2026-08-21 · midday

The same four rails, unchanged, on a green tape. The runbook swept clear. One new plumbing finding: rails.py has been testing every GOOGL price level against the GOOG quote.

What to do

280P x5 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 · ~$18,381 debit ($36.76 net per spread at midday marks)
Your playbook: “long leg never above $260 (must sit at/above the $230-250 floor zone, never below $230 where protection stops paying before history bottoms)”
Same standing order as prior runs, repriced to today's marks. The 8x Dec-28 300P also trip this rail and are the separate matter. No order is proposed against them.

Still over, unchanged

These fire every run until you act on them. No order attached.
rail now limit
program_cap $175,059 $100,000 cap 1.8x, $75,059 over
gross_leverage 1.746x 1.50 red / 1.30 amber red
net_cash / NLV -0.069 -0.05 amber / -0.10 red amber

rail	now	limit
program_cap	$175,059	$100,000 cap	1.8x, $75,059 over
gross_leverage	1.746x	1.50 red / 1.30 amber	red
net_cash / NLV	-0.069	-0.05 amber / -0.10 red	amber

Why each one reads the way it does

program_cap
Identical to prior runs to the dollar, which is the expected reading. Defined max loss is a function of strikes and entry credits, not current marks, so it moves only when a position actually closes. Nothing closed.
gross_leverage
The IPS row this proxies measures total DELTA notional / NLV; this rail measures GROSS position value. rails.json carries the note '1.57 at signing on gross; true delta leverage ~1.0x'. A red here is not the same statement as the policy row being red.
net_cash / NLV
Cash is -$97,114 on NLV $1,413,139. Cash is unchanged from yesterday to the dollar; the ratio moved only because NLV rose. Signed at -14.8% and already red that day, so this is the improved end of a known accepted state.

Everything else clear

Nearest trigger is tranche 3 confirm at $360, 4% away.

All 14 triggers swept, with today's number

price trigger	now	fires at	distance
GOOGL · tranche 3 confirm	$346.08	≥ $360	4.0% up
GOOGL · tranche 2 weakness	$346.08	≤ $317	8.4% down
GOOGL · thesis break	$346.08	≤ $280	19.1% down
GOOGL · crash / Rule 8	$346.08	≤ $260	24.9% down
GOOGL · Rule 6 close shorts	$346.08	≥ $434	25.4% up
MU · collar roll	$964.36	≥ $1,250	29.6% up

level trigger	now	fires at	distance
Dec-28 30-delta put IV	36.5%	> 50%	+13.5 pt
excess liquidity / NLV	74.6%	< 40%	-34.6 pt
widest long call spread	42.8%	>= 90% of width	+47.2 pt
Dec-28 340P extrinsic	$57.26	< $5	far
NLV drawdown, 1 week	not evaluable	-8% in a week	series began 2026-08-20

calendar	date	away
wash-sale window opens for the 6/05 loss lots	2026-09-17	27d
deploy the $100k reserve at market regardless	2026-10-29	69d
GOOG lots vest long-term, add collar put leg 9	2027-03-13	204d

Spot prices

	last	day
GOOG	$343.11	+1.5%
GOOGL	$346.08	+1.6%
MU	$964.36	-1.0%
NVDA	$215.10	-0.8%
META	$549.85	+0.7%
AMZN	$259.70	-0.2%
RDDT	$152.72	+1.6%
INTC	$90.13	-2.2%
VOO	$704.26	+0.5%
BRK B	$496.67	-0.0%
EQT	$53.69	-0.4%
TSLA	$365.70	+6.0%
DIS	$107.54	+0.2%

The check itself

never worked rails.py tests every GOOGL price level against the GOOG quote.
rails.json lists the campaign underlyings as ["GOOG", "GOOGL"], and rails.py takes the first one that has a quote, then breaks. So thesis_break, t2_weakness, t3_confirmation, washout_buy, no_chase, rule6_close_shorts and rule8_crash_powder are all measured against GOOG. Every one of those levels is written as a GOOGL level: the runbook header says 'Reference price: GOOGL $347' and Part 2 says 'GOOGL closes >= $434'. Today GOOG is $343.11 and GOOGL is $346.08, a 0.86% gap, so the rail reads tranche 3 as 4.9% away when against GOOGL it is 4.0% away. No rail has flipped on this yet. The report page has been showing the GOOGL number against these levels the whole time, so the page and the script have quietly disagreed. This run does not edit rails.json.
will happen again get_price_snapshot returns an empty prior_close for almost every ticker.
12 of 13 tickers came back with 'prior-close': {}. Only EQT populated it. Re-requested with the 'change' field and derived prior close as last minus change, then validated the derivation against EQT, the one ticker that returned both: 53.69 minus -0.20 equals 53.89, matching its reported prior close exactly. Costs one extra round of connector calls every run.
never worked Runbook Part 3 'NLV falls 8%+ in one week' is still not evaluable.
Yesterday's run identified this and started logging NLV into rails-log.jsonl. The series now holds two points, 2026-08-20 and today. The row becomes testable on 2026-08-27.

Rails that could not be evaluated

stressed_excess_liquidity - needs IBKR Risk Navigator (-30% price, +50% IV) with the TWS gateway up. Never yet measured.
top2_name_delta_ratio - needs per-leg option deltas, which summary.json does not carry. ~0.90 at signing and already red.
short_dec28_340p_extrinsic - rails.json still classes this unmeasured pending a live Dec-2028 chain pull, but the position mark answers it: $57.26 of extrinsic against the <$5 watch.
GPU spot pull and the forecast pass are open-run only and were not run at midday.

Where the numbers came from

IBKR connector: get_account_positions, get_account_summary, 16 get_price_snapshot calls plus the Dec-28 340P IV
scripts/investing/rails.py against investing/rails.json
investing/googl-mu-runbook.md Parts 2-4