Live journal

**Day summary.** It was a quiet day with one trade. Equity is about $103.5k and cash about $25.8k, so roughly 75% is invested. That is above the 40-70% band the mandate asks for. The diversifiers led: AMZN +3.29% (closed $262.43, high $262.80, a new high for its September range) and MSFT +2.38%. The semis drifted lower: AMD -2.03%, TSM -1.03%, MU -0.66%, NVDA -0.52%. The only fill was the tsla-swing trim. It sold 15.60 TSLA shares for $6,000 at $384.53, after the stock opened at $382.38 and reached $388.56. TSLA closed at $382.70 (+2.05%). Nothing is waiting for approval except the new proposal below.

**Per-strategy check.**
- **tsla-swing:** The trim fired in a way that contradicts this strategy's own stated plan. The pre-trade analysis recommended skipping it for three reasons: price was only $0.45 above the trigger, the position (8.25%) was already below its 10% target, and Q3 earnings are on 10/21. The trim went through anyway. It sold about 5.8% of equity and left about 2.4% (6.50 shares averaging $361.93). The strategy text promises a 6% minimum after trims, but the rule uses 4%, and one $6,000 sale is bigger than the gap between them. The rule re-armed at 15:08. At today's weight it can't fire again, since the stake is below the 4% minimum. Rebuying needs the stock to fall about 5% to under $362. The take-profit (sell all above $410) is about 7% away and the stop (below $335) about 12% away. I proposed a fix (#146).
- **amzn-swing:** Weight is about 11.5% after today's rally. The trim needs weight above 15%, so it stays inactive, which is how the 10/6 change was meant to work. The sell-all take-profit above $268 is only about 2.1% away. If it fires it closes a position bought at an average of $245.90, which would be roughly a +9% win. That is acceptable for a swing.
- **msft-diversifier-swing:** Up 2.38% and still working as the low-correlation leg against the semis. Weight is below the 8% minimum, so the trim stays inactive.
- **aapl-swing:** Down 1.11%. The position (48 shares averaging $333.31) is inside the rule range. The stop below $310 and the trim above $348 are both far off.
- **amd-swing:** Down 2.03%, about 2.6% weight. The 10/6 trim already overshot. The dip-buys at <$600 and <$560 are within reach on a moderate pullback.
- **mu-oversold-swing, tsm-swing-contest, nvda-momentum-swing:** Small moves on the day. Every rule is well away from price. Positions are near cost (MU) or in profit (TSM, NVDA).
- **googl-swing, meta-swing:** Small moves and nothing near a trigger.

**Lessons.**
1. The trim rule overshot for the fourth time in a week (MSFT 10/5, AMD and AMZN 10/6, TSLA 10/9). The cause is that the dollar amount of one trim is bigger than the gap between the current weight and the minimum weight. I saved this to memory as a lesson. The fix: set the minimum to at least target minus one sale, and only trim once weight is above the minimum plus one sale.
2. Because these strategies run in confirm mode, I can't tell from the log why the trim executed despite the skip advice. Either it was approved by hand or the advice wasn't shown at approval time. Approvals should be checked against the attached analysis.

**Proposals.**
- **#146 tsla-swing v2:** Trim only when weight is above 13% and price is above $390, and sell $4,000 each time. That keeps the position around 9% after a trim and needs a real break above the $386-389 highs. Entries, stop, take-profit and the conditions that would end the thesis are unchanged. I added no new buy ahead of the 10/21 earnings report.
- **Not proposed:** fixes for the other three strategies that overshot. AMZN and MSFT are now protected by their minimum-weight conditions. AMD's damage is done, and its dip-buys rebuild the position if the price comes back.

**Day summary (2026-10-08, paper):** No rules fired and nothing is pending. Chip stocks fell together. News reports blamed higher bond yields and oil prices, plus Samsung's Q3 operating profit, which jumped but still came in under very high expectations. MU fell -4.79% to close at $1,035.84, AMD -3.90% to $620.68, TSM -3.00% and NVDA -2.94%. AMZN fell -2.25% and MSFT -1.35%. AAPL rose +1.11%, and META, TSLA and GOOGL were roughly flat. Equity is about $103.2k (roughly $103,171–103,176) and cash is $19.8k, so about 81% of equity is invested. That is above the 40–70% band, but this paper phase aims for high activity. Every position still shows an open gain. Caveat: the position values the portfolio tool reported look the same as the session-start snapshot. They may not include today's close yet, so treat the equity figure as approximate.

**Per-strategy check:**
- **mu-oversold-swing:** Today wiped out most of 10/7's rebound. The 10.09 shares bought at $1,032.67 average are now about breakeven, at ~10% weight against a 10% target. No rule is close to firing:
  - dip-buy-1 is blocked by its weight cap.
  - dip-buy-2 (<$970) is 6.4% below.
  - The stop (<$900) is 13% below.
  - The trim (>$1,100) is 6.2% above.
  
  The $1,010–1,035 area has now held as a low in three dips over eight sessions. The thesis still holds.
- **amd-swing:** Holds 4.18 shares, ~2.5% weight against a 12% target. dip-buy-1 (<$600) is now only 3.3% below, so it is the nearest buy in the book. If it fires, the $8k tranche would take the position to about 10%. That's acceptable.
- **tsm-swing-contest / nvda-momentum-swing:** They fell with the sector, but no rule was reached. NVDA is still above its dip-buy levels, and TSM's dip-buy is still below the price.
- **aapl / amzn / googl / meta / msft / tsla swings:** These moved within their normal daily range and triggered nothing. AAPL rising on a day chips sold off is consistent with its role as a diversifier.

**Lessons:** When the whole sector drops 3–5% in one day on macro news, the buy ladders should stay quiet, and they did. Losses from the chip names were mostly offset by the non-chip names. One down day is not a reason to move the buy levels closer to the price. I saved this lesson to memory and added a MU note on today's support level, replacing an outdated entry.

**Proposals:** None. Nothing measurably diverged from the strategies' stated assumptions.

**Day summary.** The account had one trade today. MU opened with a gap down to $1,017 (low $1,011), then reversed and closed at $1,088 (+4.06%) on above-average volume (29.9M). Its first buy-on-dip rule (price under $1,030) triggered near $1,015. Because the strategy needs my approval before trading, the order only filled at 14:01 ET, at $1,034.43. That is above the rule's own price, but the stock still closed 5.2% above the fill. TSM fell 2.10% to $472.15 on light volume (6.4M) after its 10/5 record. META fell 2.38% to $721.31 and closed near the day's low. AMZN (+1.42%), AAPL (+0.91%) and GOOGL (+0.81%) rose, so the names outside semiconductors offset some of the semiconductor weakness. About 81% of equity is invested (cash is $19.8k of $104.4k), still above the 40–70% target band. No orders are waiting for approval.

**Per-strategy check.**
- **mu-oversold-swing:** The fill matched the design: buy into a real intraday flush, as on 10/1. MU is now 10.09 shares at an average cost of $1,032.67. That is about 10.5% of the account against a 10% target. The first buy-on-dip rule no longer applies because MU is above its 6% weight limit. The partial-sell rule above $1,100 is 1.1% away and is the most likely next trade. The stop at $900 is sound.
- **tsm-swing-contest:** This looks like a normal pullback after the record, on the lightest volume in two weeks. The position is about 8%, still up roughly 6% against cost. No rule is close to triggering.
- **meta-swing:** META is about 7.7% of the account and $1.31 above the $720 buy level. That buy rule can't fire anyway, because it requires META to be under 5% of the account. The stop at $615 is far away. Nothing to act on.
- **amzn-swing:** The partial-sell rule reset this morning. It will not sell because AMZN is about 11% of the account, under its 15% sell threshold, which is how the 10/6 change intended it to work.
- **aapl-swing, googl-swing, msft-diversifier-swing, amd-swing, nvda-momentum-swing, tsla-swing:** All moved less than 1%. None of their rules were close to triggering. The known AMD and NVDA problem remains: their buy levels sit far below the market, so those rules can't trigger (already recorded in memory).

**Lessons.** Approval delay on high-volatility names consistently puts fills on the wrong side of the trigger: the 9/21 MU take-profit filled at $1,038 against a rule of above $1,050, and today's buy filled at $1,034 against a rule of below $1,030. I saved this as a lesson: treat triggers as zones of about ±1% (about a third of a typical day's range), and don't count fills inside that zone as rule errors. I updated the MU memory with today's flush-and-reversal fill and the new position size.

**Proposals.** None. Every strategy behaved as written today. Investment above the target band is a portfolio-level issue that is already being addressed through sell rules, not a flaw in any one strategy.

**Day summary:** It was a quiet, mildly risk-on day. Account equity is about $104.0k and cash is about $27.8k, so roughly 73% is invested, slightly above the 40-70% band. Two trims filled, both well placed:
- **AMD:** sold 9.18 shares @ $653.39 against a record close of $649.42 (+2.80%, day high $658.52).
- **AMZN:** sold 19.54 shares @ $255.92 against a close of $256.29 (+1.95%), at the top of its $244-260 range since September.

MU (-1.73%) and TSM (-0.73%) were the weakest names, and no stop came close. The pending queue is empty.

**Per-strategy check:**
- **amzn-swing:** The first trim was correct: weight went from about 16% to about 11.2%. The rule then re-armed and fired again at 18:08 and 19:38 ET with weight already below the 15% target. Its sell condition only checks that weight is above 10%, and each $5,000 sale removes about 5 points. So a second sale would have taken the position to about 6%, below the 10% minimum it was supposed to protect. The sentinel declined the 19:38 fire on its own judgment, not because a rule blocked it. This is measurably wrong, so I proposed #141.
- **amd-swing:** The trim worked as designed and booked about +9% on those shares. With a 5% floor and a $6,000 sale, weight went from about 8.4% to about 2.6% in one fire, against a 12% target. The trim can't fire again now. The buy levels sit 8-14% below the price, which is deliberate while AMD is stretched after a big run. I'm accepting the small position for now; no revision.
- **aapl / googl / msft / tsm / nvda / mu / meta / tsla:** No rules fired and all moved less than 2%. All are holding inside their stated ranges. NVDA's full take-profit (>248) and TSM's (>510) remain the closest triggers to watch.

**Lessons:** I saved a new lesson ("trim-arithmetic"). A trim rule that re-arms can push weight down to its trigger level minus one sale. So the trigger should sit at or above the target weight, or at least the minimum weight plus one sale, or the "floor" doesn't protect anything. AMD and AMZN both showed this today. I also updated the AMD memory note.

**Proposals:** #141 amzn-swing v3 raises the trim trigger from weight above 10% to above 15% (the target). One $5,000 sale then leaves at least about 10%. Price levels, buy rules, stop and take-profit are unchanged.

**Day summary (2026-10-05, paper):** It was a strong day for AI and mega-cap stocks. TSM closed at a record $485.91 (+2.78%), NVDA at a record $238.90 (+2.12%), TSLA rose 2.20%, META 1.90% and MSFT 1.48%. MU slipped 1.02%, and AAPL, AMD and AMZN ended about flat. Two trades went through, both at 13:40 ET:
- **MSFT:** a trim sold 11.33 shares at $529.48, 0.8% above the $525.18 close.
- **TSM:** a trim sold 8.34 shares at $479.67, about $6 below the $485.91 close.

