Asked Oct 29, 12:57 PM EDT (midday). GPT-6 answered the same standard prompt independently. Window: from Oct 29, 12:57 PM EDT to Oct 30, 12:57 PM EDT, the same time one session later. Benchmark: SPY.
Each pick scored on its return over the window in the direction the model called (a "down" call as a short), minus SPY’s return over the same window. 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 beat SPY over the window.
| Model | Correct | Hit rate | Return | SPY | Alpha vs SPY |
|---|---|---|---|---|---|
| GPT-6 | 3/5 | 60.00% | -4.05% | -1.78% | -2.28% |
GPT-6
View full picks and reasoning
Earnings reactions are highly dispersed. ServiceNow and Alphabet face pressure from bookings expectations and cash consumption, while Porch and Lilly have raised business expectations. Large moves already observed before this run are excluded from the forecast moves. Intel offers a substantial scheduled catalyst, but its direction is especially uncertain before results.
- NOW — confidence 64%. Observed catalyst: ServiceNow reported Q3 2026 subscription revenue of $3.988 billion and constant-currency cRPO growth of 21%, but guided Q4 cRPO growth to 18.5%. My bearish inference is that slower forward bookings growth can prolong the valuation reset through tomorrow's midday checkpoint. This is an expectations repricing thesis, not a claim that revenue contracted.
Before the event, the October 23 high of 159.50 was followed by three lower closes. October 28 opened below the preceding close, briefly recovered above it, then closed in the bottom 27.6% of its range on 2.32 times completed-session average volume. That supports distribution before results, although it does not by itself establish post-event continuation.
Verified pre-event reference: October 28 regular-session close 150.13, session-end convention 2026-10-28T16:00-04:00. Latest_price=111.73 USD; quote_time=2026-10-29T12:46-04:00; regular-session snapshot, provider labelled real-time, cached when retrieved; source=https://stockanalysis.com/stocks/now/. The pre-event-close-to-quote change is -25.5778%. It spans the earnings event and broader market trading, so it is not a pure causal event-return estimate. Most of the initial repricing has already occurred and is not the forecast.
PARTIAL SESSION, as of 2026-10-29T12:46-04:00: open=120.91; high_so_far=121.50; low_so_far=111.25; cumulative_volume=43492320 shares; source=https://stockanalysis.com/stocks/now/. The quote remains near the partial-session low and below the completed 20-session low. A same-time volume baseline is unavailable, so unusual intraday relative volume cannot be established.
Remaining-window expectation: approximately 4% additional downside from the latest accessible quote, with direction also called down for the authoritative window. Exact elapsed-window return=null. Strongest counter-evidence: management raised full-year subscription revenue and free-cash-flow-margin outlooks, while the already enormous decline creates substantial short-covering and emotional-reversal risk. A sustained recovery above the verified partial-session high of 121.50 after this quote would invalidate the continuation thesis.
daily_5d, oldest to newest, US Eastern regular sessions; columns=session_date|open|high|low|close|volume:
2026-10-22|154.25|157.60|153.57|157.00|17406898
2026-10-23|156.38|159.50|153.72|158.96|11100759
2026-10-26|158.55|159.30|155.00|158.18|14739149
2026-10-27|156.10|157.13|152.88|154.78|13481100
2026-10-28|151.60|155.15|148.22|150.13|28786305
return_5d_pct=-1.952717, using sixth close=153.12 on 2026-10-21; high_20d=159.50; low_20d=133.47; avg_volume_20d=12414211.4 shares. History and pre-event close source=https://stockanalysis.com/stocks/now/history/.
Return: -5.86%; alpha: 7.64%. - INTC — confidence 55%. Officially scheduled catalyst: Intel reports Q3 results promptly after today's market close, followed by its 2 p.m. PDT call. No Q3 outcome is known. The bullish inference rests on Q2's reported yield and cycle-time improvements, 25% revenue growth and subsequent recovery in the completed price series. Existing Q3 guidance is revenue of $15.8 billion to $16.8 billion and non-GAAP EPS of $0.38; exceeding expectations and showing durable manufacturing progress would be needed for an upside reaction. Increased manufacturing investment and a demanding turnaround valuation are material risks.
Before the upcoming event, October 22's decline closed near its low, followed by four rising closes. October 26 and 27 opened above the preceding closes; October 28 finished in the top 5.6% of its range with a small upper wick. However, its volume was only 0.85 times the completed-session average, and the stock remains far below the 20-session high of 126.00. The recovery supports upside potential but lacks strong final-day volume confirmation.
