Chapters
00:00 Intro
01:23 QQQs Break the Descending Wedge
04:59 $MSFT & $AMZN Earnings: AI ROI Confirmed
06:22 The Leopold Blowup & Why Positioning Matters
11:08 $META's Gap Fill & GP's December Calls
14:06 The Calendar Flip & a 25x SPX Call
17:46 $DELL's "SanDisk Moment" & $HPE
20:06 $IGV, $NOW & $ORCL's Reversal
24:18 V-Bottom or Retest?
29:47 SPX 7741, Metals & Amazon Absorbing Bezos
33:37 Crypto, $CRWV & $RDDT
37:11 Contraction to Expansion & Rate Cuts
40:15 Money Left on the Table: Trading Psychology
48:41 Memory Earnings & Final Picks: $DDOG, $IBM & $S
Market Hours #197 went out live on Tuesday, August 4, with Ben, Nick and GP, two days into August and right after Microsoft and Amazon earnings had flipped the tape. The QQQs had just broken out of a descending wedge, Microsoft was up about 30% in four days, and the crew argued that positioning, not fundamentals, drove both the selloff and the snapback. From there it was Meta's gap fill, Dell, software and Oracle, whether this was a V-shaped bottom, the SPX approaching GP's 7741, and an honest talk about leaving money on the table.
Nick's setups, levels and trades
Nick said the move played out the way they'd drawn it up: when price goes back into a range, it's easy to get chopped up on the first pull-in. He stacked the indices and found the SPY and ES looked better than the QQQs and NQ, which had put in "a really pretty looking descending wedge right off the 200-day." He had expected a little more sideways chop just under 700 before it popped, but called the day a breakout of the wedge and a 50-day pocket pivot, on volume already close to the daily average. In his view Friday's bar — the QQQs taking out Thursday's high — was "your face rip signal." The SPY, IWM and Dow all looked phenomenal, and the IWM had coiled right into its 50-day.
He wouldn't call the indices a V-bottom — to him they chopped sideways in a descending wedge and dropped the floor before ripping out — though individual names did V out. He expects it to look a lot like April through June: a melt-up in the indices, respecting the 10-day at first and then seeing how they behave as the 20-day catches up. "It sure looks to me like we're just getting started." He also pointed out the QQQs were up about 9% from the bottom versus 5.8% for the SPY, even though the QQQs were the only index not yet at all-time highs.
On Dell, he said you aren't late: with the 10-, 20- and 50-day converging, you could risk the gap or down to about 405, and HPE looks just as good. His advice was to pay more attention to what isn't participating than to what is. On Meta and Microsoft, a push from here would invalidate the bigger rounding-top look; Meta's spikes toward 500 came on less and less pressure, and he read the fast gap fills after earnings in Meta and Reddit as a sign buyers "just wanted more." He also pointed out the reversal in CoreWeave.
On the upcoming memory reports (Micron, SanDisk, STX), how he feels depends on the market going in: if things rip hard into them, he'd lean toward taking profits after the reports, while watching for Meta-style action where they get smacked and bought right back. His pick was Datadog, which he thinks "looks unbelievably good" — it held up well after its rip off the lows near 100, coiled into the 50-day, gave back only 20–25 points from about 270, and was breaking out on the day. He also shared that he'd had to step away from journaling for a while because it turned into a shame spiral, and that he and John were doing a social-media cleanse for the rest of August, working on replacing judgment with curiosity.
Ben's setups, levels and trades
Ben said the real story was Microsoft and Amazon keeping guidance and CapEx up while showing big growth in Azure and AWS, the pipelines for the AI buildout. He recalled saying months earlier that the last time Amazon made a bet this big was the AWS buildout, and it took two quarters for the stock to recover — "history doesn't always repeat, but it sometimes rhymes." On the Leopold fund blowup, he joked that it might be a psy-op, and pointed out the fund still grew from about $220 million to about $10 billion.
He pushed the crew on whether a check-back was coming after a V-shaped rally. His own bigger-picture view: multiples in software names like ServiceNow and Intuit have been compressed, and "to every contraction, there has to be an expansion." With the Bank of Japan news and possible rate cuts ahead, he thinks there are 12 to 18 months of this kind of environment, closer to a 2019-style market. He sees the start of rate cuts as a signal that it's time to take profits — "you got to ride it till the music stops."
He admitted his calculator hurt: holding all of his Dell from a roughly $30 cost basis would have been worth about $448,000 more, but he said he never had that money to begin with and tries not to let it bother him. His closing words: clean your desk, the only way is through, and don't forget about SentinelOne.
GP's setups, levels and trades
GP's thesis was that "positioning is all that matters." Microsoft went from a $98 billion run rate with 37% growth to $100 billion and 41% — "nothing really changed" except that an over-leveraged fund that was long AI names and short mega caps and software got blown out. Once that seller was gone, Microsoft put in 30% in four days. He said Tesla is in the same boat, after record sales that the market ignored. On Oracle, forced selling from the credit downgrades and CDS blowout had to finish before the stock could launch. He noted Larry Ellison has survived every crisis since the savings-and-loan era, and said the risk-reward made sense around 116–117 at multi-year lows. Microsoft went from the 350 pivot to about 499 in a few days.
On Meta, he compared the post-earnings red candle to the tariff beatdown: a nasty candle to test the bulls, then a takeoff. It filled its gap in three days, and he bought December 700 calls, risking about 20% of the capital before he'd stop out. He thinks Dell is in its "SanDisk moment" and still cheap, and expects ServiceNow to get past 165 quickly. He said he bought IGV August 140 calls on the day.
On the market, he expected memory and the Q2 momentum names (except Dell) to consolidate. He thinks AMD's earnings that night may mark the top of their mean reversion, while the money flows into the mega caps and the QQQs could keep lagging the SPY. The SPX was approaching his 7741 level, a 161.8% extension that often overshoots but is a major inflection point, so he'd like to see consolidation there. He also said the metals had all broken out from the bases they built around last Oktoberfest's levels, comparing it to gold's last cycle, where a big move was followed by consolidation and then a wave-three expansion. Amazon was absorbing Bezos' roughly $4 billion stock sale while down only about 2%. He also flagged declining volume in the memory names, using AMD as an example, as it pushed back into resistance.
He was candid about a trade that stung: he bought 100 SPX 7675 calls for about 60 cents the day before, sold them around $15, and watched them run past $60–75 — about $600,000 left on the table by his count. He said the process was right — get the capital out, then manage a big winner — and he'd make the same trade again. The lesson was to let the feelings sit, not try to revenge-short, and come back fresh. His pick was IBM, which has an island-reversal look if it gaps up and doesn't fill. His last word: "Be kind to yourselves."
Subscribe for the weekly live show at YouTube.com/@MarketHoursPod.
This is not financial advice. Market Hours is for education and entertainment only. Always do your own research.
Follow Market Hours: YouTube @MarketHoursPod · X @markethourspod · TikTok @markethourspod · StockTwits @markethourspod