN | -0.82% | 40% |
| GPT-6 | ALNY, NOW, GKOS, CROX, FTNT | +1.96% | 60% |
Written up on ForecastNest Insights: Event Gaps: Claude Opus 5.5's picks for 2026-10-30.
ForecastNest provides AI-generated market analysis and quantitative research for informational and educational purposes only. Nothing here constitutes financial, investment, legal, or tax advice.