Both rules re-armed after their 90-minute wait. Neither fired again, because each position had dropped below the weight limit that lets the rule sell. Cash rose from about $6.8k to $16.8k, so the share of the account invested fell from about 93% to about 84%. That is still above the 40–70% target band, but moving the right way. Nothing is waiting for approval.

**Per-strategy check:**
- **tsm-swing-contest (v12):** The revision made on 10/2 (sell $4k once the stock is over 11% of the account; full-exit level raised to above $510) worked on its first test. It took a partial profit at a record high and kept most of the position. TSM is now about 8.3% of the account (17.74 shares at an average cost of $435.64, up about 11.5%). The full-exit level is about 5% above the price, the first buy level (<$447) about 8% below, and the stop (<$418) is fine. Selling near the open gave up about $6 a share as the stock kept climbing, which is the accepted cost of selling in steps.
- **msft-diversifier-swing (v4):** This was the first move through $525 since the 9/25 breakout, and the sale price was good. MSFT is now about 6.9% of the account, up about 5%, which is below the 8% level that allows another trim. The full-exit level (>$545) is 3.8% above the price, the buy levels (<$508 and <$495) are below it, and the stop at <$480 is fine. No change needed.
- **nvda-momentum-swing (v11):** NVDA has broken above the $205–234 range the strategy is built around. The position is small (about 2.5% against a 15% target), and the buy levels (<$222 and <$213) are 7–11% below the price, so they are unlikely to fill. Normally that would call for a fix. I left it alone because the account is already above its target invested range. One thing to watch: the full-exit rule at >$248 is only 3.8% above the price and would sell the whole position.
- **tsla-swing:** Closed at $378.73, about 1.7% under the trim level (>$385). Rules are consistent with the thesis.
- **meta-swing:** Closed at $741.90, between the buy level (<$720) and the trim level (>$800). Nothing to do.
- **aapl-swing, amzn-swing, googl-swing, amd-swing, mu-oversold-swing:** Quiet day for all five, no rules near firing and no changes to their theses. AMD and MU remain the volatile, short-term-trade names.

**Lessons:**
- Trims that only fire while a position is above a set share of the account are working as a funding source. Each name sells once on a strong day, the core position stays, and cash builds for the buy ladders with no manual step. If that share limit sits just under the current weight, one sale per day is the normal result, not a bug.
- Buy-on-pullback ladders get stranded when a stock breaks out (NVDA again). That is acceptable while the account is over-invested, but the full-exit levels need watching so they don't quietly sell a whole small position.

**Proposals:** None. No strategy's behaviour clearly diverged from its stated assumptions. The NVDA issues are on the watch list, to be re-anchored if the price gets near $245.

**Day summary**
A broad, strong risk-on session: nine of ten positions closed green. Equity $102,404 (+$10 vs session start on a mark-to-market basis), cash down to $6,844 — roughly 93% deployed, well above the phase-3 40-90% band. TSM led with +2.99% to a record close of $472.92 (open $465.64, high $474.79) on unremarkable volume (8.9M, inside the 7.7-11.3M two-week range), TSLA +4.65% on a Q3 delivery beat (486,532 delivered vs ~462k consensus; earnings 10/21), AMD +2.95% to $633.91, NVDA +1.34% to a new high close of $233.95 on heavy 129.7M volume. The only loser was MU, -2.05%: it opened $1,107.45, printed $1,107.86, then faded to $1,074.89 — and that fade is exactly where the day's single trade happened. One fill: mu-oversold-swing/trim-strength (weight>0.04 and price>1100 → sell $6,000) executed at 13:37 ET, 5.4329 sh @ $1,104.38, then re-armed at 15:07 ET. That fill was 2.7% above the close and within 0.3% of the day's high — the best execution of any trim this book has produced. Pending queue was empty at the open and nothing else triggered.

**Per-strategy check**
- mu-oversold-swing (v9): Working as designed and the day's clear success. The $1,100 trim threshold was set on 9/26 "above today's high" and MU reached it twice since; today it sold into the spike at $1,104.38 rather than chasing. Remaining 2.358 sh @ $1,026.89 (2.5% weight, +$112). Dip-buy-1 (<1030) is ~4.2% below market and genuinely reachable given this name's ~$40 ATR. No change warranted.
- tsm-swing-contest (v11): Measurably off — revision proposed. At $472.92 the >475 full take-profit is 0.44% away and would liquidate the book's largest winner (+$954, +7.5%) and its designated patient anchor on a further 0.4% move. Meanwhile v11's 15% trim floor blocks any partial trim at 12.0% weight, so at a record high the strategy's only live exit is the full one. Entries (<447/<435) and the <418 stop remain sound.
- nvda-momentum-swing (v11): Thesis of a $205-234 oscillation is now being tested on the upside — close $233.95 is at the top of that range, and trim-strength (>232, floor 8%) did NOT fire because weight is only 2.5% after earlier trims. The rule is structurally dead at current sizing, not broken. Option exits (>227) remain triggered. Watch, do not re-anchor on one day.
- tsla-swing (v1): Thesis explicitly named delivery guidance as the invalidation test; today's print beat and the stock added +4.65%. Position 22.1 sh @ $361.93 (+$182, 8.0%). dip-buy-1 (<362) is now below market and dormant; trim at >385 is the next live level. Consistent.
- amd-swing (v9): Deliberately set entry ceilings below market because AMD was extended; at $633.91 dip-buy-1 (<600) is 5.3% below and trim (>650) 2.5% above. Position +$462. The pullback-only design is being respected rather than deadlocked — in range on both sides, which is the fourth-time-lucky version of this name's old problem.
- amzn-swing (v2), googl-swing (v2), aapl-swing (v2): All post-sizing-revision and all quiet, which is the intended behavior. AMZN 16.0% weight (above the 15% target, both dip-buys correctly blocked), GOOGL 8.0% vs 15% target with dip-buy-1 (<338) ~2% below market, AAPL 15.6% weight at essentially flat cost. No divergence.
- meta-swing (v1): +0.30%, position +$91 at 7.9% weight — at target. Entry levels (<720/<670) sit below market as designed; trim at >800 far off. Nothing to do.
- msft-diversifier-swing (v4): +0.92%, +$443, weight 12.6% vs 15% target. dip-buy-1 (<508) and the entry-call (<505) are both close to market and still live. Working.
- crm-fade-swing: archived, no rules, no action.

**Lessons**
Two durable observations. First, the MU trim is the cleanest evidence yet that a trim threshold set deliberately *above* the prevailing high — rather than at a round number near market — gets filled into spikes instead of into fades: $1,104.38 against a $1,074.89 close. Second, and the real finding of this pass: a full-position take-profit placed only ~4% above market decays into a disguised stop within a single strong session. TSM's >475 was written five sessions ago at $456 and is now 0.44% away; the same geometry, applied to a 15%-target anchor, would undo weeks of patient accumulation because price went up 3% on average volume. Full exits belong at blow-off distance; staged trims belong near market. A third, softer point: at 93% deployed with ten ladders armed, trims are no longer just profit-taking — they are the book's only funding mechanism, so a blocked trim rule now has a second-order cost across every other strategy.

**Proposals**
One revision queued: #135, tsm-swing-contest v12. It moves the full take-profit from >475 to >510 (~7.8% above market, beyond any measured move from the $424-462 base) so it is a genuine blow-off backstop again, and lowers the trim floor from 15% to 11% with the trim trigger at >468 and a $4,000 tranche (max 2 fires/day, 90m cooldown) so strength can be harvested in ~4pp steps from today's 12.0% weight instead of jumping straight to a full liquidation. Entries, the <418 stop, and all other price anchors are untouched. Awaiting your approval on the Pending page — I cannot apply it myself. No other strategy showed a measurable divergence from its stated assumptions today, so no other proposals.

Day summary
Two rules fired, both as designed, and both bought real weakness rather than chasing. mu-oversold-swing/dip-buy-1 filled 7.791 MU shares @ $1,026.89 ($8,000) at 14:22 ET — MU had opened $1,054.34 and flushed to $1,022.90 after FQ4 FY2026 (record revenue and cash flow, beat on the print, but heavier capex guidance hit the stock after hours on 9/30), then reversed to close $1,097.39 (+3.03%) on 44.5M shares, the heaviest volume in weeks. The fill landed 0.4% off the session low and was +6.9% (+$522) by the close. googl-swing/dip-buy-1 filled 23.747 GOOGL shares @ $336.88 ($8,000) at 16:52 ET as GOOGL sold off from a $350.79 open to a $335.51 low, closing $338.24 (-1.70%) — a press report of internal skepticism about the next Gemini model erased an intraday gain. That fill was also 0.4% above the low and 0.4% below the close. Elsewhere the book was quiet: AMD +0.65%, TSM +0.66%, NVDA +1.09%, MSFT -0.02%, META +0.10%, TSLA -0.20%, AMZN -0.37%, AAPL -0.81%. Equity $100,867 with $844 cash — the book is now ~99% deployed, which is above the mandate's 40-70% band and is the single most important fact of the day. Pending queue was empty at session start; my revision is now the only item.

Per-strategy check
- mu-oversold-swing (v9): working exactly as written. dip-buy-1's 0.06 cap sits below the ~8% weight one $8,000 fire creates, so the rule self-blocks after one fire — correct geometry. It re-armed at 16:22 ET but price was already back above $1,030, so no second fire. trim-strength (>1100, floor 4%) missed by $1.10 (high $1,098.90) and is the book's next live rule. Stop <900 is ~18% below, wide but intentional for a ~3.7% ATR name.
- googl-swing (v1): thesis validated on the first fill, geometry not. dip-buy-1 capped at 0.10 while an $8,000 tranche is ~8% of equity, so the re-armed rule could fire again and carry weight to ~16%, past the 15% target; dip-buy-2's 0.18 cap plus another tranche could then breach max_weight 0.20. With $844 cash any further $8,000 fire would be margin-funded. Revision proposed.
- aapl-swing (v2) / amzn-swing (v2): already carry the $5,000-tranche fix. AAPL -0.81% to ~$330, dip-buy-1 (<335, cap 0.09) is blocked by weight (~15.7%), not price — correct. AMZN -0.37%, weight ~16%, also weight-blocked. No action.
- amd-swing (v9): price near $616, dip-buy-1 (<600) and trim (>650) both out of reach; deliberately so while AMD is extended. No action.
- meta-swing / tsla-swing (v1 each): positions ~8% each, dip-buy-1 caps 0.05 and 0.06 respectively — those caps sit below current weight, so entries are correctly blocked. Both still carry $8,000 tranches, but because the caps bind before the tranche fires, they do not have GOOGL's defect. Left alone.
- nvda-momentum-swing (v11): NVDA +1.09%; dip-buy-1 (<222) and trim (>232) untouched. entry-call and take-profit-options-early remain in triggered state and will not fire without a re-arm — correct, since the option leg is already closed out.
- msft-diversifier-swing: flat on the day (-0.02%), entry-call triggered/dormant, no rule near. tsm-swing-contest: +0.66%, trim re-armed after the 9/30 scale-out at $460.99; price below $460 so nothing fired. crm-fade-swing: archived, no rules.

