Portfolio Plan Portfolio 3.5M Portfolio 100K Portfolio 2M Crypto Convexity Recommendations Crash Liquidation

PORTFOLIO 2M, THE UPSIDE + BALLAST COMPLEMENT TO PORTFOLIO 100K: 11 FORTRESSES, $2.14M ACCOUNT

Chain marks 2026-09-22 SGT pre-market (fortress_v28 CSV replay, no live IBKR; Jan-2029 LEAPS; run dir PORTFOLIO_2M) · horizon to LEAPS expiry ~2.3 yr · sibling: 3-account plan

SNAPSHOT

The job: sit beside the all-capped income book and cover what it gives away. That book writes every contract, so on a big run its calls go underwater, new premium stops, and its upside is capped near one cycle. This book holds the 11 names from the 3.5M list the income book does not own. About one third of the max loss is written, all of it on the ballast sleeve (LLY, LMT, JPM, AMZN), whose calls stay out of the money when the income book runs; the other two thirds, every upside name, runs uncapped. ML $1,985,349 = 0.93x cash (the no-liquidation calibration). Sizing: the 3.5M page's approved weights for these names, scaled to $2.0M, no name above 15% (GLD's excess spread across its sleeve), whole contracts from fortress_v28.

Invested / Max Loss

$1,197,649

ML $1,985,349 = 0.93x cash · NLV $2,140,000

Notional Exposure

$4.20M

2.0x on the account

Structure P&L @ Expiry

$1,068,332

all legs at +10%/yr drift, ~2.3 yr

Idle cash

$942,351

$2,654/mo @ 3.38% (bench 3.88% −0.5%)

Hedge rent / yr

$125,275

billed on ALL contracts in both scenarios

Liquidation Crossover

none (survives >-95%)

ML ≤ cash · crash-vol, TIMS-race

Written (capped)

38% of max loss

LLY, LMT, JPM, AMZN fully written · 41 contracts

Free to run (uncapped)

62% of max loss

GLD, XBI, CAT, AMD, SPCX, IREN, SNDK · 101 contracts

Uniform +50% rally, structure gain

$1,409,864

every name +50% · uncapped ceiling $2,278,930

HOW IT COMPLEMENTS THE INCOME BOOK (the scenario you asked about)

Melt-up: every income-book name +30%. Its calls go underwater, so new premium stops while the hedge rent keeps billing; its structure gain is capped near one cycle. Each 2M name moves by its measured beta to the income book (table below), and a written 2M call only keeps paying if its name stays inside one cycle (~+8%). Selloff: every income-book name −30% with crash vol; 2M names move by the same betas. This book keeps most of its income through the melt-up because its written calls sit on names that do not run with the income book.

Book	Income / mo (normal)	Income / mo income book +30% melt-up	Structure gain in that melt-up	MTM in an income-book −30% selloff (crash vol)
Portfolio 100K (income book) $1,600,000 account	$106,551	-$6,527 -6% of normal	$241,858 +15% of its NLV	-$709,009 -44% of its NLV
Portfolio 2M $2,140,000 account	$24,477	$16,809 69% of normal	$643,766 +30% of its NLV	-$552,280 -26% of its NLV
BOTH BOOKS $3,740,000 account	$131,029	$10,282 8% of normal	$885,625 +24% of its NLV	-$1,261,288 -34% of its NLV

WHY THESE 11 NAMES, AND WHICH ONES CARRY THE CAPS

Measured against the income book's own daily returns (180 trading days of Yahoo closes, the income book's positions weighted by max loss). Ballast names barely move with it on its best or worst days, so calls written on them stay out of the money when it runs, and they cushion it when it falls. Upside names run with it (SNDK, IREN, AMD, SPCX most), which is exactly the upside the income book caps away, so they are left uncapped. The four ballast names are the fully written third.

