August 2026 · Interim Note
AMD and a Leveraged Nebius Bet Add $293 to the Book, Even With Palo Alto’s Worst Day in a While
The S&P 500 rose 0.32% Tuesday, which again tells you almost nothing about my book. AMD added $112, NBIL (a 2x leveraged Nebius fund) added $69, and Meta added $51 — against Palo Alto Networks (−$55) and Sandisk (−$25) pulling the other way. Net impact: +$293.43. 21 of 32 positions closed higher. No trades.
Where the $293 came from
Figure 1 · Dollar attribution, ranked
Eight winners covered Palo Alto and Sandisk with room to spare
Share counts × the change in each position’s closing price, August 24 to August 25, 2026. Every position moving more than $1 is included; the remaining 16 positions netted to a small positive not shown individually. Source: Massive Market Data closes, applied to my own share counts.
This is a different shape from most recent sessions: instead of one dominant winner, three separate positions each cleared $45, and a fourth — NBIL, my 30-share position in GraniteShares’ 2x leveraged Nebius fund — had its best single day since I bought it. Palo Alto Networks fell 3.13% to $339.90, its worst session in weeks, and I don’t have a dated headline explaining it; logging it as unexplained rather than guessing, same as today’s News article does for CrowdStrike.
| Position | Close | Day | Book impact | vs. my cost |
|---|---|---|---|---|
| AMD | $479.18 | +4.91% | +$112.18 | +140.39% |
| GraniteShares 2x Nebius (NBIL) | $24.65 | +10.34% | +$69.30 | +167.06% |
| Meta Platforms (META) | $570.05 | +1.97% | +$51.07 | +143.31% |
| Micron Technology (MU) | $932.97 | +2.48% | +$45.08 | +186.42% |
| Space Exploration Technologies (SPCX) | $137.95 | +2.19% | +$23.60 | +2.19% |
| Nebius Group (NBIS) | $221.97 | +5.24% | +$22.12 | +146.29% |
| Palo Alto Networks (PANW) | $339.90 | −3.13% | −$55.00 | +111.99% |
| Sandisk Corp (SNDK) | $1,480.77 | −0.83% | −$24.70 | +206.44% |
| CrowdStrike (CRWD) | $185.38 | −2.78% | −$10.60 | +95.63% |
The position keeping the most weight right now
Nebius rose 5.24% to $221.97 on continued Motley Fool coverage of its Nvidia and Goldman Sachs backing — institutional validation, not a new event, but the kind of thing that keeps a stock’s momentum going. I hold it two ways: 2 shares of NBIS directly, and 30 shares of NBIL, a 2x-leveraged daily fund tracking it. NBIL rose 10.34% Tuesday — almost exactly double NBIS’s move, which is the fund doing what it says on the label. I did not add to either position, and I want to be honest with myself about why NBIL deserves a closer look than a routine winner does.
What “2x daily” actually means in dollars
A 2x leveraged ETF like NBIL is built to deliver roughly twice NBIS’s daily return, before fees — not twice its return over any longer period, because daily resetting means gains and losses compound in a way that erodes value in a choppy, sideways market even if the underlying stock ends up flat. That is a structural fact about how the fund is built, not a prediction about what NBIS will do next. Below is the arithmetic for what a range of NBIS daily moves would mean for my 30-share NBIL position at Tuesday’s $24.65 close, plus how Tuesday’s actual numbers compared to the theoretical 2x.
| If NBIS moves… | NBIL, theoretical 2x | Dollar impact, 30 shares |
|---|---|---|
| −10% in one session | −20% | −$147.90 |
| −5% in one session | −10% | −$73.95 |
| Flat | 0% | $0.00 |
| +5% in one session | +10% | +$73.95 |
| +10% in one session | +20% | +$147.90 |
| Tuesday, actual: NBIS +5.24% | Theoretical: +10.48% Actual: +10.34% | +$69.30 |
Assumptions, stated plainly. The scenario rows ignore the fund’s expense ratio and any tracking error, and hold NBIL’s price fixed at Tuesday’s $24.65 close rather than compounding from a moving base — a simplification, not a forecast of what NBIS will do tomorrow. The bottom row is not a scenario; it is what actually happened, included specifically to show the gap between the clean 2x math (+10.48%) and the real result (+10.34%), a 14-basis-point difference that is fees and tracking cost, not a modeling error on my part. The wider point this table exists to make: a 30-share position in a 2x fund carries roughly the dollar risk of a 60-share position in the underlying, and I should size it with that in mind rather than treating “30 shares of NBIL” as a small position because the share count is small.
AMD’s move is worth a separate note. It rose 4.91% to $479.18 after an analyst doubled his price target to $1,250, per today’s News article — a number I am not treating as a plan, since a single analyst’s target is not a reason to change position sizing on its own. My 5-share position is now up 140.39% against my average cost, which is the kind of gain that makes me want to double-check I’m not holding it out of inertia rather than conviction. For now, the answer is conviction: nothing about the fundamentals changed today, so nothing about the position does either.
No trades today
The Blackout Rule and Holding Period exist for exactly this kind of session. AMD, NBIL, and Meta are all up sharply and I did not trim any of them; Palo Alto Networks and Sandisk both fell and I did not add to either. Same rule, applied in both directions. I made no trades.