Lessons
- Pullback-only entries are earning their keep in this regime when the trigger is an event, not a trend break: both fills today landed within 0.4% of the session low on narrative-driven flushes (MU capex guidance, GOOGL Gemini report) and both theses were intact by the close. This is the opposite of the old "pullback-only deadlock" failure recorded for AMD/TSM — the difference is that the thresholds sit just under market, not 10-20% below.
- The re-arm era has a new recurring defect: a tranche worth ~8% of equity under a cap of 0.10 is not a cap at all, because one fire leaves the rule still eligible. I have now found this on three strategies (AAPL, AMZN, GOOGL). The rule of thumb to apply everywhere: every cap must sit at least one tranche below the weight it is protecting.
- Cash discipline is now the binding constraint, not rule logic. At ~99% deployed, the next $8,000 fire borrows. Tranche size has to be read against available cash, not just against target weight.

Proposals
- Revision #133, googl-swing v2 (queued, awaiting your approval): tranches cut from $8,000 to $5,000 (~5% of equity); dip-buy-1 cap 0.10 -> 0.09, dip-buy-2 cap 0.18 -> 0.13, trim floor 0.08 -> 0.10, trim size $6,000 -> $5,000. All price levels, the <308 stop, the >365 take-profit and the invalidation test are unchanged. This mirrors the fix already applied to aapl-swing v2 and amzn-swing v2. No other revisions proposed — the remaining strategies' caps bind correctly or their levels are intentionally out of reach.
- Not proposed but worth your judgment: with cash at $844, you may want to disarm or shrink one of the remaining $8,000 entry tranches (META, TSLA, NVDA, AMD) rather than let the first one that triggers be funded on margin.

**Day summary.** A quiet, broadly green session for the book: equity $100,239.91 with $16,844.37 cash (~17% uninvested), seven of eight positions in the green on cost. Day changes: AAPL +1.10%, AMD +0.69%, AMZN +1.01%, MSFT +0.77%, NVDA +0.51%, TSLA +0.56%, TSM -0.15%, META -1.84%. One trade fired, on TSM: trim-strength sold $5,000 (10.846 sh) @ $460.99 at 15:37 ET, a good fill 0.9% above the $456.27 close, and the rule re-armed at 17:07 ET after its 90-minute cooldown. Pending queue was empty at session start; I have added two revisions to it.

**Per-strategy check.**
- *tsm-swing-contest (v10, 26.08 sh @ $433.27, $11,918, 11.9% of equity, +$617).* Thesis confirmed: nine sessions of grinding higher from $434.67 (9/18) to today's $461.87 high with no 2%+ down day — the lowest-beta anchor behaving exactly as written. The trim executed correctly and at a good price, but its geometry is wrong: a 10% floor with a $5,000 tranche (~5pp of a $100k account) took the position from ~17% to 11.9%, i.e. *below* its own 15% target, and the re-arm means any print above $460 tomorrow would cut the "largest, most patient anchor" to ~7% purely because price rose. Revision proposed. Entries <447/<435 sit 2-5% below market; stop <418 is ~4.5 ATR away and remains sound.
- *aapl-swing (v1, 48.00 sh @ $333.31, $15,982, ~15.9% of equity, -$18).* Price recovered +1.10% to $333.02 after the narrative-driven 9/29 drop to $329.40; still inside the $329-341 band, position essentially flat. The defect is on the entry side: dip-buy-1 has a 10% cap but an $8,000 tranche, so on 9/29 it fired twice in 90 minutes ($334.53, then $332.09) and took AAPL from zero to 15.9% in one day, past the 15% target, with a third permitted fire able to breach the 20% max. Revision proposed. Note dip-buy-1's condition is true again at today's close (weight is the only thing blocking it), which is exactly why the sizing fix matters now.
- *meta-swing (v1, 11.11 sh @ $719.86, $8,025, ~8.0%, +$25).* The day's only loser (-1.84% to $725.18), and the highest-ATR name as advertised (~$30, ~4%). It has round-tripped from the 9/24 high of $779.82 back into the mid-$720s, which is why the 9/28 entry at $719.86 was a pullback buy rather than a chase. Weight is at the 8% target; dip-buy-1's 5% cap now correctly blocks further buying, so no sizing defect here. Stop <615 sits below the entire post-9/8 breakout. No action.
- *googl-swing (v1, no position).* GOOGL closed $344.08 after a $352.60 intraday spike on elevated 32.2M volume — a wide, reversing range. dip-buy-1 (<338) has not triggered because price has held above it since 9/29's $338.02 low. Zero position here is the rule working, not failing, but note this strategy shares the same $8,000-tranche/10%-cap arithmetic as AAPL; if it fires it will produce the same one-day overshoot. Flagged for the next revision cycle rather than pre-emptively rewritten.
- *nvda-momentum-swing (v11, 10.77 sh @ $209.38, $2,459, 2.5%, +$204).* Closed $228.38 (+0.51%), the fourth straight session in $224-233. trim-strength (>232) sold $6,000 @ $232-ish earlier in the week, which is why weight is now only 2.5% against a 15% target — the trim floor of 8% with a $6,000 tranche is the same exit-side defect as TSM, only already realised. dip-buy-1 (<222) is 2.8% below market. entry-call and take-profit-options-early remain triggered/dormant. Worth a sizing revision, but I ran out of tool budget to verify the option leg's current state, so I am not proposing one blind.
- *amd-swing (v9, 13.37 sh @ $598.50, $8,166, 8.1%, +$166), amzn-swing (v2, 65.07 sh @ $245.90, $16,177, 16.1%, +$177), msft-diversifier-swing (25.00 sh @ $499.96, $12,826, 12.8%, +$326), tsla-swing (22.10 sh @ $361.93, $7,843, 7.8%, -$157), mu-oversold-swing (flat).* All inside their stated ranges with no rule within reach; AMD's deliberately-low entry ceilings (<600/<560) and AMZN's post-v2 $5,000 tranches are behaving as designed. AMZN at 16.1% is above its 15% target and its dip-buy-1 cap of 8% correctly blocks adds. No action.
- *crm-fade-swing.* Archived, no rules, nothing to review.

**Lessons.** The durable conclusion today is that tranche size and weight bound are one mechanism, not two, and the error is symmetric. A cap only bounds a position if the tranche is small relative to it; a trim floor only bounds a trim if the tranche is small relative to the floor-to-target distance. With ~$100k equity an $8,000 tranche is ~8pp of weight, so a 10% cap permits ~16% after one fill and ~24% after a re-armed second — that is precisely what happened to AAPL on 9/29 (0% → 15.9% in one session). The mirror case appeared today on TSM: a 10% floor with a $5,000 tranche cut a 17% anchor to 11.9%, below its own target, with the rule re-armed to keep cutting on strength. The design rule I am adopting for all re-arming rules: cap + tranche ≤ target weight, trim floor ≥ target weight, tranche ≈ 4-5% of equity at most, 180m cooldowns on dip buys, ≤2 fires/day. NVDA (2.5% vs 15% target, ground down by an 8%-floor/$6,000 trim) and GOOGL (unfired but carrying the AAPL geometry) are the two remaining strategies with this flaw and are next in the queue. Two memory writes were rejected today for exceeding the 500-character limit, so these conclusions live only in this report — I will re-write them shortened at the next opportunity.

**Proposals.** Two revisions queued for approval, both sizing-only with every price level and fundamental thesis left untouched: **#129 aapl-swing v2** — tranches $8,000 → $5,000, dip-buy-1 cap 10% → 9%, dip-buy-2 cap 18% → 13%, cooldowns 90m → 180m, dip-buy-1 3 → 2 fires/day, trim floor 8% → 10% at $5,000, so cap + tranche stays inside the 15% target and 20% max. **#130 tsm-swing-contest v11** — trim floor 10% → 15% (the target), trim tranche $5,000 → $4,000, 3 → 2 fires/day, so trimming only occurs at or above target and one fire moves weight ~4pp instead of ~5pp from an already-below-target base. No new positions, no level changes, and no proposal for NVDA/GOOGL until I can verify their current state properly.

**Day summary**
A risk-off tape (S&P 500 -0.8%, 30-year Treasury yield at a 24-year high) produced a split book. Equity closed at $99,871 with $11,844 cash (~88% deployed). Winners: META +3.24%, TSM +0.91%, AMZN +0.21%. Losers: AAPL -2.66%, TSLA -1.29%, NVDA -0.72%, AMD -0.05%, MSFT -0.05%. Only one strategy acted: aapl-swing's dip-buy-1 fired twice (13:37 ET, $8,000 @ $334.53; 15:07 ET after a 90-minute re-arm, $8,000 @ $332.09), building a 48.00-share AAPL position at an average cost of $333.31 — roughly 1.2% above the $329.40 close, since AAPL faded all day and closed at its low ($328.70–$337.06 range, 36.1M shares). The AAPL weakness was narrative rather than fundamental: a Bank of America note flagging META's shopping agent as a threat to iPhone-ecosystem commerce, plus reports of product/management-process changes under CEO Ternus. Notably, the same META story that hurt AAPL is part of why META was the book's best performer — the two positions are now partly on opposite sides of one news item. No pending items; the day's only rule state change of consequence is that AAPL is now at ~15.8% of equity versus its 15% target.

**Per-strategy check**
- aapl-swing (v1, 48.00 sh @ $333.31, ~15.8%, -$180): thesis said "lowest realized volatility of any name considered, ~2.2% ATR" — today's -2.66% one-way fade is at the edge of that assumption, though a single session doesn't break it. Rules behaved exactly as written but the *sizing geometry* is off: two $8,000 tranches (~8% of equity each) against a cap of `position_weight < 0.10` took the position from flat to above its 15% target in 90 minutes, and a third fire was blocked only because the cap finally bound. The stop (<$310, ~5.9% / ~2.5 ATR below) and dip-buy-2 (<$322) remain sensibly placed. The entries were not wrong in level — $334.53 and $332.09 are inside the intended pullback zone — but the pace left no dry powder for the deeper tranche.
- amzn-swing (v2, 65.07 sh @ $245.90, ~16.1%, +$50): the 9/28 sizing revision to $5,000 tranches is the correct shape and dip-buy-1 re-armed today without firing (close $246.67, just above the <$246 trigger). Working as intended.
- meta-swing (v1, 11.11 sh @ $719.86, ~8.2%, +$199): entered 9/28 at $719.86, closed +3.24% at $738.79. At target weight; dip-buy-2 (<$670) and trim (>$800) both far away. No action correct.
- amd-swing (v9, 13.37 sh @ $598.50, ~8.1%, +$119): flat day, at the 8% cap on dip-buy-1, so nothing could fire. Trim at >$650 and stop at <$520 unchanged.
- tsm-swing-contest (36.93 sh @ $433.27, ~16.9%, +$865): best open P/L in the book and the steadiest performer again today (+0.91%) — the low-beta anchor role keeps verifying.
- msft-diversifier-swing (v4, 25.00 sh @ $499.96, ~12.7%, +$217): flat at $508.65-ish; dip-buy-1 (<$508) is within ~0.2% and is the most likely next fill in the book. Its $8,000 tranche against a `< 0.10` cap carries the same overshoot risk AAPL just demonstrated.
- nvda-momentum-swing (v11, 10.77 sh @ $209.38, ~2.5%, +$194): quote $227.21. Two option exits sit in `triggered` state and the stock is only ~2% under dip-buy-1 (<$222). Position is far below its 15% target after prior trims, which is acceptable but means NVDA is now a minor contributor.
- googl-swing (v1, no position): closed $340.92, dip-buy-1 needs <$338 — it traded $338.02 intraday and missed by two cents. Thresholds are live and realistic; nothing wrong.
- tsla-swing (v1, 22.10 sh @ $361.93, ~7.8%, -$193): -1.29%, above the 6% cap so dip-buy-1 is blocked; stop <$335 intact.
- mu-oversold-swing: flat, dormant thresholds as designed.
- crm-fade-swing: archived, no rules. Correct.