Name	Role	Beta to income book	Corr	Income book's 10 best days	Income book's 10 worst days	Implied move in its +30% melt-up	History
LLY	Ballast (fully written)	-0.02	-0.02	+0.1%	+0.7%	-1%	178 days
LMT	Ballast (fully written)	+0.00	+0.00	-0.1%	-0.1%	+0%	178 days
JPM	Ballast (fully written)	+0.20	+0.28	+1.1%	-0.8%	+6%	178 days
AMZN	Ballast (fully written)	+0.39	+0.35	+0.7%	-1.9%	+12%	178 days
XBI	Upside (uncapped)	+0.42	+0.44	+2.2%	-1.7%	+13%	178 days
GLD	Upside (uncapped)	+0.43	+0.42	+1.8%	-2.3%	+13%	178 days
CAT	Upside (uncapped)	+0.57	+0.43	+2.6%	-3.6%	+17%	178 days
SPCX	Upside (uncapped)	+1.05	+0.37	+4.2%	-0.6%	+32%	69 days, short history
AMD	Upside (uncapped)	+1.34	+0.57	+4.8%	-7.4%	+40%	178 days
SNDK	Upside (uncapped)	+1.66	+0.47	+6.8%	-9.3%	+50%	178 days
IREN	Upside (uncapped)	+1.79	+0.52	+4.9%	-10.4%	+54%	178 days

BOOK COMPOSITION BY THEME

Exposure (notional)

$4.20M notl · $0.94M cash

Capital (NLV)

$2.14M NLV · 44% cash

Semis AI & Software Crypto & Fintech Space Defense Industrials Healthcare Financials Energy Gold Cash

Left: market exposure (leveraged notional by theme). Right: the same dollars un-leveraged: deployed invested capital by theme plus idle cash, summing to the account NLV, so the grey Cash slice is its TRUE share of capital (44% of the account, earning ~$2,654/mo at 3.38%).

MARK-TO-MARKET vs DRAWDOWN (the book alone)

Every underlying haircut the same %, crash-vol adjusted, structure legs only. Crossover none (survives >-95%): paper drawdowns, not liquidation events.

Underlying drawdown	0%	-10%	-20%	-30%	-40%	-50%
Portfolio 2M	$0 0% NLV	-$425,228 -20% NLV · 2.0x	-$757,326 -35% NLV · 1.8x	-$1,010,933 -47% NLV · 1.6x	-$1,205,100 -56% NLV · 1.4x	-$1,353,122 -63% NLV · 1.3x

The x-multiplier is the NLV move divided by the underlying move (effective downside leverage). It falls as the drawdown deepens (2.0x at −10% → 1.3x at −50%): the hedge puts gain and the structure self-de-levers, which is the convexity that bends the curve flat.

MARK-TO-MARKET vs RALLY: the upside this book exists to keep

Every name moved the same %, ballast names included. Red = this book (ballast written, upside free), pale = everything uncapped. The income-book melt-up, where the ballast names stand still, is in the complement table above.

▬ Portfolio 2M (ballast written) ▬ Uncapped ceiling

Underlying rally	+10%	+25%	+50%	+75%	+100%
Portfolio 2M	$418,151 +20% NLV · 2.0x	$796,243 +37% NLV · 1.5x	$1,409,864 +66% NLV · 1.3x	$2,007,299 +94% NLV · 1.3x	$2,595,118 +121% NLV · 1.2x
Uncapped ceiling	$460,017 +21% NLV · 2.1x	$1,152,176 +54% NLV · 2.2x	$2,278,930 +106% NLV · 2.1x	$3,376,921 +158% NLV · 2.1x	$4,457,346 +208% NLV · 2.1x

Cap-aware: the four ballast names are fully written, so their structure gain stops near one CC cycle (~+8%); the seven upside names run free. At +50% on every name the book keeps $1,409,864 of structure gain against an uncapped ceiling of $2,278,930 (+100%: $2,595,118 vs $4,457,346). The gap is almost entirely the four written ballast names, which carry the most upside per dollar of max loss; that is the price of their income, accepted because they are the names least likely to be running when the income book is.

THE BOOK: 11 FORTRESSES, THE WRITTEN THIRD ON BALLAST

LLY, LMT, JPM and AMZN are fully written; GLD, XBI, CAT, AMD, SPCX, IREN and SNDK run free (their rows show the hedge rent they still pay, with no premium against it). Net income = gross CC on the written contracts x each name's graded keep rate, minus the hedge rent on every contract. The Max Loss column shows each name against its budget.