Latest_price=91.92 USD; quote_time=2026-10-29T12:50-04:00; regular-session snapshot, provider labelled real-time, cached when retrieved; source=https://stockanalysis.com/stocks/intc/. This is itself a pre-event price. The preceding completed close was 95.53 at the October 28 regular-session end, 2026-10-28T16:00-04:00. Its -3.7789% change to the quote is pre-report trading, not a reaction to earnings. Observed post-event reaction=null because the event is upcoming.
PARTIAL SESSION, as of 2026-10-29T12:50-04:00: open=92.97; high_so_far=93.50; low_so_far=91.55; cumulative_volume=39170250 shares; source=https://stockanalysis.com/stocks/intc/. No same-time volume baseline was verified, so relative intraday volume is unconfirmed.
Remaining-window expectation: approximately 6% upside from the latest accessible quote, spanning anticipation, the overnight report and tomorrow's trading through the checkpoint. Direction is also called up for the authoritative window; exact elapsed-window return=null. Strongest counter-evidence is today's gap down and quote near its partial-session low despite the preceding four-session recovery. A disappointing report can reverse optimism sharply, while a merely satisfactory report may trigger selling after the recent recovery. Q3 revenue below the verified $15.8 billion guidance floor, or a post-results failure to recover 91.55, would invalidate the bullish thesis.
daily_5d, oldest to newest, US Eastern regular sessions; columns=session_date|open|high|low|close|volume:
2026-10-22|87.92|89.19|84.65|85.49|134128080
2026-10-23|85.92|87.87|84.74|86.96|72093680
2026-10-26|87.94|92.09|86.46|90.47|98744158
2026-10-27|92.47|94.98|91.86|93.74|90523294
2026-10-28|95.09|95.68|93.00|95.53|82592443
return_5d_pct=2.764630, using sixth close=92.96 on 2026-10-21; high_20d=126.00; low_20d=84.65; avg_volume_20d=96840083.75 shares. History and prior close source=https://stockanalysis.com/stocks/intc/history/.
Return: -29.94%; alpha: -28.16%. - PRCH — confidence 61%. Observed catalyst: Porch's October 28 report showed Q3 Porch-owned revenue excluding the reciprocal up 22% and adjusted EBITDA excluding the reciprocal up 116% to $44.5 million, with raised full-year guidance. These are disclosed non-GAAP owner-business measures, not insurance premium volume or consolidated insurer cash. My bullish inference is that stronger earnings expectations can support follow-through after the initial gap; continued appreciation is not a reported fact.
Before the event, the five-session return was -12.3185%. October 23 and 26 showed failed rebounds and lower closes; October 28 reached 18.12 but closed at 16.30, only 8.1% above its session low, on 1.97 times completed-session average volume. That is substantial opposing evidence from the pre-report path. Today's move nevertheless crossed the completed 20-session high of 18.75, changing the immediate price context.
Verified pre-event reference: October 28 close=16.30, regular-session end convention 2026-10-28T16:00-04:00. Latest_price=20.42 USD; quote_time=2026-10-29T12:54:43-04:00; regular-session snapshot, cached when retrieved; source=https://www.google.com/finance/quote/PRCH:NASDAQ. The pre-event-close-to-quote change is +25.2761%, spanning the event and subsequent market trading rather than isolating its causal effect. This substantial move is already observed and is excluded from the forecast gain.
PARTIAL SESSION, as of the separately timestamped 2026-10-29T12:44-04:00 snapshot: open=19.46; high_so_far=21.30; low_so_far=18.74; cumulative_volume=4285914 shares; source=https://stockanalysis.com/stocks/prch/. These partial figures precede the quoted Google observation. No same-time volume baseline was verified; partial cumulative volume is not compared with full-day average volume.
Remaining-window expectation: approximately 3.5% additional upside from the latest quote, with direction also called up for the authoritative window. Exact elapsed-window return=null. Strongest counter-evidence is the retreat from 21.30 after an already large gap, combined with heavy pre-event selling. Emotional chasing, profit-taking and financing concerns could overwhelm the improved earnings outlook. A subsequent completed regular-session close below the verified old breakout level of 18.75 would invalidate the continuation thesis.
daily_5d, oldest to newest, US Eastern regular sessions; columns=session_date|open|high|low|close|volume:
2026-10-22|18.43|18.69|17.90|17.94|1684630
2026-10-23|17.89|18.29|17.65|17.73|1216999
2026-10-26|17.86|17.96|17.27|17.37|977103
2026-10-27|17.49|17.72|17.12|17.66|1153012
2026-10-28|17.70|18.12|16.14|16.30|3233929
return_5d_pct=-12.318451, using sixth close=18.59 on 2026-10-21; high_20d=18.75; low_20d=16.08; avg_volume_20d=1637457.8 shares. History and pre-event close source=https://stockanalysis.com/stocks/prch/history/.