**Lessons**
1. The durable lesson of the day is a sizing lesson, not a market call: on a ~$100k book, an $8,000 tranche is ~8% of equity, so a re-arming buy rule capped at `position_weight < 0.10` can legally fire twice and land near 16% — above the strategy's own 15% target and, with dip-buy-2's `< 0.18` cap, potentially in the 26–34% range against a declared `max_weight` of 0.20. The cap must be set below (max_weight − tranche), not at it. amzn-swing v2 already uses the corrected shape; aapl, googl, msft, nvda, tsla, meta, amd and tsm swings do not.
2. Re-arming ladders average *into* a trend, not into a washout. Both AAPL fills printed above the close on a day that closed at its low. That is the expected cost of mechanical scale-ins and is tolerable — but it argues for smaller tranches and wider spacing between fires rather than a 90-minute cooldown on a name whose whole thesis is low volatility.
3. Correlation inside this book is not what the theses assume. One BofA note simultaneously drove AAPL to the worst performance and supported META, the best. Diversification by sector label is weaker than diversification by exposure to the same news flow.
4. I failed to persist any of these conclusions to curated memory this pass: three memory writes were rejected for exceeding the 500-character limit and I ran out of tool calls before rewriting them shorter. The operational lesson for myself is to draft memory entries at roughly 400 characters and to write them early in the pass, not after all the research. The findings above are therefore recorded only in this report.

**Proposals**
No strategy revision was queued this pass — I hit the tool-call limit before I could submit one, and I will not fabricate having done so. The revision I judge warranted, and intend to submit at the next opportunity, is to aapl-swing: keep the thesis, price levels, stop (<$310) and take-profit (>$362) unchanged, but cut the dip-buy tranches from $8,000 to $5,000, tighten dip-buy-1's cap from `position_weight < 0.10` to `< 0.10` with the position already past it (i.e. effectively retire it for now) and set dip-buy-2 at `position_weight < 0.15` so the deeper tranche cannot push past the 20% max, and lengthen the re-arm from 90m to 180m. The same tranche-versus-cap correction should follow for msft-diversifier-swing (nearest to firing, <$508 is ~0.2% away) and then googl-swing, nvda-momentum-swing and tsla-swing. Nothing in today's data justifies changing any price threshold, stop, or fundamental thesis — the geometry is the problem, not the levels.

**Day summary.** A macro risk-off session, not a stock-specific one: media and index data point to the 10-year Treasury yield near a two-decade high, rising odds of an October Fed hike, plus Iran/oil tension; the S&P 500 fell ~0.4% and the Nasdaq 100 about 1%. The book's ladders read this as a buying opportunity everywhere at once. Nine fills: six dip-buys across six strategies (AMZN twice, MSFT, TSLA, TSM, AMD, META), plus NVDA's trim and its option take-profit. Roughly $48,000 was deployed in one day, taking the account to ~72% invested with ~$27.8k cash, above the mandate's 40-70% band. Equity ended ~$99,774. Position day-changes: NVDA +1.68% and TSM +0.50% green; META -4.79%, TSLA -3.94%, AMD -3.61%, AMZN -1.41%, MSFT -1.35%. Net open P/L is slightly positive (+$1,074) mostly on the older TSM and MSFT lots; today's new entries in META, TSLA and AMD are underwater by construction because each fill happened mid-slide.

**Per-strategy check.**
- *nvda-momentum-swing (v11)* — the day's clear win and fully consistent with thesis. NVDA gapped to $229.70, hit $233.21 and closed $228.86 (+1.68%) on 135M shares, the heaviest non-expiry volume in weeks. take-profit-options-early (>227) closed the 2 Oct-16 $220 calls at $15.10 against a $4.65 entry (~+225%, ~$2,090 on a $930 budget), and trim-strength (>232) sold $6,000 of stock at $232.53, leaving 10.77 sh (~2.5%). The deliberately tight option exit — the geometry v10 built specifically to stop theta bleed — did exactly its job. No change needed.
- *tsm-swing-contest (v9)* — thesis validated, sizing not. TSM was one of two green names on a red tape, confirming its low-beta anchor role. But dip-buy-1 fired at $445.63 for $8,000 while weight was already 8.7%, landing the position at 16.8% — past its own 15% target — and two more fires were still permitted the same day. Revision proposed.
- *amzn-swing (v1)* — levels fine, arithmetic broken. dip-buy-1 fired at 13:37 @ $245.89, re-armed after 90m and fired again at 19:52 @ $245.91: two fills eleven cents apart, zero to 16.1% weight in one session against a 15% target. Close $246.15 sits mid-range ($244.30-260.11), so nothing about the thesis failed. Revision proposed.
- *msft-diversifier-swing (v4)* — dip-buy-1 filled $8,000 @ $503.97, bringing the position to 25.00 sh @ $499.96, ~12.8% weight, +$241 open. The pullback from the 9/25 breakout close ($516.17) into the low $500s is precisely the "scale into weakness" case the v4 thesis described, and MSFT again fell least among the non-green names (-1.35%). entry-call skipped once more as unaffordable — that leg has now failed to price for weeks and should eventually be deleted rather than left to log skips. Same cap-plus-tranche overshoot risk exists here but the position is not yet past its 15% target; leaving it for now rather than proposing three near-identical revisions in one pass.
- *meta-swing (v1)* — first entry, 11.11 sh @ $719.86 (8.0% = exactly target). The day was an outside-down reversal (open $750.42, high $750.58, low $713.19, close $715.62), so the fill was mid-slide rather than at a support test — the cost of a static price trigger on a 4%-ATR name. dip-buy-2 at <$670 and stop at <$615 remain correctly placed below the 9/21 breakout. No change.
- *amd-swing (v9)* — dip-buy-1 filled $8,000 @ $598.50, 13.37 sh, ~8.1% weight, +$121. The v9 thesis explicitly set entries well below market because AMD was extended after +25% in two weeks; price came to the level and the rule worked as designed. Close about -3.6%. No change.
- *tsla-swing* — dip-buy-1 filled $8,000 @ $361.93, 22.10 sh, ~7.9% weight, -$101. TSLA fell from a $386.70 high on 9/23 to a $356.80 low today, so the $362 trigger caught the third day of a slide; the fill is near the low of the last two weeks. No change.
- *aapl-swing, googl-swing* — no fills and correctly so: AAPL $338.40 versus a <$335 trigger, GOOGL $342.75 versus <$338. Both sat just above their thresholds all day. Inaction is the right outcome.
- *mu-oversold-swing* — flat, entries far below market, nothing monitored today. Unchanged from prior conclusion to stay out through the FQ4 print.
- *crm-fade-swing* — archived, no rules, nothing to review.

**Lessons.** Two written to memory. First, a sizing rule that generalizes across the book: on a re-arming dip-buy, the weight cap plus the tranche size must stay under the target weight, or the position overshoots by construction — an $8,000 tranche is ~8pp of this account, so a 10% cap legally permits an 18% position on a 15% target, and today it produced AMZN at 16.1% and TSM at 16.8%. The fix is smaller tranches, not different prices. Second, and more important strategically: diversifying by ticker does not diversify the trigger. Six "independent" per-name ladders all fired within one session because the move was rate-driven multiple compression that hit every long-duration growth name simultaneously. The real constraint is an aggregate cash floor, not per-name caps, and I should expect the whole book to buy at once on macro days.

**Proposals.** Two revisions queued for approval, both purely mechanical — every price level, stop, target and fundamental thesis left untouched. #125 amzn-swing v2: tranches to $5,000, dip-buy caps to 8%/11%, cooldowns to 180m, dip-buy-1 to 2 fires/day, trim matched at $5,000 with a 10% floor. #126 tsm-swing-contest v10: same treatment with caps at 9%/13%. Neither takes effect unless you approve it on the Pending page; I have no path to apply either myself.

**Day summary — 2026-09-25 (paper, equity $98,232.56, cash $74,824.48 = 76.2%)**
No trades, no observations. One position did all the work: MSFT closed $516.17, +3.66%, high $519.40, on 37.1M shares — the heaviest non-expiry session of the month — on the expanded Copilot release (Home / Code / Autopilot), two days after a sell-side upgrade to Buy with a $575 target. That close breaks the $486-518 band MSFT had traded since 8/28. The semis legs were inert: NVDA +0.22% to $225.07, TSM -0.17% to $450.37 on 5.7M (very light), AMD (no position) +0.22% to $630.63 on 17.0M, decelerating from 25-44M last week. Open P&L across the four legs is about +$1,978, of which the 2 NVDA Oct-16 $220 calls are +$990 (+113%) on a $930 cost. The notable fact of the day is that the book's designated diversifier was its only mover, which is exactly what it was bought for — but it is also the smallest of the three stock legs.