Portfolio 2M: Income / Month (CC + cash)

$24,477

CC $21,823 + cash int $2,654 · 41 written / 101 uncapped

Portfolio 2M: Total P&L at Expiry

$1,677,716

structure gains + net income · 34%/yr on cash

Income ROI on cash

14%/yr

on Invested 25% · on Max Loss 15%

Hedge coverage

3.1x

kept CC $387,152/yr vs rent $125,275/yr

Ticker (marks)	Type	Contracts (inc / unc)	Spot	Spot @ Expiry (+10%/yr)	LC / SP / HP	Invested	Max Loss (vs budget)	Notional	Lev	Net Δ/sh (ex-CC)	Structure P&L @ Expiry	Net Income / Mo	Net Income to Expiry	Total P&L	ROI/yr IC	ROI/yr ML	ROI/yr Notl
LLY 09-22	BALANCED rented hedge	5 all capped	1155.99	1,443.27	870 / 1150 / 920	$142,650	$257,650 budget $261,000 · 99%	$577,995	4.05x	1.09	$143,986	$9,384 keep 68%	$262,039	$406,025	122%	68%	30%
LMT 09-22	BALANCED rented hedge	10 all capped	537.24	670.75	440 / 520 / 430	$95,050	$185,050 budget $184,000 · 101%	$537,240	5.65x	1.08	$135,702	$6,095 keep 68%	$170,209	$305,911	138%	71%	24%
AMZN 09-22	BALANCED rented hedge	12 all capped	260.30	324.99	195 / 260 / 215	$76,800	$130,800 budget $135,000 · 97%	$312,360	4.07x	1.08	$79,186	$6,200 keep 73%	$173,129	$252,315	141%	83%	35%
AMD 09-22	BALANCED rented hedge	4 0 inc / 4 unc	609.00	760.35	270 / 420 / 350	$135,000	$163,000 budget $140,000 · 116%	$243,600	1.80x	1.07	$61,138	-$365 keep 76%	-$10,192	$50,946	16%	13%	9%
SPCX 09-22	BALANCED rented hedge	14 0 inc / 14 unc	152.49	190.39	95 / 130 / 100	$78,904	$120,904 budget $117,000 · 103%	$213,486	2.71x	1.05	$54,637	-$615 keep 74%	-$17,164	$37,472	20%	13%	8%
IREN 09-22	BALANCED rented hedge	32 0 inc / 32 unc	46.95	58.62	23 / 33 / 20	$75,296	$116,896 budget $117,000 · 100%	$150,256	2.00x	1.03	$38,701	-$406 keep 54%	-$11,344	$27,357	16%	10%	8%
SNDK 09-22	PIONEER rented hedge	1 0 inc / 1 unc	1766.51	2,205.52	1400 / 1900 / 910	$23,230	$122,230 budget $175,000 · 70%	$176,651	7.60x	1.07	$57,322	-$378 keep 70%	-$10,561	$46,760	86%	16%	11%
JPM 09-22	BALANCED rented hedge	14 all capped	352.25	439.79	290 / 350 / 285	$86,660	$177,660 budget $174,000 · 102%	$493,150	5.69x	1.12	$123,045	$5,267 keep 68%	$147,079	$270,124	134%	65%	24%
GLD 09-22	BALANCED rented hedge	16 0 inc / 16 unc	395.87	494.24	265 / 395 / 340	$209,200	$297,200 budget $300,000 · 99%	$633,384	3.03x	1.22	$157,590	-$979 keep 74%	-$27,335	$130,254	27%	19%	9%
XBI 09-22	BALANCED rented hedge	29 0 inc / 29 unc	158.30	197.64	115 / 158 / 129	$124,584	$208,684 budget $210,000 · 99%	$459,070	3.68x	1.14	$115,072	-$1,480 keep 68%	-$41,335	$73,737	25%	15%	7%
CAT 09-22	BALANCED rented hedge	5 0 inc / 5 unc	812.53	1,014.46	510 / 720 / 610	$150,275	$205,275 budget $187,000 · 110%	$406,265	2.70x	1.10	$101,953	-$900 keep 68%	-$25,139	$76,814	22%	16%	8%
PORTFOLIO 2M (11 fortresses)		142 41 inc / 101 unc				$1,197,649	$1,985,349 budget $2,000,000 · 99%	$4,203,457	—	—	$1,068,332	$21,823	$609,384	$1,677,716	60%	36%	17%

