August 2026 · Interim Note
Monday Reversed Friday: +$265.28, and SanDisk Alone Did 61.6% of the Work
The book rose $265.28 Monday, recovering a little more than half of Friday’s $430.79 loss. Sandisk did most of it by itself: 2 shares, up $81.72 apiece, for a $163.44 contribution — 61.6% of the entire day’s gain. Palo Alto Networks (+$52.70), Micron (+$51.74), AMD (+$25.70) and CrowdStrike (+$25.20) rounded out the winners; Meta (−$26.30) and Alphabet (−$21.72) were the largest drags. 17 of 32 positions closed higher. No trades.
Where the $265 came from
Figure 1 · Dollar book impact by position
One position did more work than the next five combined
Share counts × the change in each position’s closing price, August 28 to August 31, 2026. The ten largest single-name movers are shown individually; the remaining 22 positions are bucketed as “Other 22” and net to +$15.10. Bars are independent day-over-day columns, not a cumulative waterfall. Source: Massive Market Data closes, applied to my own share counts.
This is close to the mirror image of Friday: one name did most of the work, rather than one name doing most of the damage. Sandisk rose 5.50% to $1,566.70, per a same-day Zacks Rank #1 upgrade citing a 17.1% rise in 60-day consensus earnings estimates — today’s News article covers the mechanism in full, alongside CrowdStrike’s guidance-driven rally on the other side of the tape. My 2-share Sandisk position is now up 224.23% against a $483.21 average cost — the single best return in the book, and the position I have thought hardest about trimming for exactly that reason.
| Position | Close | Day | Book impact | vs. my cost |
|---|---|---|---|---|
| Sandisk (SNDK) | $1,566.70 | +5.50% | +$163.44 | +224.23% |
| Palo Alto Networks (PANW) | $382.13 | +2.84% | +$52.70 | +138.35% |
| Micron (MU) | $958.73 | +2.77% | +$51.74 | +194.24% |
| AMD | $470.72 | +1.10% | +$25.70 | +136.14% |
| CrowdStrike (CRWD) | $231.00 | +5.77% | +$25.20 | +143.77% |
| ServiceNow (NOW) | $147.99 | +2.27% | +$19.68 | +74.60% |
| WhiteFiber (WYFI) | $18.92 | +6.59% | +$16.38 | +41.94% |
| Estée Lauder (EL) | $102.30 | −1.05% | −$19.62 | +48.00% |
| Cheesecake Factory (CAKE) | $109.49 | −2.30% | −$20.64 | +141.17% |
| Alphabet (GOOGL) | $339.35 | −2.09% | −$21.72 | +23.59% |
| Meta Platforms (META) | $572.34 | −0.98% | −$26.30 | +144.26% |
The position I keep coming back to
A same-day Motley Fool piece noted that Appaloosa Management’s David Tepper exited his entire Sandisk position in Q2 2026, according to the fund’s 13F filing, after the stock had run up more than 3,000% over a trailing year — the article frames it as profit-taking on a historically cyclical memory name at a stretched valuation, not a fundamentals call. I hold 2 shares against a $483.21 average cost, up 224.23% as of Monday’s close. That is a genuinely large unrealized gain sitting in a name whose entire investment case rests on a NAND-pricing cycle that has, historically, always turned. I have not trimmed it. I am not trimming it today either — the Holding Period exists so that a single strong day, or a single billionaire’s 13F, doesn’t force a decision I haven’t actually thought through.
No trades today
The Blackout Rule and Holding Period exist for exactly this kind of session. A one-day 61.6%-from-one-name gain is not new information about any company’s fundamentals, any more than Friday’s loss was. I made no trades.