**Per-strategy check**
- msft-diversifier-swing (v2, confirm): thesis working, rules stale. Weight is 4.8% against an 8% target (9.128 sh @ $492.98) because entry-initial was consumed on 9/8 and add-on-dip < 478 has been unreachable for 17 sessions. take-profit > 520 now sits 0.7% above market and sells the whole leg — the strategy is on track to book a half-size +5.5% win at the moment its premise starts paying, leaving the book all-semis plus cash. stop-loss < 465 is 9.9% (~5 ATR) below market. Revision proposed, see below. Pending #69 (entry-call, fired 9/16 at $490.30) is stale: approved today it would buy a call struck ~4% off $516, not off $490 — a different trade than the rule intended. My recommendation is to reject it and leave the rule armed for a real dip.
- nvda-momentum-swing (v10, confirm): stock leg sound, option leg still bleeding. Shares 36.574 @ $218.74 = 8.4% weight, +$227, all stock levels far away (TP $248 = +10%, stop $204 = -9%). The calls are the problem: v10 lowered the early exit from > 230 to > 227 on 9/23, and the two sessions since printed highs of $224.94 and $226.94 — short by $0.06. The calls went $13.05 (9/22) → $10.50 (9/23) → ~$9.60 today, roughly $690 of open gain returned on a sub-1% stock move, with ~21 DTE. No revision filed: v10 is two days old, today's premium was actually flat (+0.41%), and re-cutting the same level a third time would be reacting to noise rather than to a measured divergence. If the stock stalls in $223-227 for another two or three sessions while premium keeps eroding, the honest fix is a time-based or premium-based exit, not another price notch — flagging that now so it is not a surprise.
- tsm-swing-contest (confirm): behaving exactly as assigned. 18.976 sh @ $421.58 = 8.7% weight, +7.2%. -0.17% today on the lightest volume in weeks; take-profit > 460 is +2.1% away and remains the book's next likely fill. Given confirm-mode approval lag, read it as an exit near $455-465. Stop < 420 is ~6.7% below and still structurally sound. No change warranted.
- amd-swing (v8, confirm): flat since the 9/21 take-profit at $609.58. AMD closed $630.63 today; the armed entry window $465-482 is ~24% below market and monitors nothing. This is the fourth session I have logged that fact. I am not proposing a re-anchor: the user's standing preference is pullback-only, and chasing a name up 24% from the exit would violate it. The strategy is correctly dormant rather than broken.
- mu-oversold-swing (v8, confirm): flat since the 9/21 exit at $1,038.41. entry-dip $915-960 is dormant, FQ4 earnings land around 9/30. Staying flat through the print is the right posture after a +14.9% four-day swing.
- crm-fade-swing: archived, no rules, nothing to review.

**Lessons**
Written to memory: an option profit exit placed above the recent closing range is not an exit, it is a bet on a new high. The NVDA calls have now demonstrated this twice — > 230 was approached three times and cleared zero times, and > 227 was missed by six cents twice — while theta removed real money both times. Future option exits should sit at or below the last close and be sized off premium behavior, not off a stock chart level. Also recorded: MSFT's diversification premise is now verified in both directions (it held up on the 9/10 semis drawdown and led on today's rally), which upgrades it from assumption to observed behavior; and the pullback-only deadlock has now appeared in a fourth name, which is why today's fix is a retest window rather than a lowered target.

**Proposals**
One revision queued, #88, msft-diversifier-swing v3 (requires your approval on the Pending page; I cannot apply it): take-profit 520 → 540 (the measured move of the broken $486-518 range, ~+9.5% on entry); stop-loss 465 → 488, just under the 9/10 and 9/16 lows, so the exit trigger is a failed breakout rather than a number 5 ATR away; and a new add-on-retest rule buying $3,500 in the $498-512 window with a weight cap of 5.5%, which lifts the leg to ~8.3% on a pullback into the broken level and cannot fire above $512. Option rules are kept intact, with stop-loss-options aligned to < 488 and take-profit-options deliberately left at > 520, below the new share target, because the calls expire and the shares do not. Separately, and needing no revision: I recommend rejecting pending #69 as a stale trigger.

**Day summary.** No trades, no observations. Equity $98,033.60, cash $74,824.48 (76.3%) — the book is still mostly cash after the 9/21 AMD and MU take-profits. A quiet, mixed session: TSM +1.00% to $451.05 (vol 6.3M, the lightest in weeks) and AMD ran +2.38% to $629.26, while our three remaining long legs were flat-to-down — NVDA -0.41% to $224.58 on 70.1M shares (the lightest volume in a month), MSFT -0.53% to $497.93, and the NVDA Oct-16 $220 calls -7.05% to $9.76. Open P&L across the four legs remains positive (+$1,779 total), and nothing today touched a thesis-level invalidation on any strategy. The two things genuinely worth attention are not price moves: the option leg's continuing decay and a pending order that has gone stale.

**Per-strategy check.**

- *nvda-momentum-swing (v10, confirm).* The stock leg (36.574 sh @ $218.74, $8,204, 8.4% — target met) is behaving exactly as written; nothing near the $248 take-profit or the $204 stop. The option leg is where the strategy is measurably diverging from its own reasoning. v10 was written yesterday specifically because the >$230 early exit had been approached three times and cleared zero times; it lowered that exit to >$227. Today NVDA's high was $224.90 — a fourth consecutive session printing a high in the $224.90-$229.98 band with no option exit fill. Meanwhile the calls went $13.05 (9/22) → $10.50 (9/23) → $9.76 today: the stock is only -1.9% from its 9/22 close but the calls are -25%, roughly $660 of open gain surrendered in two sessions with ~22 days to expiry. The leg is still +110% (+$1,022 on $930), so this is about protecting a large realized-able gain, not a loss. The structural problem is the geometry v10 left in place: the exit sits ~1.1% above market while the option stop sits ~4.6% below ($214), so on a decaying asset most of the premium can bleed away in the gap between the two without any rule acting. v10's own logic ("every day it is not hit costs theta on a leg whose gain is now the largest single P&L item in the book") applies with more force today than yesterday, because two more days of theta have now been paid.
- *msft-diversifier-swing (v2, confirm).* The diversifier premise held again in a mild way — MSFT closed -0.53% while AMD rose 2.38% and TSM 1.00%; low correlation, today on the negative side. The stock leg (9.128 sh @ $492.98) is +0.7%, and no stock rule is near: add-on <$478, take-profit >$520, stop <$465 against a $497.93 close. The problem is pending order #69: the entry-call rule fired on 2026-09-16 when MSFT closed $490.30, and it has now sat in the queue for six sessions while the stock rose ~1.6% to $497.93. Because confirm-mode option orders are re-priced at approval, approving it today would not buy the contract the rule intended — an OTM-4% call off $497.93 strikes near $518 rather than the ~$510 implied at trigger time, with six fewer days of time value for the same $1,200. The trigger condition (price < 490) is no longer true. My recommendation is to decline #69 and let the rule re-arm for the next genuine dip below $490.
- *tsm-swing-contest (v8, confirm).* Working as designed and the best-behaved leg in the book. 18.976 sh @ $421.58 = $8,552 (8.7%, +7.2%). Closed $451.05 on 6.3M shares — participating in the sector with roughly a quarter of AMD's amplitude, its assigned low-beta anchor role. The >$460 take-profit is ~2.0% away and remains the book's most likely next fill; given confirm-mode approval lag (MU filled 1.1% below its own trigger on 9/21), read it as "exit near $455-465". The <$420 stop, repaired in v8, sits ~7% below and just under average cost. No change warranted.
- *amd-swing (v8, confirm).* Flat since the 9/21 take-profit at $609.58. AMD closed $629.26 (+2.38%, high $630.79, vol 24.5M), a new high above the 9/22 close. The armed entry-dip window ($465-482) is now ~24-26% below market and monitors nothing — the fourth instance of the pullback-only deadlock on this name. I am deliberately not proposing to raise it: the user's standing instruction is that a dormant threshold on a satellite swing is an acceptable outcome (missing a satellite trade has no real cost), and re-anchoring upward after a +24% five-session run is exactly the level-chasing v7/v8 warned against. Noted, not fixed.
- *mu-oversold-swing (v8, confirm).* Flat. MU quoted $1,080.53, now ~12% above the take-profit that closed the swing at $1,038.41. The entry-dip window ($915-960) is ~13% below market and dormant, correctly so. FQ4 earnings land around 09-30; staying flat through the print rather than chasing remains the plan.
- *crm-fade-swing.* Archived, no rules, no action.

**Lessons.** Two written to memory today. First, a stock-level note that NVDA's option leg has now failed to reach an exit trigger on four consecutive sessions while giving back ~$660 of gain in two — evidence that an exit set just above the recent high is a lottery ticket on a new high, not an exit plan, and that lowering it by $3 was not enough. Second, a new durable lesson: confirm-mode pending option orders never expire and are re-priced at approval, so a stale one silently becomes a different trade than the rule authorized. The standing discipline I will apply going forward — if the triggering condition is false at review time, decline it and let the rule re-arm. This is the option-specific corollary of the 2026-08-17 lesson that a trigger is only a proposal, never a commitment.

**Proposals.** None queued this pass. I reached the tool-call limit for this reflection before I could submit the NVDA v11 revision I had concluded was justified, so I am recording the intent here rather than leaving a half-formed proposal on the Pending page. The case for it is documented above and in memory: v10's >$227 option exit has now been missed a fourth time while the leg lost 25% of its value in two sessions, and the ~1.1%-above / ~4.6%-below geometry around a wasting asset is the measurable divergence. The fix I would propose is to tighten the option-only exit to roughly NVDA > $224-225 — inside the band the stock has actually closed in for four sessions, so an ordinary up-session realizes the gain — and to raise the option stop from $214 toward ~$218-220 so that a further slide protects a meaningful share of the remaining premium instead of acting after most of it is gone. Stock levels, sizing and the fundamental thesis would be untouched; this separates the expiring leg from the non-expiring one, which is what v10 set out to do but under-delivered. Two items need the user's hand in the meantime: pending #69 (MSFT call) should be declined as stale, and the NVDA option decision is live now, not at $248.

**Day summary.** No trades today. Equity $98,213.63, cash $74,824.48 (76.2%) — still cash-heavy after the 9/21 AMD and MU take-profits. The five-session semiconductor run paused: NVDA closed $225.51 (-1.47%, range $224.02-228.95) on 80.2M shares, the lightest volume of the whole stretch; AMD $614.61 (-1.47%, 15.8M, barely a third of Monday's 44.5M); TSM $446.74 (-1.16%, 8.6M). MSFT was the only green leg, $500.59 (+0.52%), again inverting the sector — its diversifier role working in the intended direction for once. Volume shrinking as prices fall reads as an orderly give-back rather than distribution, and nothing today touched a thesis-level invalidation. The sharpest single move in the book was the NVDA Oct-16 $220 calls: marked $10.50, -21.67% on the day (-$510 of open gain) while the underlying fell 1.5% — the leverage cutting the other way for the first time. The sentinel logged repeated `previous_close` NaN quote-validation errors across all five active strategies on both the 18:39 and 19:09 passes; these are data-feed faults in the quote object, not rule failures, but they mean I cannot assume every rule was evaluated on every pass today.

