AI-stock damage snapped back 70% to 75% within two days, with VIX peaking at 20.66 on the 29th, QQQ reclaiming 200-day support near 662 and reaching 722, and the S&P 500 printing another all-time high at about a 20 PE. The core thesis is that the market keeps rewarding the strongest AI, chip, and crypto names, while weaker names and crowded narratives keep lagging or getting sold into. BTC stayed stuck around 64K, ETH is still below its 200-day at 2085, SOL remains weak, and the video keeps rotating between relative-value trades, option setups, and a few high-conviction catalysts. Later segments broaden that into NVDA, MU, MRVL, GOOG, AMD, TSLA, copper, and several smaller crypto and small-cap names, all framed through exact levels, kill zones, and breakout targets.
AI-stock losses recovered 70% to 75% of the damage in two days after one of the worst Julies ever for AI names.
VIX peaked at 20.66 on the 29th, then printed a weird one-day spike straight to the 200-day for no reason, treated as a textbook sold-off signal.
QQQ was called the bottom last week: it sold off to the 200 EMA, nearly touched it within an hour timeframe, hit the 200-day around 662, then ripped back above 720; QQQ was at 722 in the segment.
The S&P 500 broke another all-time high on the day of the segment and was described as trading at a PE of 20.
BTC was still around 64K and change, stuck near 64,000 since February; 59K was named strong support, with 62K to 65K the expected chop range until ETF buying or Saylor buying hits in force.
ETH had a 1500 killbox double-buy signal, was creeping up, and sat about 8% below the 200-day moving average at 2085; once above the 200-day, it was expected to hold and potentially run to 4,000, 4,300, or 5,000.
SOL was described as having completely sucked wind, with the 200-day at 85 and falling all the time because price stayed low.
MSTR was flat while BTC rose, framed as MSTR ATM-ing since July 2 for nearly five weeks to raise cash and give confidence that STRC has legs; STRC was at 94, after the prior drop to 72 and buys at 74, 75, and 73.
TSLA had support at 298 to 297 on the 29th, bounced to 327, then slipped again; a big unlock was due the next day, with the post-unlock float expected to be 1.6x the IPO float and more sell pressure than buy pressure.
NVDA recovered 190 support and rose 3.43% on the day; its market cap was shown at 5.31T, described as more than five trillion and close to 5T, while valuation was called an all-time low PE and "never been cheaper."
MU was said to have crashed to 739 six days earlier, offered a great kill zone around 721, then surged to 920 and met resistance at level five, with level six next.
MRVL bounced clean off level three, ran to 219, sat near 210, and pulled back a little; it was also described as up 36% to 36.7% in two days, with a great kill zone at 164 and a call for 1,000 within a couple of years.
GOOG was hit by a rumor that "Deis Sabis" was leaving and taking 4 or 5 AI research engineers, opening at 382 before tanking to 355; over the past year GOOG was said to have smoked MSFT by over 100%, with GOOG viewed as undervalued at 450 and MSFT as fairly valued.
AMD broke down through 480 to 420 last week, then recovered after yesterday's earnings scare about higher-than-expected capex; the setup was framed as a breakout back toward a new all-time high at 500 because AI capex spending is not slowing down.
Copper was called headed to a new all-time high and then printed one within a week up to 68,278; options and LEAPS were held because copper demand was tied to AI and electrification.
SUI was compared with SOL as a better-adoption L1 for AI agents: 800,000 users vs 3.5 million on SOL, 7 million daily transactions vs 76 million on SOL, 250% better daily active users, and basically no fees; SOL was called the cheapest horse on fully diluted market cap per daily active user.
HYPE traded around 57 to 60, made a new all-time high up to 77, and the trade plan was to buy at 50 and sell at 75.
UMAC was pitched as a drone play bought at 19 about six days ago, with strong growth, mostly profitable quarterly results except Q4 2025, and room to bounce from 8 back toward 15 near its prior high.
GLW was renamed Corning Incorporated, criticized for glassmaking and a PE of 73; MU was contrasted as available at a PE of five.
A layered options trade on SL used $17.50 LEAPS and short $20 puts for a $1 net debit, turning a $19 stock into a $9/share position by the next night; ALAB calls sold at the 500 strike for about $60 fell to three or less, and backtests showed 88.82% ATR-model correctness and 80.72% trend-model correctness.