2026-09-08 · midday

The options book reopened, and that is the whole point of this run. This morning's page carried a caveat on every option-derived number: the check fired nearly three hours before the open, so it priced live stock against Friday's frozen option marks and warned you to read the mixed rows with care. That caveat is now settled with measured numbers instead of a guess. On live Dec-2028 marks the long-leg repair costs $18,793.92, against the $18,769.60 the last six runs carried on frozen marks - a difference of $24.32, or 0.13 percent. The three call-spread width readings each moved about half a point. So the warning was right to make and small in size, and the standing order does not change. The implied-vol reading also got a second witness: option-midpoint-iv came back valid for the first time in seven runs at 35.38 percent, against the 35.68 percent this week has leaned on from a single field. Same three structural rails, every runbook trigger clear, nearest level GOOG tranche 2 at $317, 5.4 percent below. INTC is up 9.47 percent and has no written rule, so it gets named and nothing else.

What to do

280P x5 and 300P x8 sit above the $260 long-leg ceiling.
BUY TO CLOSE 5 GOOGL Dec15'28 360 PUT / SELL TO CLOSE 5 GOOGL Dec15'28 280 PUT
one spread, limit, day · ~$18,793.92 debit ($37.5878 net per spread on live marks)
Your playbook: “long leg never above $260 (must sit at or above the $230-250 floor zone, never below $230)”
Seventh read of the same standing order, and the first one priced on a live chain since Friday. The six runs before this quoted $18,769.60 off frozen marks; the live number is $18,793.92, $24.32 more, so nothing about the order or its size changes. Only the 5x 280P leg is reachable by this order; the 8x 300P leg has no written repair. Reprice before sending - the chain is moving now.

Still over, unchanged

These fire every run until you act on them. No order attached.

rail	now	limit
long_leg_band	280P x5, 300P x8	$230-$260 band	13 contracts outside
net_cash_ratio	-0.065	amber -0.05, red -0.10	amber
gross_leverage	1.751x	amber 1.30x, red 1.50x	red
top2_name_delta_ratio	0.66x	amber 0.70x, red 0.85x	clear
total_delta_ratio	0.94x	amber 1.30x, red 1.50x	clear

Why each one reads the way it does

long_leg_band
The rail the action item above addresses. No leg has been traded. The marks behind it are live for the first time since Friday.
net_cash_ratio
Cash -$93,230.12 against NLV $1,426,566.44. The cash figure has not changed by a cent since Friday; the ratio moves only with NLV.
gross_leverage
Gross $2,497,745.83 over NLV $1,426,566.44. Carried caveat: this divides GROSS position value by NLV against thresholds the IPS wrote for delta notional, and the delta rows in this same table read 0.66x and 0.94x, so the red reading overstates true leverage.

delta by name

Delta notional per name as of 2026-09-08: stock at delta one, each option leg at the Black-Scholes delta implied by its own mark. Top two: GOOG, MU.

name	delta notional	of NLV
GOOG	$803,790	56.3%	1543 sh, 16 legs
MU	$140,689	9.9%	525 sh, 2 legs
META	$88,872	6.2%	144 sh, 0 legs
VOO	$86,725	6.1%	123 sh, 0 legs
RDDT	$79,433	5.6%	513 sh, 0 legs
NVDA	$50,513	3.5%	285 sh, 2 legs
INTC	$31,461	2.2%	300 sh, 0 legs
AMZN	$25,714	1.8%	100 sh, 0 legs
BRK B	$20,176	1.4%	40 sh, 0 legs
EQT	$11,056	0.8%	200 sh, 0 legs
TSLA	$369	0.0%	1 sh, 0 legs
DIS	$105	0.0%	1 sh, 0 legs

Everything else clear

Nearest trigger is tranche 2 weakness (nearest written level, and what rails.py reads) at $317, 5% away.
No position moved more than 5% today. Biggest was INTC +9.47%.
INTC +9.5%, now $104.87. No written rule for it.
Named, not acted on. INTC is on the no-written-rule list, so this check proposes nothing for it. It was +3.47% when the run fired pre-market and is +9.47% now.

All 26 triggers swept, with today's number

price trigger	now	fires at	distance
GOOG · tranche 2 weakness (nearest written level, and what rails.py reads)	$335.21	≤ $317	5.4% down
GOOGL · tranche 3 confirmation	$338.50	≥ $360	6.3% up
GOOG · tranche 3 confirmation	$335.21	≥ $360	7.4% up
GOOG · thesis break, close regardless of price	$335.21	≤ $280	16.5% down
GOOG · washout, stop selling spreads and buy stock	$335.21	≤ $260	22.4% down
GOOG · crash powder, deploy the $100k reserve	$335.21	≤ $260	22.4% down
MU · collar roll-up	$1,019.73	≥ $1,250	22.6% up
GOOGL · close every short put	$338.50	≥ $434	28.2% up
GOOGL · long spreads near max, apply the buy-it-fresh test	$338.50	≥ $500	47.7% up
GOOG · shares capped by the 600Cs, roll the collar up	$335.21	≥ $600	79.0% up