STRESS BATTERY vs THE MARGIN & LIQUIDATION FIELD GUIDE (module_risk_cockpit, crash-vol + skew, 1% steps to −95%, run 2026-09-25)

ML-to-cash: ML $1,985,349 / $2,140,000 = 0.93x; cash−ML floor $154,651. Worst simulated path: NLV $614,795 (at -94%).

Attack	Portfolio 2M crossover
Base, flat vol	NONE
Crash-vol ramp (vol ~2x at -30%, skew-aware)	NONE
Ramp + house margin hike x1.6	NONE
Ramp + house margin hike x2.0	NONE
Ramp + house margin hike x3.0	NONE
Ramp + GLD decorrelation fail (beta forced to 1.0)	NONE
t+12mo decay (LEAPS aged 1yr, hedges rolled) + ramp + hike x2	NONE
~$100k PM-floor race (NLV below the portfolio-margin floor)	not within −60%

Verdict (this book alone): on every price path to −95%, under crash vol, a tripled house requirement, a GLD decorrelation failure and twelve months of decay, the account cannot be margin-called (8/8 attacks: NONE).

If both books share ONE account: ML $3,467,539 on $3,740,000 of cash = 0.93x; crossover NONE past −95% (crash-vol ramp, same engine).

Hedge rent: $125,275/yr, billed on every contract, written or not; kept premium covers it 3.1x. That thin coverage is the price of leaving two thirds of the book uncapped. The hedges must never lapse.

WHAT THIS PAGE SAYS, IN PLAIN ENGLISH

Two books, two jobs. Portfolio 100K is the income engine: everything written, ~$106k/mo, upside capped. This book is the upside engine: 11 different names, seven of them free to run, plus a fully written ballast sleeve that pays when the income engine cannot.

The melt-up case: if the income book's names run 30%, its new premium stops (-$6,527/mo after rent), but this book's structures gain $643,766 and its ballast calls keep paying $16,809/mo. Together the two books keep earning and still capture the run.

The selloff case: this book is ballast, not a hedge. It falls too (-$552,280 when the income book loses -$709,009), because its upside names share the income book's beta; the ballast sleeve and GLD soften it. Neither account can be liquidated (ML ≤ cash).

Why the written third sits on the ballast names: it keeps $16,809/mo flowing when the income book melts up (writing a third of every name instead would keep only $1,092/mo, since the upside names' calls go underwater alongside the income book's), and it leaves every upside name free. It also pays $1,612/mo more in normal times than a third of every name would. The trade-off accepted: in a uniform broad rally where the ballast names run too, a third-of-every-name placement would keep $240,649 more at +50%, because LMT, JPM, AMZN and LLY carry the most upside per dollar of max loss.

ASSUMPTIONS AND CAVEATS

Marks: fortress_v28 CSV replay (no live IBKR) of the 2026-09-22 17:53 SGT chain export, i.e. pre-market frozen quotes, Jan-2029 LEAPS. Re-price live in RTH before ticketing.

Betas and best/worst-day means are 180 trading days of Yahoo daily closes against the income book as it stands now (ML-weighted); they are a regime snapshot, not a law. SPCX has only 69 days of history. Scenario moves are linear in beta; real melt-ups and crashes are not.

Income is modelled: v28 DTE-prorated premium x each name's graded keep rate, minus hedge rent. In the melt-up, a written call is treated as paying only if its name moves no more than one cycle (~+8%); a call that goes underwater is counted at zero new premium, not at a roll credit.

Read-only hypothetical; nothing here places orders.