**Per-strategy check.**
- *nvda-momentum-swing (v9, confirm, target 8%)* — Stock leg 36.574 sh @ $218.74 = $8,245, 8.4% of equity, target met, +$245. No rule fired; correctly so, as $225.51 sits between add-on (<214) and take-profit (>248). The measurable divergence is the option leg. v9 added the >$230 early close_options on 9/17 precisely to give the calls a reachable exit, and NVDA has now printed highs of $228.50, $229.98 and $228.95 on three consecutive sessions — within $1.05 of the trigger, cleared zero times — while the calls round-tripped $7.60 → $13.05 → $10.50. With 23 DTE the exit is not "$248", it is "somewhere near $227-230", and the >$230 formulation requires a marginal new high to pay out. The stop side is the same defect mirrored: at <$204 a $220 call with three weeks left would be worth roughly $1, so the stop only fires after the premium is already gone. Proposed v10 (#86) below.
- *amd-swing (v8, confirm, target 4%)* — Flat since the 9/21 exit at $609.58. AMD fell to $614.61 on collapsing volume; the ladder (entry-dip $465-482 armed, entry-step <$505 triggered) sits 22-25% below market and monitors nothing. This is the fourth consecutive session I have recorded the pullback-only deadlock here. I am deliberately not proposing a re-anchor: the strategy's own v7/v8 thesis commits to not chasing levels, the user's standing preference is to accept non-fills on satellites, and a 25% gap after a +25% weekly sector run is not evidence the levels are wrong — only that the setup is absent. Support to watch if it does retrace: the unfilled 9/21 gap $559.91-583.88, then the 9/18 low $541.53.
- *tsm-swing-contest (v8, confirm)* — 18.976 sh @ $421.58 = $8,477, 8.6%, +6.0%. First down day after four up, and again the smallest amplitude in the book on a sector-wide decline, matching its assigned low-beta anchor role in both directions. The >$460 take-profit is ~3.0% away and remains the book's next likely fill; given confirm-mode approval lag (the 9/21 MU fill came 2.4% below its own trigger) read that rule as "exit near $455-465" rather than $460. The <$420 stop is ~6% below and remains sound.
- *msft-diversifier-swing (v2, confirm, target 8%)* — 9.128 sh @ $492.98 = $4,568, 4.7%, +1.4%. Stock rules untouched (TP >$520, add-on <$478, stop <$465). The live issue is Pending #69: the entry-call (price < 490 → buy_call $1,200, dte>=30, otm 4%) fired on 9/16 and has sat unapproved for five sessions. MSFT is now $500.59, 2.2% above the rule's own condition, and the contract is re-priced at approval — so approving it would buy a setup the rule never authorized. Recommendation: decline #69 and re-arm rather than hold it.
- *mu-oversold-swing (v8, confirm, target 4%)* — Flat since the 9/21 exit at $1,038.41 (+14.9%). MU $1,071.88, above the exit; entry-dip $915-960 is ~12-14% below and dormant. FQ4 earnings land around 09-30 and the plan of record is to stay flat through the print. Correct no-action day.
- *crm-fade-swing* — archived, no rules, no action.

**Lessons.** Two written to memory. First: an option leg whose only exits are stock-price levels tends to decay rather than exit — the >$230 trigger was approached three sessions running and never hit while the position surrendered 21.67% in a day; short-dated legs need an early take-profit *inside* the stock's recent closing range and a stop around 2-2.5 ATR below market, kept separate from the share leg's wider target. Second: confirm-mode option orders do not age out of the pending queue, so a trigger can drift far out of its own condition (#69, five sessions, now 2.2% offside) and should be declined rather than approved late. I also tried to refresh the NVDA and TSM stock files but both are at their 3,000-character budget; they need pruning next pass before new entries will fit, so today's price detail lives only in this report.

**Proposals.** One queued: **#86, nvda-momentum-swing v10** — move `take-profit-options-early` from `price > 230` to `price > 227`, and `stop-loss-options` from `price < 204` to `price < 214`. Both changes are option-only; the share leg's $248 target and $204 stop are untouched, as are entry-call (<212), add-on-dip (<214, weight<0.063) and the fundamental thesis. Note the new option stop deliberately coincides with add-on-dip, so a real dip de-risks the expiring leg while adding to the non-expiring one. This is a reduction in the profit target, not a chase. It awaits your approval on the Pending page; I cannot apply it myself. No revision proposed for amd-swing, mu-oversold-swing, msft-diversifier-swing or tsm-swing-contest — their dormancy today is the designed outcome, not a defect.

**Day summary — 2026-09-22 (paper, equity $98,881, cash $74,825 / 75.7%)**
No trades and no observations today. The semis melt-up extended a fifth session: AMD $623.77 (+1.34%, high $624.52) on 26.6M shares versus 44.5M on 9/21 — price still pushing on clearly decelerating volume; TSM $451.89 (+1.52%) on 9.0M; NVDA $228.87 (+0.66%) with a high of $229.98; MU $1,096.16. MSFT was the lone decliner at $498.00 (-0.72%), after opening at $507.52 and giving the whole gap back — the non-semis diversifier again behaving as designed, currently on the negative side. Five-session sector context from external reporting: AMD +24.75%, INTC ~+25%, NVDA +7.78%, TSM +6.77%, with no new catalyst today beyond the 9/21 ~10% chip price-hike report. All four open legs are profitable: NVDA stock +$345, TSM +$558, MSFT +$44, and the NVDA Oct-16 $220 calls marked $13.05 versus $4.65 cost (+181%, +$1,680) — the book's largest single gain, now ~2.6% of equity.

**Per-strategy check**
- **nvda-momentum-swing (v9, confirm):** The day validated v9 with uncomfortable precision. NVDA's high of $229.98 came two cents under the >$230 `take-profit-options-early` rule that v9 added only on 9/17 to stop a profitable one-month call decaying into expiry. Rules behaved correctly — nothing fired. With 24 DTE and the calls at +181%, ~$230 is the live decision point, not the $248 share target. No revision: the rule is right, it simply has not been touched yet.
- **tsm-swing-contest (v8, confirm):** Working exactly to thesis. TSM participated at roughly a quarter of AMD's amplitude, which is its assigned low-beta anchor role. Position 8.7% of equity (+7.2%). The >$460 take-profit is now only +1.8% away and is the book's most likely next fill; given confirm-mode approval lag it should be read as "exit near $455-465", not $460 exactly. The <$420 stop sits ~7% below, below blended cost — sound. No change needed.
- **msft-diversifier-swing (v2, confirm):** Stock leg fine (barely profitable, low-vol as claimed). The defect is in the queue, not the rules — see Proposals.
- **amd-swing (v8, confirm):** Flat since the 9/21 take-profit at $609.58 (+20.7% on the $2,000 leg). The strategy is now fully inert: `entry-step` is `triggered` and cannot re-fire without a manual re-arm, and `entry-dip` ($465-482) sits 23-25% below market. This is the third documented instance of the pullback-only deadlock on this name. I ran out of tool calls before I could finish drafting a re-anchoring revision, so I am flagging it rather than proposing an unverified one.
- **mu-oversold-swing (v8, confirm):** Flat and correctly dormant. MU at $1,096 is ~14% above the $915-960 entry window, and `entry-dip` is `triggered` anyway. FQ4 earnings ~09-30: staying flat through the print after a +15% run remains the right call.
- **crm-fade-swing:** Archived, no rules, no action.

**Lessons**
- Written to memory: confirm-mode pending orders decay, because the contract and price are set at *approval*, not at trigger. MSFT's `entry-call` fired 2026-09-16 at $490.30 and still sat unapproved today with MSFT at $498 — six sessions and +1.6% above its own trigger condition. Approving it now would buy a re-priced 4%-OTM call off a dip that no longer exists: a different and worse trade than the rule intended. Standing rule: an unapproved signal order older than ~2 sessions, or whose condition no longer holds, is EXPIRED — reject it and let the rule re-fire on a real dip. This is the same family as the 9/21 lesson where MU's take-profit filled 1.1% below its trigger after a two-hour approval lag.
- Also recorded: AMD's advance on decelerating volume (26.6M versus 44.5M) is loss of fuel, not yet reversal; first real support is the unfilled 9/21 gap at $559.91-583.88, then the 9/18 low of $541.53.
- Pattern confirmed: fixed take-profit targets over-deliver on gappy names (AMD exited at $609.58 against a $560 target), while the same rigidity leaves re-entry ladders stranded far below market once the move runs. The book is now 75.7% cash with only four legs monitored — high cash here is a consequence of untriggered ladders, not a sizing error, but on AMD it has become a monitoring gap.

**Proposals**
- No strategy revision queued today. TSM v8, NVDA v9 and MU v8 all matched their stated assumptions, and I will not propose a revision merely because I can.
- One action for the user on the Pending page: **reject option_order #69 (msft-diversifier-swing / entry-call, buy_call $1,200)**. Its condition (price < 490) was true on 9/16 and is now stale by six sessions and +1.6%; the contract would be re-priced at approval against a stock that has recovered. Rejecting it leaves the rule free to fire again on a genuine sub-$490 dip. I cannot reject it myself — I am advisory only, and every order passes the user's confirmation and the risk gate.
- Flagged for tomorrow, not yet proposed: amd-swing needs either a manual re-arm of `entry-step` or a re-anchored entry structure near the $560-584 gap shelf, otherwise the strategy monitors nothing at all. I want to verify the level against tomorrow's volume before drafting it.

**Day summary**
A broad semiconductor melt-up. A supply-chain report of a ~10% chip price increase sent AMD +9.95% to $615.52 — its first $1T market cap — on 43.3M shares, and dragged the whole book higher: NVDA +2.30% to $227.38 (104M shares), TSM +2.42% to $445.18, MSFT +1.59% to $501.61. Two take-profit rules fired and both filled: AMD sold 3.9612 sh @ $609.58 (entry $504.90 on 9/10, +20.7%, ~+$415) and MU sold 3.8718 sh @ $1,038.41 (entry $903.97 on 9/14, +14.9%, ~+$520). Equity closed at $98,559 with cash back up to $74,824 (75.9%). Both satellite swings are now closed, and the four remaining lines — NVDA shares, the NVDA Oct-16 $220 calls, TSM and MSFT — are all profitable. The calls, marked ~$12.10 against a $4.65 cost (+160%, +$1,490), are the single largest P&L in the book.

**Per-strategy check**
- amd-swing (v8): thesis validated end-to-end. v8 explicitly refused to raise the $560 target after a +11% run, calling that level-chasing; the gap-up open then delivered +20.7% instead of the planned +11%. The repaired <$500 stop was never needed. Strategy is now flat with entry-step in `triggered` state and the entry-dip window ($465-482) sitting ~23% below market — it monitors nothing until re-armed and re-anchored. No revision proposed: nothing is measurably wrong with a strategy that just completed its plan, and re-anchoring an entry ladder onto a +$110 one-day breakout is exactly the mistake its own thesis warns against.
- mu-oversold-swing (v8): also completed as designed, +14.9% on the satellite leg. One measurable imperfection, but it is execution, not rule design: the >$1,050 trigger was flagged at ~09:39 ET and the confirm-mode order filled at ~11:47 ET @ $1,038.41 — 1.1% below its own trigger and 2.4% off the day's high. Now flat ahead of FQ4 earnings (~09-30), which resolves the open risk question the v8 thesis had flagged. Staying flat through that print is the right default; a re-entry should wait for guidance and ideally a retrace into the $915-960 base.
- nvda-momentum-swing (v9): working as written. The stock leg is at ~8.4% weight with the add-on correctly dormant behind the weight<0.063 gate. v9's new take-profit-options-early (>$230 -> close_options) is now only 1.1% from being hit, which vindicates the v9 diagnosis that the old >$248 option exit was unreachable. Live watch item: in confirm mode that option exit needs prompt approval, because an at-the-money October call gives back gains quickly once the underlying stalls.
- tsm-swing-contest (v7): the one genuine divergence today. At $445.18 the <$398 stop is ~10.6% / ~5.4 ATR below market while the $460 target is 3.3% away — the open trade risks the full $470 gain plus 5.6% of capital to make 3.3%. This is the identical stale-stop defect already fixed on AMD and MU.
- msft-diversifier-swing (v2): +1.59% today, but it rose *with* semis rather than against them, so today says nothing new about the low-correlation premise. Rules untouched: add-on <$478, TP >$520, stop <$465 are all far away. The stranded item is Pending #69 — the entry-call (price<490, buy_call $1,200) from 9/16 — whose condition is no longer true at $501.61; approving it now would buy a contract the rule never actually authorized at this price. Recommend rejecting it rather than letting it age further.
- crm-fade-swing: archived, no rules, nothing to review.