level trigger	now	fires at	distance
Dec-28 30-delta put IV	35.68%	> 43%	+7.3 pt
GOOGL IV, crash trigger	35.68%	> 50%	+14.3 pt
11x Dec-28 300/400 call spread, mark vs width	41.53%	>= 90%	+48.5 pt
8x Sep-27 300/440 call spread, mark vs width	38.06%	>= 90%	+51.9 pt
10x Dec-28 350/500 call spread, mark vs width	28.83%	>= 90%	+61.2 pt
GOOGL Dec-28 340P extrinsic	$56.08	< $5	$51.08
Excess liquidity / NLV	74.39%	< 40%	+34.4 pt
NLV change on the week	+3.16%	-8% in one week	+11.2 pt
Defined max loss vs program cap	$175,059	$200,000	$24,941 headroom
Biggest single-name move today	INTC +9.47%	>= 5%	fired, no written rule

calendar	date	away
Wash-sale window opens for the 6/05 GOOG loss lots	2026-09-17	9d
Micron fiscal Q4 print (not a runbook trigger row)	2026-09-30	22d
Alphabet Q3 print · scheduled thesis adjudicator	2026-10-28	50d
Reserve deploys at market if no crash by this date	2026-10-29	51d
Earliest date the GPU 90-day tripwire test can evaluate at all	2026-11-16	69d
First GOOG lot vest, collar put leg 9	2027-03-13	186d

Spot prices

	last	day
GOOG	$335.21	-0.0%
GOOGL	$338.50	+0.0%
MU	$1,019.73	+0.3%
AMZN	$257.14	-0.5%
BRK B	$504.40	-0.3%
DIS	$104.59	-0.7%
EQT	$55.28	+0.2%
INTC	$104.87	+9.5%
META	$617.17	+0.1%
NVDA	$225.98	-1.9%
RDDT	$154.84	+0.2%
TSLA	$369.17	+4.3%
VOO	$705.08	-0.4%

The check itself

found this run the first positions pull of this run carried GOOG at Friday's close while the live quote was 335.18.
At 12:39 New York the positions endpoint returned GOOG at 335.30999755 with daily_pnl of -$0.0038 on 1,543 shares. That is Friday's official close to eight decimal places, on the single largest equity line in the book, at a moment when the snapshot endpoint had GOOG trading at 335.18 with a last-trade timestamp seconds old. Every other equity row had moved. A re-pull two minutes later returned 335.1549988 with daily_pnl -$239.17, matching the snapshot's own plprice field exactly, so the row self-corrected and the book of record for this run is the second pull. Size it honestly: the stale row overstated that line by about $200, which is 0.014 percent of NLV and moves no rail. The finding is not the $200, it is that the check has no consistency test between the positions feed and the quote feed, and a run that pulled once would have shipped the stale figure without noticing. Not changed - adding that test is an edit to the check.
resolved the shut-book option-midpoint-iv artifact is gone.
For six consecutive runs the GOOGL Dec-28 340P returned option-midpoint-iv with annualIv -15.8745 and isValid false, a shut-book artifact, and every one of those runs fell back to the implied_vol field's 0.3567735. Today the contract reads top_status DELAYED rather than FROZEN_DELAYED and option-midpoint-iv is valid at 0.3538424. The two independent fields now agree to 0.30 vol points. That matters because the IV rails - the 43 percent tranche-2 alternate trigger and the 50 percent Part 3 crash trigger - have been running on a single unwitnessed field all week, and they now have a second source that agrees.
resolved the mixed-marks caveat this morning's page carried is now measured rather than assumed.
The open run flagged that every option-derived row mixed live pre-market stock against Friday's frozen option marks, and said so without being able to size it. Now it can be sized. The long-leg repair debit, which had printed identically to four decimals for six runs, moves from $18,769.60 to $18,793.92 - $24.32, or 0.13 percent. The 11x Dec-28 300/400 width reading moves from 42.05 to 41.53 percent, the 8x Sep-27 300/440 from 38.51 to 38.06, the 10x Dec-28 350/500 from 29.29 to 28.83, each about half a point. The 340P extrinsic reads $56.08 against the $54.73 the morning quoted, and that one moved mostly because the spot leg had updated under a frozen mark rather than the other way round. So the caveat was correct to state and worth about a tenth of a percent on the number Gabriel would actually trade.
will happen again the cron has no timezone anchor and no trading-day guard.
Unchanged from this morning, and this run is the other half of the evidence. The slot is written '35 6,12 * * *' with no timezone, so it fires at machine local time on a Central-clock machine. The open slot, defined as five minutes after the US open, fired 2h53m before it. This midday slot, defined as one hour before the close, fired at 12:41 New York - roughly three hours into the session and about three hours early. On this run the drift is benign, because three hours into a session is a perfectly good time to read a live book. It is still the same missing anchor. Not changed: the schedule is Gabriel's to edit.
will happen again deploy.sh still needs an arm64 node pinned ahead of it on PATH.
Unchanged. /usr/local/bin/node is a Mach-O x86_64 binary sitting ahead of /opt/homebrew/bin on PATH while /opt/homebrew/bin/node is arm64, and the npx workerd cache wrangler pulls is darwin-arm64 only, so the default PATH fails the upload. Worked around per-run by pinning PATH=/opt/homebrew/bin ahead of the deploy call. The permanent fix belongs in the shell profile, not in this check.
will happen again markets.dottl.xyz has no Access gate.
Re-verified previously and unchanged: the full book, positions, strikes and account totals are readable by anyone with the URL. Left as-is pending Gabriel, because putting a gate on it is his call and not this check's.