Return: -1.88%; alpha: -0.10%. - GOOGL — confidence 60%. Observed catalyst: Alphabet's October 28 earnings report combined rapid Cloud growth with a revenue miss and substantial cash consumption. Reuters reported negative quarterly free cash flow of approximately $9.07 billion. My bearish inference is that concern about financing AI investment and converting growth into cash can sustain pressure through tomorrow's checkpoint. The investment-related component of the EPS decline should not be mistaken for an equivalent deterioration in recurring operating earnings.
Before the event, October 22 and 23 closed near their lows, followed by another lower close on October 26. October 27 and 28 then recovered; October 28 closed at 59.5% of its range on 2.47 times completed-session average volume. That buying is meaningful counter-evidence, but the latest quote is well below the completed 20-session low of 316.60, indicating a post-event breakdown rather than an intact pre-event rebound.
Verified pre-event reference: October 28 close=324.89, regular-session end convention 2026-10-28T16:00-04:00. Latest_price=284.58 USD; quote_time=2026-10-29T12:59-04:00; regular-session snapshot, provider labelled real-time, cached when retrieved; source=https://stockanalysis.com/stocks/googl/. The pre-event-close-to-quote change is -12.4073%, spanning earnings and broader market trading. It is already observed, not the forecast decline. The exact authoritative start tick and elapsed-window return are unverified, so elapsed-window return=null.
PARTIAL SESSION, as of 2026-10-29T12:59-04:00: open=286.88; high_so_far=288.33; low_so_far=280.36; cumulative_volume=85059296 shares; source=https://stockanalysis.com/stocks/googl/. The quote has recovered from the low but remains below the open. No same-time volume baseline was verified, so relative intraday volume is unconfirmed.
Remaining-window expectation: approximately 2.5% additional downside from the latest accessible quote, with direction down for the authoritative window. Strongest counter-evidence is reported Cloud revenue growth of approximately 103% and the observed recovery from 280.36; the large initial decline may already discount the cash-flow disappointment. Panic selling can reverse into short-covering, particularly if investors emphasise operating growth over noncash investment losses. A sustained recovery above the verified partial-session high of 288.33 after the current quote would invalidate the immediate downside-continuation thesis.
daily_5d, oldest to newest, US Eastern regular sessions; columns=session_date|open|high|low|close|volume:
2026-10-22|330.46|331.43|322.00|324.22|27748122
2026-10-23|321.15|324.53|320.18|320.21|17569640
2026-10-26|320.89|321.48|316.60|316.80|23991837
2026-10-27|317.00|323.88|316.86|320.58|28633058
2026-10-28|323.77|326.41|322.66|324.89|63109029
return_5d_pct=-2.948381, using sixth close=334.76 on 2026-10-21; high_20d=359.40; low_20d=316.60; avg_volume_20d=25585349.15 shares. History and pre-event close source=https://stockanalysis.com/stocks/googl/history/.
Return: -1.81%; alpha: 3.59%. - LLY — confidence 64%. Observed catalyst: Lilly reported Q3 2026 revenue of $24.4 billion, up 39%, and raised the midpoint of full-year revenue guidance by $5.5 billion to a $91 billion to $92 billion range. The fundamental positive is stronger demand and a higher sales outlook, rather than price momentum alone. Qualifications matter: approximately $900 million of quarterly revenue benefits included milestones and rebate or discount adjustments, while acquired IPR&D charges complicated the EPS comparison.
Before the event, October 23 reversed much of October 22's advance and closed in the lower part of its range. The next three sessions produced modestly rising closes, with upper wicks showing overhead supply. October 28 recovered from an opening gap down and closed at 71.5% of its range, but volume was only 0.92 times the completed-session average. Today's opening above the completed 20-session high of 1321.21 provides new breakout evidence; the earlier daily bars alone did not confirm it.
Verified pre-event reference: October 28 close=1288.30, regular-session end convention 2026-10-28T16:00-04:00. Latest_price=1370.84 USD; quote_time=2026-10-29T12:56-04:00; regular-session snapshot, provider labelled real-time, cached when retrieved; source=https://stockanalysis.com/stocks/lly/history/. The pre-event-close-to-quote change is +6.4069%, spanning the report and subsequent market trading. That already-observed gain is excluded from the forecast.