**Lessons**
1. Fixed take-profits over-deliver on gapping momentum names. AMD's $560 target was reached by an open at $583.94 and filled at $609.58 — roughly double the planned gain. The discipline of *not* chasing the level after a +11% run was what allowed the gap to be captured.
2. Confirm-mode exit latency is a measurable, symmetric cost. Today's two exits were both flagged at ~09:39 ET and both filled at ~11:47 ET; the trending name (AMD) gained ~8% from the delay and the fading name (MU) lost 1.1% versus its trigger. Take-profit levels should be set so a 1-3% adverse drift still leaves the trade acceptable, and the trigger price must never be read as the realized price.
3. The sentinel note "queued analysis unparseable — left for human review" does not mean a rule failed — both such notices today ended in clean fills. This is the third confirmed instance of that pattern.
4. After a winning swing exits, the strategy's entry ladder becomes dead weight: AMD's and MU's entry windows now sit 12-23% below market. That is the correct dormant state, not a defect to be fixed by raising thresholds — but it does mean the book has no live buy triggers at all and is purely in harvest mode.

**Proposals**
- Queued revision #82, tsm-swing-contest v8: move stop-loss from <$398 to <$420, placing it just under the blended average cost ($421.58) and below the 9/17 open, which converts the trade to roughly break-even risk (worst case ~-0.4% on $8.4k instead of -5.6%). At ~2.9 ATR / 5.7% below the close, it is wider than TSM's worst recent session (-3.41% on 9/14), so the 9/03 error — a <$410 stop tagged intraday at $409.17 on a day TSM closed $417.02 and kept climbing — is not repeated. The $460 target is deliberately not raised, the $406-414 add-on window and its weight<0.063 gate are untouched, and no option legs are added. Awaiting your approval on the Pending page.
- No revision for amd-swing or mu-oversold-swing: both just executed their plans correctly and are flat. Re-anchoring either entry ladder onto today's breakout would be level-chasing.
- Housekeeping recommendation (needs your action, I cannot do it): reject Pending #69, the stale MSFT entry-call whose price<490 condition is no longer true at $501.61.

**Day summary (2026-09-18, paper account)**
Equity $97,182, cash $68,389 (70.4%), a second consecutive day with no fills. All six positions are in profit: MU +$432, TSM +$236, AMD +$214, NVDA +$109, MSFT +$11, and the NVDA 10/16 $220 call +$860 (about +92%). Day changes: MU +3.92%, AMD +2.70%, NVDA +1.34%, TSM +0.93%, MSFT -0.80%. Closes: AMD $559.82, MU $1,015.80, NVDA $222.27, TSM $434.24, MSFT $493.78. Semis extended the risk appetite that returned on 9/17 (AMD on its raised $3T TAM outlook, MU reclaiming $1,000 on its heaviest volume since 9/04 at 32.57M shares). MSFT moved opposite the semis for a third straight day, which keeps confirming its premise as a low-correlation non-semi diversifier -- it is simply on the negative-contribution side right now. Having no fills is by design, not a failure: every add-on is correctly dormant under the weight<0.063 gate, and the account is in a pure take-profit/stop-loss discipline phase.

**Per-strategy check**
- amd-swing (v7): entered 3.9612 shares @ $504.90, now +10.9%, weight about 2.3%. The intraday high of $559.91 missed the >$560 take-profit by $0.09 -- a normal discrete sampling outcome, and I am not lowering the threshold. Real defect found: the <$438 stop now sits 21.8% below market (about 6.4 ATR) and no longer protects anything. -> submitted v8 (#76).
- mu-oversold-swing (v7): 3.8718 shares @ $903.97, now +12.4%, weight about 4.0% (target reached). Take-profit >$1,050 is only 3.4% away while the <$866 stop is 14.7% below market (about 3.9 ATR) -- the asymmetry is badly inverted. -> submitted v8 (#77). Also flagged: FQ4 earnings around 09-30 fall inside the holding period, and a gap cannot be defended by a stop.
- nvda-momentum-swing (v9): stock leg 36.574 shares @ $218.74 (about 8.3%, target reached); option leg 2x $220C marked around $8.95 (+92%). The >$230 option take-profit added in v9 sits +3.5% from market, one of the closest rules in this account to triggering, and its design intent (realizing the option at the upper edge of the actual range) is approaching as expected. No change needed.
- tsm-swing-contest (v7): 18.976 shares @ $421.58, +0.93% to close $434.24, weight about 8.5%. The $406-414 add-on window is dormant under weight<0.063 and is now far from market; take-profit >$460 (+5.9%) and stop <$398 (-8.4%) -- that stop distance is still reasonable for TSM's low volatility (the lesson from v4's <$410 being swept intraday). No change.
- msft-diversifier-swing (v2): 9.128 shares @ $492.98, now +0.2%. Pending #69 (price<490 -> buy_call $1,200) has a stale trigger price (spot is $493.78); approving it would re-price a purchase on a condition that no longer holds, so I recommend dropping that pending item. The strategy itself has no defect; no revision proposed.
- crm-fade-swing: archived, no rules, no action.

**Lessons**
Wrote a new lesson, stop-geometry-decays-after-a-run: stop levels "expire" as the market moves. A stop placed by ATR at entry degrades, after the name runs 10%+, into a line that permits giving back the entire gain. Two cases showed up on the same day (AMD 6.4 ATR, MU 3.9 ATR), and with MU's take-profit only 3.4% away the asymmetry is fully inverted. From now on, daily review treats "how many ATR is the stop from market" as a standing check: beyond about 3 ATR with a meaningful gain already, move it up below the nearest consolidation -- but never raise the take-profit target at the same time. Chasing targets higher is an error pattern this account has recorded repeatedly; tighten the downside only, never chase the upside. Also wrote three same-day factual notes for AMD, MU and the portfolio.

**Proposals**
- #76 amd-swing v8: move the stop and the option stop from <$438 up to <$500 (below the 9/08-9/15 consolidation, about 3.1 ATR), cutting the $2,000 stock leg's worst case from about -13% to about -1%; the >$560 take-profit and the $465-482 entry window are untouched.
- #77 mu-oversold-swing v8: move the stop from <$866 up to <$900 (below the daily lows of the 9/14-9/16 base, about 3.0 ATR, still wider than MU's roughly 3.8% typical daily range), cutting the worst case from about -$147 to about -$14; the >$1,050 take-profit and the $915-960 entry window are unchanged; the FQ4 gap risk is recorded in the thesis for the user to decide.
- Recommend dropping #69 on the Pending page (MSFT buy_call, trigger condition no longer valid).
Both revisions take effect only after the user approves them on the Pending page; I cannot apply them myself.

_(Originally written in Chinese on 2026-09-18 and translated to English on 2026-09-20. Figures unchanged.)_

**Day summary (2026-09-17)**
A broad semis risk-on session with no fills and no new triggers. Every position closed green and, for the first time, all six holdings (including the option) are in profit: AMD $545.09 (+6.36%, vol 27.9M, on AMD raising its 2030 TAM outlook to ~$3T from ~$2T), MU $977.50 (+5.50%), TSM $430.02 (+2.94%), NVDA $219.34 (+2.54%, vol 88M), MSFT $497.75 (+1.52%), NVDA 10/16 $220C +37.74% (mark ~$7.60 vs $4.65 cost, ~+63%). Equity $96,563, cash $68,389 (70.8%) — the cash level is the arithmetic result of ~28% aggregate target weights, not idle indecision.

**Per-strategy check**
- nvda-momentum-swing (v8): stock leg 36.574 sh @ $218.74 = $8,018 ≈ 8.3%, target met; add-on correctly dormant at the new weight<0.063 gate (yesterday's fix already doing its job — NVDA rose away from <214 anyway). One real defect found: the call's only exits are stock levels +13% / -7% away with 29 DTE. Revision proposed.
- tsm-swing-contest (v7): 18.976 sh @ $421.58, +$154, ~8.4%. Add-on window $406-414 now ~4% below price and dormant by design; TP $460 is +7%, stop $398 is -7.4%. Nothing off.
- mu-oversold-swing (v7): 3.872 sh @ $903.97, +$282 (+7.5%). Stop $866 now 11% below price, TP $1,050 +7.4%. FQ4 earnings ~09-30 is the real test; the gap-aware entry window has done its job.
- amd-swing (v7): 3.961 sh @ $504.90, +$159 (+8.0%). The $465-482 deep window is now 12-15% below price — dead for this leg, but I am deliberately not raising it (that is the repeat mistake this book has already logged). TP $560 is only +2.7% away and is now the most likely rule to fire.
- msft-diversifier-swing (v2): 9.128 sh @ $492.98, +$41. The diversifier premise held both ways this week: negative contributor while semis rallied 9/16, mildly positive today. Add-on <$478 and stop <$465 untouched.

**Lessons**
1. An option leg must have its own price exit inside the range the stock actually trades. Shares have no expiry and can wait for $248; a one-month call cannot. Written to memory.
2. A pending option proposal whose trigger price no longer holds is stale, not a bargain — the contract is re-priced at approval. Pending #69 (MSFT call, triggered at $490.30, now $497.75) should be left to lapse rather than approved.
3. Confirmation, not action: a day where five ladders sit dormant because their weight gates were tightened yesterday is the fix working, not a missed opportunity.

**Proposals**
- #73 nvda-momentum-swing v9 (queued, needs your approval): adds `take-profit-options-early: price > 230 -> close_options`, an option-only exit at the top third of the post-earnings $207-234 base. Stock levels, sizing, stop and thesis unchanged; the old >248 close_options is kept as a backstop.
- No revision proposed for AMD/TSM/MU/MSFT — nothing measurably diverged today.

**Day summary (2026-09-16, paper, equity $95,394 / cash $68,389 = 71.7%)**
No trades today. Semis rebounded for a second day after 9/14's sector-wide AI risk-off: AMD +1.65% ($512.50), TSM +1.02% ($417.98), NVDA +0.82% ($213.90), MU -0.11% ($926.55); the NVDA Oct-16 $220 call marked +10.36% (+$210). The one decliner was MSFT, -1.37% to $490.30 — its second straight down day against a rising semis book, which confirms the low-correlation premise it was bought for while currently sitting on the negative side of it. One sentinel event: MSFT's entry-call fired, logged "queued analysis unparseable — left for human review," and is now Pending #69. The structurally important fact today is that the book has finished building: NVDA (8.2%) and TSM (8.3%) have both reached their 8% targets, so the account has moved from a laddering phase into a hold-and-discipline phase.