Rails that could not be evaluated

prior_close was served on only 5 of 13 equities again. The other 8 were derived as last minus the broker's own change field, and all 13 cross-check to the cent against the prior closes the 06:37 run derived independently from get_price_history. No price was estimated, but the field is unreliable and the check leans on a derivation for most names.
The GPU tripwire is defined at a 20 percent fall on an instrument that moves 20 percent or more on roughly half of all days. Even after 2026-11-16 it will be a coin flip rather than a signal. It does not close by waiting.
The GPU tripwire cannot evaluate at all until 2026-11-16 - the 90-day lookback has no row to compare against, the daily series having started 2026-08-18.
The SemiAnalysis composite is five months stale at 2026-04 $2.82 and the runbook's tripwire prose is defined on it, while later months are paywalled. The anchor and the instrument are both unavailable.
2026-08-29 is still a hole in gpu-spot-daily.csv.
stressed_excess_liquidity - needs IBKR Risk Navigator (-30% price, +50% IV) with the TWS gateway up. Never yet measured, and the IPS amber's 7-day stress re-run has never run.
short_dec28_340p_extrinsic is still listed unmeasured in rails.json because it needs a live Dec-2028 chain pull. The runbook's own version of the row was swept by hand and is clear at $56.08 against the $5 trigger, and unlike the last six runs that reading is on a live mark.
gross_leverage divides GROSS position value by NLV against thresholds the IPS wrote for delta notional, so the red reading overstates true leverage. Unchanged and unresolved.
rails.py reads GOOGL levels off the GOOG quote, which is why the tranche-2 and tranche-3 distances differ between the two tickers in this report.
rails.py keys option-leg matching on contract_description with no contract_id fallback.
The 11x Dec-28 300/400 line is 11 long 300C against 1 short 400C, not eleven matched spreads.
The MU collar roll cannot be executed as the runbook writes it: Part 2 says 'BUY back 1400C' but there is no 1400C - the collar is 5x Dec-28 1740C short against 5x 900P long. Only the trigger row and IPS Rule 7 line 242 are stale; the book-summary line and the tax documents are already correct. First reported 08-18, and the trigger is 22.6 percent away.
The runbook's own 2026-09-04 note already says the MU roll-up as written is not premium-neutral: selling the 900P and buying the 1150P costs about $79k net on the 5-lot, while the roll down-and-out is a $41k credit. Re-price on the day.
The 8x GOOGL 300P long leg sits outside the $230-260 band and has no written repair order. Only the 5x 280P leg is reachable by the standing action item.

Where the numbers came from

Market OPEN and roughly three hours into the session. This is the first run since Friday 2026-09-04's close with a live options chain, and that is what makes it worth reading rather than a repeat of the morning.
Run label decided from the rails log rather than the clock: an entry already carried today's date reading markets-open, so this is the midday run. Section 4b (GPU) and section 5 (forecasts) were skipped accordingly - both ran at 06:37.
get_account_positions, get_account_summary and 14 get_price_snapshot calls all succeeded. Positions and summary were pulled twice; see plumbing for why, and the second pull at 12:41 New York is the book of record.
All 13 equities returned top_status REALTIME with last-trade stamps 1788885575-1788885737 (12:41:15-12:42:17 New York).
The GOOGL Dec-28 340P returned top_status DELAYED rather than FROZEN_DELAYED, and its option-midpoint-iv came back valid for the first time in seven runs. implied_vol's 0.35684919817367244 was used for continuity with the prior runs; option-midpoint-iv reads 0.3538424 and the two agree to 0.30 vol points, so the choice changes no verdict.
Positions diffed against the open run's saved copy: 32 rows, same 32 contract_ids, zero quantity changes, all 32 marks moved. The 20 option rows moving is the news - they had been frozen at Friday's close for six consecutive runs.
prior_close was served on 5 of 13 (GOOG, BRK B, EQT, RDDT, VOO); the other 8 were derived as last minus the broker's change field. All 13 agree to the cent with the prior closes the open run derived independently from get_price_history ONE_DAY outside_rth=false on 2026-09-04. No price estimated.
contract_description present on all 32 rows, so defined max loss prints the complete $175,059 and uncapped_shorts was measured rather than skipped: 10 GOOG short against 10 long, 30 GOOGL short against 38 long, zero uncapped.
Every Part 2, 3 and 4 runbook trigger row was walked by hand against this run's numbers, including the three call-spread width percentages, the 340P extrinsic and the weekly NLV test. All clear. The weekly NLV test used the 2026-09-01 open reading of $1,382,809.14 from the rails log.
The delta block is rails.py's output copied verbatim, as the prompt requires.
No GPU pull and no forecast pass: both belong to the open run and both ran this morning.