High Risk

Recent Performance — Last 5 Sessions

Today = vs our Oct 20 mark ($1,376.03). That mark was published to the website on Tuesday; Tuesday was not emailed or texted, so measured against Monday’s $1,431.36 — the last figure that actually reached you — the book is +3.0%. Live marks (Wed, 7:51 MDT, twenty-one minutes after the open, 9:51 Eastern); the daily move is measured against our Oct 20 (Tuesday) published mark. Tuesday was published to the website but was not emailed or texted, so the last figure that actually reached you is Monday’s — three bases today, and all three print. The direction matters, so read it before the numbers: Tuesday’s tape kept rising after we marked at 9:42 Eastern, so Tuesday’s official close landed above our mark. That makes today’s headline, which is measured against our mark, the more flattering of the two readings — your own screen measures from that official close and will look worse than our number. Both readings are printed, for the book and for every position. October MTD is measured from Wednesday Sep 30’s official month-end close. October MTD vs Sep 30 close ($1,733.06). Tuesday’s official close was $1,441.72 — 4.77% above our mark — so on the close basis today is +2.2%. Sessions before Jul 14 reflect the prior book; the current holdings were established at the Jul 14 open.

Baseline note — read this before the headline, because the headline overstates what has happened to you. Tuesday’s letter was never sent. It was built, audited and published to the website, but no email and no text went out. So the last number you actually hold for this book is Monday’s $1,431.36, not Tuesday’s $1,376.03. Three bases today, and they diverge sharply. Today’s +7.1% is measured against our Oct 20 mark — a mark taken on one of the worst mornings this book has had. Tuesday’s official close was $1,441.72, and from that close the book is +2.2%. Measured from the $1,431.36 you were last sent, it is +3.0%. A +7.1% headline next to a +3.0% move since you last heard from us is not a contradiction — it is the arithmetic of measuring from a bottom we marked and never mailed. The number to trust is +3.0%. October MTD is measured against the Sep 30 official month-end close of $1,733 for the current whole-share model. September is locked closed in the monthly table at +19.50%.

Position Scoreboard — Sorted by Portfolio Weight

Today = vs our Oct 20 mark, with the vs-Tuesday’s-close reading beneath it wherever the two differ. Since Fill = vs SOXL $179.62 / TQQQ $75.19 / MSTR $94.78 / BE $205.15 / NBIS $172.19. Weight = % of unit. Live marks (Wed, 7:51 MDT, twenty-one minutes after the open, 9:51 Eastern); the daily move is measured against our Oct 20 (Tuesday) published mark. Tuesday was published to the website but was not emailed or texted, so the last figure that actually reached you is Monday’s — three bases today, and all three print. The direction matters, so read it before the numbers: Tuesday’s tape kept rising after we marked at 9:42 Eastern, so Tuesday’s official close landed above our mark. That makes today’s headline, which is measured against our mark, the more flattering of the two readings — your own screen measures from that official close and will look worse than our number. Both readings are printed, for the book and for every position. October MTD is measured from Wednesday Sep 30’s official month-end close.

Portfolio Allocation by Category

Live Track Record — Since January 22, 2026 (Actual)

▲ Current month through today; July, August and September 2026 are locked closed; return measured from the Jan 22, 2026 entry. * Jan–Mar 2026 are estimated partial-period returns carried from our May 2026 updates; they predate our daily archive (begins 15 April 2026) and are shown as estimates, not as-published daily marks. Apr–Sep cells are locked closed from official month-end closes; close-basis restatement pending the anchor review. No backtests, no hypotheticals. Units are price-return; dividends are not accrued.

Calculate Your Position Size

High Risk is marked at $1,474 on the current whole-share model, and today the three bases diverge so widely that leading with the headline would mislead you. Against our Oct 20 mark the book is +7.1%. Against Tuesday’s official close of $1,441.72 it is +2.2%. And against the $1,431.36 you were last actually sent — because Tuesday’s letter was published to the website but not emailed or texted, which was our failure — it is +3.0%. Use that last one. The +7.1% is real arithmetic but it measures from a mark we took on one of the worst mornings this book has had and then never mailed, so it flatters us. October is -14.9% month to date and the year is still negative at -9.9%.

The demand question got a loud answer last night. The funding question got a worse one, and the funding question is the one that has been driving this book. Super Micro pre-announced September-quarter revenue of $18.5 to $19.5 billion against its own guidance of $14.5 to $15.5 billion — a beat of roughly a quarter on its own forecast, with gross margin guidance lifted to 12.0–12.5% — and SpaceX reported revenue up 161% to $14.58 billion with its AI segment at $7.56 billion against $813 million a year ago. Nobody can look at those two prints and argue the orders are not there. Six days ago this letter carried the opposite datapoint, Anthropic’s quarterly growth halving, and we said then that demand and financing were different risks deserving different reactions. That cuts both ways, and it cuts against us today: a strong demand quarter does not lower anybody’s cost of capital.