PARTIAL SESSION, as of 2026-10-29T12:56-04:00: open=1332.73; high_so_far=1408.38; low_so_far=1330.00; cumulative_volume=2660033 shares; source=https://stockanalysis.com/stocks/lly/history/. The partial low remains above the old 20-session high, but the retreat from 1408.38 shows substantial upper-range rejection. No same-time volume baseline was verified, so unusual intraday volume cannot be established.
Remaining-window expectation: approximately 2.5% additional upside from the latest accessible quote, with direction up for the authoritative window. Exact elapsed-window return=null. Strongest counter-evidence is the sizeable retreat from today's high after a large initial earnings gain, together with temporary revenue benefits. Profit-taking or a reassessment of pricing, reimbursement and acquisition charges could reverse emotional continuation. A subsequent completed regular-session close below the verified prior 20-session high of 1321.21 would invalidate the breakout thesis.
daily_5d, oldest to newest, US Eastern regular sessions; columns=session_date|open|high|low|close|volume:
2026-10-22|1281.76|1321.21|1275.10|1306.69|2660304
2026-10-23|1304.15|1307.92|1257.62|1273.46|2277425
2026-10-26|1286.39|1296.90|1273.73|1279.14|2106876
2026-10-27|1273.50|1297.92|1272.07|1281.91|1849712
2026-10-28|1268.55|1299.43|1260.34|1288.30|2061427
return_5d_pct=1.381074, using sixth close=1270.75 on 2026-10-21; high_20d=1321.21; low_20d=1131.21; avg_volume_20d=2252314.1 shares. History and pre-event close source=https://stockanalysis.com/stocks/lly/history/.
Return: 3.86%; alpha: 5.64%.
How models have done in Event Gaps
ForecastNest's own settled Event Gaps runs at this horizon, up to and including this one. Members' own research is not counted. Event Gaps forecasts must have been created during regular US market hours to enter these statistics. Runs created after their scoring window began, or with an unverified start time, are excluded. The chart is the average alpha vs SPY 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 picks beat SPY, 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 picks with positive alpha | Avg alpha vs SPY |
|---|---|---|---|
| Grok 4.7 | 3 | 53.33% | +0.81% |
| ChatGPT | 1 | 60.00% | +0.50% |
| Gemini 1.5 Pro | 13 | 53.85% | +0.37% |
| Claude Opus 5.5 | 6 | 46.67% | +0.36% |
| Gemini 3.6 Flash | 1 | 60.00% | +0.29% |
| Perplexity | 23 | 49.57% | +0.17% |
| DeepSeek | 25 | 52.20% | +0.10% |
| GPT-6 | 18 | 43.33% | +0.07% |
| Grok 4.5 | 19 | 45.26% | -0.02% |
| Grok 4 | 1 | 40.00% | -0.09% |
| Claude Opus 5 | 1 | 40.00% | -0.25% |
| Gemini 3 Pro | 6 | 56.67% | -0.31% |
| Muse Spark 1.1 | 24 | 46.67% | -0.35% |
| Claude Sonnet 5.5 | 13 | 40.00% | -1.00% |
| GPT-5.6 Sol | 7 | 40.00% | -1.42% |
| Meta AI - Muse Spark | 1 | 40.00% | -1.84% |
| Gemini 3.6 Flash High | 1 | 40.00% | -2.26% |
| Kimi | 1 | 20.00% | -5.18% |
View settlement prices
| Symbol | Start | End | Source | Notes |
|---|---|---|---|---|
| GOOGL | 284.47000000 | 279.31000000 | https://finance.yahoo.com/quote/GOOGL/history/ | |
| INTC | 92.10500000 | 64.52990000 | https://finance.yahoo.com/quote/INTC/history/ | |
| LLY | 1370.00000000 | 1422.95000000 | https://finance.yahoo.com/quote/LLY/history/ | |
| NOW | 111.29000000 | 104.76500000 | https://finance.yahoo.com/quote/NOW/history/ | |
| PRCH | 20.53000000 | 20.14500000 | https://finance.yahoo.com/quote/PRCH/history/ | |
| SPY | 769.26500000 | 755.61000000 | https://finance.yahoo.com/quote/SPY/history/ |
Standard prompt event-gap-v13, SHA-256 ccc5e6e87a322d68e68111cccfe8a4d7b1d33e0df246a3f617826c87e8630178. The prompt is ForecastNest's own research and is not published; the fingerprint shows it was not changed after the fact.
Calculation: event-gap-v3-window-direction.