Here is the evidence that the thing which actually broke this book has not been fixed. CoreWeave’s 9.75% notes maturing in 2031 changed hands around 80.6 cents on the dollar on Tuesday, a yield of about 15.5% — against roughly 11.3% in mid-August, and essentially unchanged from the 15.8% we reported on this same bond on Monday. That is the AI build-out’s borrowing cost parked at its highs rather than mending — not a new high, but nowhere near repaired. And the equity agrees: on a morning when Super Micro is up 11% and the semiconductor index is up 1.1%, CoreWeave itself is down about 3.3%. The company carries roughly $51.6 billion of total debt and obligations. So the market is willing to pay more for servers today and no more for the credit of the people buying them. That distinction is the whole story of October for this page, and one good demand quarter does not close it.

Which brings us to the honest reading of our own numbers, because the leverage and the mark are both flattering us this morning. The semiconductor index is up about 1.1% and the Nasdaq Composite 0.7%. SOXL is +8.7% against our Tuesday mark, +2.8% against Tuesday’s close and TQQQ is +4.6% against our Tuesday mark, +1.8% against Tuesday’s close. A three-times daily-reset fund on a one percent index day should give roughly three percent, and against Tuesday’s close that is about what SOXL delivered; the far larger number against our own mark is the dip we marked, not performance. The arithmetic that has hurt all month still applies in reverse: these funds reset exposure every afternoon, so a sustained choppy decline costs far more than three times the index and needs a much larger rally to recover it. One strong session does not undo -14.9% for the month. SOXL remains -44.2% since entry.

Tomorrow is the date that matters most to this page, and we owe you a correction on half of it. BE is +7.6% against our Tuesday mark, +4.0% against Tuesday’s close — the strongest name in the book measured from Tuesday’s close, with MSTR strongest measured from our own mark — and Bloom Energy reports third-quarter results after the close on Thursday, October 22 — confirmed by the company itself in an October 12 release, with a 5:00pm Eastern call, so the reaction lands in Friday’s session. Analysts look for revenue near $1.06–1.07 billion, roughly double a year ago, and the options market has been pricing something like a 10–13% move on the print. Part of today’s strength in the name is plainly positioning ahead of that rather than news. Now the correction. On October 19 we told you Nebius reports the same Thursday morning, flagged then as carried by data providers rather than confirmed by the company. That expectation has not survived re-checking, and we checked it again this morning rather than repeat it on trust. We can no longer find a single source putting Nebius tomorrow. Its own investor-events page still reads ‘no upcoming events yet’; there is no filing; and the providers that did carry Thursday have moved into November — one now estimates November 10 and states plainly that the company has not confirmed, another projects a November 20–30 window. So the correct reading is that tomorrow is a one-name event, not two: Bloom alone, at 16% of this book rather than the 27% we implied on October 19. That is a smaller concentration than we last described to you, and we would rather say so than let the larger number stand because it happened to sound more careful. The two remain counterparties to one another, so whenever Nebius does report it still matters here. NBIS is +3.5% against our Tuesday mark, -2.4% against Tuesday’s close and -2.8% from our fill.

MSTR did something unremarkable today, and after the last fortnight that is worth saying out loud. Bitcoin was about $82,487 at Tuesday’s equity close and roughly $83,682 at our mark, up about 1.4%. MSTR is +8.8% against our Tuesday mark, +3.0% against Tuesday’s close — so measured on the same basis as the asset, from Tuesday’s close, the share moved in the same direction as the bitcoin it holds and about twice as far. The notable part is the direction: for most of this month it has moved against the asset rather than with it. Compare that with last Friday’s session, three before this one, when it fell 9.6% on a day bitcoin fell about 1%, and you can see the index worry is quiet today rather than resolved. Its weight is 39.9% of this unit against 39.3% on Tuesday — weight moving through price, not through any trade. The dated facts are unchanged: the affected-company list publishes November 11, any exclusion takes effect only after the close on November 30, and an existing member is removed only if it fails the screen in each of its two most recent annual filing periods. We hold four shares, the review of trimming to three is closed, and no trade was made.

Bottom line: High Risk is at $1,474 a unit. All three readings, worst first rather than best: +3.0% against the $1,431.36 you were last sent, +2.2% against Tuesday’s official close, and +7.1% against our own Oct 20 mark — with October -14.9% and the year -9.9%. One position flip sign between the two bases today. NBIS is +3.5% against our Tuesday mark, -2.4% against Tuesday’s close. Nothing here is a change of view and nothing was traded. The structure is unchanged and worth seeing written down: 33% in two 3x daily-reset funds, 40% in a bitcoin treasury company carrying a dated index risk, and 27% in two AI-infrastructure counterparties, of which only Bloom reports Thursday. What we are watching, in order of how much it can move this page: Bloom Energy after Thursday’s close on October 22, which on today’s evidence is tomorrow’s only report here, with Nebius now estimated in November and still not company-confirmed; CoreWeave’s bond yield, which is the cleanest read on whether this month’s actual problem is easing and which has stayed parked at its highs rather than mending; whether a new AI issuer can price a deal at all after Seoul; the MSCI affected-company list on November 11 and the November 30 effective date, with MSTR at 39.9%; the Federal Reserve on October 27–28; and the midterms on November 3.

Portfolio Builder | Market Update | Wednesday, October 21, 2026

Trade Ledger — Entry & Exit History

Every position we open and close, newest first. Realized P/L on closes is marked against the original entry price.

Entry prices verified against the Jan 22, 2026 cost basis. “Open” = still held today. An entry marked “Closed <date>” was exited in that rotation — its realized result is shown on the CLOSE row for that date. This ledger builds our public track record going forward.