**Per-strategy check**
- nvda-momentum-swing (v7): thesis intact — $207-234 base, three closes in the lower third on ordinary 88-132M volume, no breach of the $207.25 invalidation low. Measurably off: add-on-dip gates on position_weight < 0.10 but buys $3,500 (~3.7% of equity), so a fill lands at ~11.9% — above the strategy's own 10% max_weight. NVDA traded inside the <214 condition all session, so this was live, not hypothetical. Proposed v8.
- tsm-swing-contest (v6): thesis intact and reinforced — lowest-beta leg (fell least on 9/14, participated on the rebound). Same sizing defect: 8.3% weight with a <0.10 gate and a $3,500 buy inside a $406-414 window that sits only ~1% below today's close. Proposed v7.
- msft-diversifier-swing (v2): working as designed. The <490 dip condition, rewritten on 9/11 precisely because the old one was unreachable, did its job on the first real test. Option leg is $1,200 (~1.26% of equity) with paired $520/$465 close_options rules — compliant. Pending #69 awaits the user; note the contract re-prices on approval, so if MSFT bounces back above ~$495 the strike/premium shift and it should be re-judged, not rubber-stamped. Stock add-on <$478 is -2.5% away — the only near-market equity ladder step left in the book.
- mu-oversold-swing (v7): at target (3.8% vs 4%), basing $917-945 without revisiting the 9/14 low of $902.60. Stop <$866 has ~4% slack below that low — correct geometry. No change needed. Flag: FQ4 earnings ~9/30; a high-volatility satellite should not be held naked through it.
- amd-swing (v7): underweight at 2.1% vs a 4% target, and the only remaining entry window ($465-482) is 6-9% below market — the fourth recurrence of this book's pullback-only stalemate. Deliberately NOT proposing a fix: AMD is a secondary diversifier with a small target, the cost of staying underweight is low, and raising deep thresholds to chase price is the exact mistake this book has already documented.
- crm-fade-swing: archived, no rules, nothing to review.

**Lessons**
- Weight-cap arithmetic (new, book-wide): `position_weight < max_weight` does not enforce max_weight, because the order size itself is not counted. The correct gate is `max_weight − order_size/equity`; at $3,500 per clip that is ~0.063 for a 0.10 ceiling. Written to memory and applied in both proposals.
- A target-weight-reached strategy whose add-on goes dormant is in the correct state, not broken. Dormancy re-arms usefully if a stop ever fires and the position is rebuilt smaller.
- "queued analysis unparseable" is an analysis-layer failure, not a rule failure — TSM's add-on carried the same note on 9/15 and filled normally. Treat it as an ordinary confirm-mode proposal to review.

**Proposals (both queued, neither applied — your approval only)**
- #70 nvda-momentum-swing v8: add-on gate 0.10 → 0.063. All price levels, the option leg, and both stops unchanged.
- #71 tsm-swing-contest v7: add-on gate 0.10 → 0.063. The $406-414 gap-aware window, $398 stop, $460 target unchanged.
- Still open from the sentinel: Pending #69, MSFT buy_call $1,200 (30+DTE, ~4% OTM). I am advisory only and cannot approve or place it; on the merits it is rule-consistent, correctly sized, and properly paired with exits, but it will re-price at approval time.

**Day summary (2026-09-15, paper, equity $95,176, cash $68,389 = 71.9%)**
A rebound day for the semis book after 9/14's sector-wide AI risk-off, with one exception: TSM. AMD +2.19% (close $504.20, reclaiming all of 9/14's -4.40%), NVDA +0.57% ($212.17), MU +0.39% ($927.60), the NVDA Oct-16 $220 call +5.10%, while MSFT gave back -1.64% ($497.12) and TSM fell -0.99% ($413.89) after gapping down to open $421.94 and printing a $412.55 low. One trade: TSM add-on-dip filled 8.4627 sh @ $413.58 ($3,500) at 12:11 ET. Pending queue was empty at start of pass; I have queued one revision (#65). Note the inverted correlation from 9/14 — MSFT was the only gainer that day and the only loser today, which is exactly the diversifier's job.

**Per-strategy check**
- **tsm-swing-contest (v6)** — thesis honored, and today was its first real test. The v6 bounded add-on window (price > 406 and price < 414) was written on 9/14 specifically so a gap-down open could not fill a "price < X" rule one ATR above the stop. TSM gapped exactly that way and the window filled at $413.58, leaving ~$15.6 (~1.8 ATR) of slack above the $398 stop. Position is now 18.976 sh @ $421.58, ~8.25% of equity against an 8% target — the add-on window has done its work and the strategy is fully built. The 8/24 low ($405.15) was approached but not broken, and volume fell to 8.3M from 14.3M, so the invalidation condition (heavy-volume close below $405.15) is not met. No change proposed.
- **nvda-momentum-swing (v7)** — no rules fired; NVDA closed $212.17 inside the stated $207-234 base. add-on-dip (<214) and entry-call (<212) are both technically live at today's price, which is why I checked: they were consumed on 9/14 and position_weight is now ~8.2%, at target, so further adds would push past the 10% max. Stop <204 is ~3.9% below. The one real blemish is already logged: 9/14's one-sided "<214" add-on filled at $209.38 on a gap, only 0.76 ATR above the $204 stop, versus the designed 1.4 ATR. TSM's window proved the fix; NVDA got the diagnosis first but TSM got the cure.
- **amd-swing (v6)** — two measurable defects, revision queued. entry-step/entry-call both require position_weight == 0, unreachable since the 9/10 fill, yet entry-call still displays as armed; and it is independently unexecutable ($900 cannot buy a 30+DTE 5%-OTM contract on a $500 stock — skipped 9/10, the fourth such skip in this book). Also, 9/14's $480.33 intraday low dipped under the <$482 deep entry without filling.
- **msft-diversifier-swing (v2)** — working as designed, no action. -1.64% to $497.12 on the day the semis bounced; 9/14 it was +1.97% while MU fell -5.25%. Position 9.128 sh @ $492.98, roughly flat (+$40). add-on <478 (-3.8%), TP >520, stop <465 all untouched. Its paired $1,200 call leg at <490 remains armed and is the one non-NVDA option leg I still consider plausibly affordable.
- **mu-oversold-swing (v7)** — no rules fired, position 3.8718 sh @ $903.97, +$92. The v7 gap-aware window (price > 915 and price < 960) is already validated in principle by TSM today. Stop <866 is ~6.6% below; TP >1050 is ~13% above. FQ4 earnings ~09-30 is the next real event and deserves a pre-earnings review.
- **crm-fade-swing** — archived, no rules, no action.

**Lessons**
- Bounded entry windows beat one-sided thresholds on high-priced, high-ATR names. Recorded as a durable rule: for stocks above ~$300 with ATR of 2-3%, write entries with both a lower and an upper bound and place the stop at least 1 ATR under the window's floor. TSM's fill at 1.8 ATR of slack versus NVDA's at 0.76 ATR is the controlled comparison, one week apart, same book.
- The sentinel samples discrete prices; it does not see intraday extremes. AMD's $480.33 low on 9/14 did not trigger the <$482 rule, just as MU's quote clearing >$1010 on 9/04 did not fill the take-profit. Put levels inside real trading bodies, not on wick extremes, and never raise a threshold to chase a miss that was never actually fillable.
- The "queued analysis unparseable — left for human review" sentinel log is cosmetic. It appeared at 15:06 on the TSM rule and that same hard rule filled normally five minutes later. This is at least the fifth occurrence; I am treating it as noise in the analysis-text path, not an execution risk, and will stop investigating it unless a hard rule actually fails to fire.
- Option affordability is a structural constraint of this book, not an outage. Four skips now (TSM 9/10, AMD 9/10, MU 9/14) all on $400+ underlyings. Under the 2%-of-equity premium discipline (~$1,900), only NVDA at ~$212 reliably supports a call leg. Strategies on expensive underlyings should be written as stock-only rather than carrying dead option entries.

**Proposals**
- Queued revision #65: **amd-swing v6 → v7** (awaiting your approval on the Pending page; I cannot apply it). Three changes, no change to price levels or the fundamental thesis: (1) mark entry-step `triggered` — it is factually consumed; (2) delete entry-call outright, as it is both permanently unreachable and unaffordable, making AMD an explicitly stock-only leg; (3) convert entry-dip from `< 482` to the bounded window `> 465 and < 482`, so a gap opening at $445-455 cannot buy less than half an ATR above the $438 stop. The paired take-profit-options/stop-loss-options exits are deliberately kept — they no-op without an option position and preserve an exit if one is ever opened manually. $482/$560/$438 and the ~$4,000 total commitment are unchanged. Accepted trade-off: a gap straight through the window means no fill at all.
- No revision proposed for NVDA, MSFT, MU or TSM. NVDA's spacing flaw is real but its add-on is already consumed at target weight, so rewriting the geometry now would be cosmetic; the right moment is the next time that strategy needs to open an entry. I would rather leave it documented in memory than churn a strategy file for no behavioral change.
- For your consideration, no action taken: the book is now ~72% cash with TSM (8.25%) and NVDA (8.2%) both at their swing targets. Every remaining ladder step sits 4-7% below market. That is the design working, not a failure — but MU's FQ4 earnings around 09-30 is the one dated event that could move a satellite position, and it is worth deciding before then whether that position is held through the print or trimmed into it.

**Day summary — 2026-09-14 (paper, equity $95,108, cash $71,889 = 75.6%)**

A sector-wide AI risk-off session, not a fundamental event: public calls by AI-lab leaders (Anthropic/OpenAI) for slower frontier development triggered a broad semiconductor selloff, with memory hit hardest. Book moves: MU -5.25%, AMD -4.40%, TSM -3.41%, NVDA -3.36% (on 123M shares, above the prior two sessions' 83-106M), MSFT +1.97%. Three rules fired, all hard rules on auto-execution paths, none of them my doing:

- MU entry-dip: 3.8718 sh @ $903.97 ($3,500)
- NVDA entry-call: 2× NVDA 2026-10-16 $220C @ $4.65 ($930)
- NVDA add-on-dip: 16.716 sh @ $209.38 ($3,500)
- MU entry-call ($800, dte≥30, otm 6%): skipped — "no affordable option contract"

Net P/L on positions: -$210. Invested weight rose to ~24%.

**Per-strategy check**

- **nvda-momentum-swing (v7)** — Rules behaved as written, but the *geometry* v7 was specifically built to fix got undone by the gap. v7 moved the add-on to <$214 and the stop to <$204 precisely to keep ~$10 (~1.4 ATR) of spacing. The gap-down open filled the add-on at $209.38, leaving $5.38 (~0.76 ATR) to the stop — a 2.6% further decline liquidates the whole 36.57-share position and closes the new calls. The intraday low $208.93 is 0.8% above the stated invalidation level (8/24 low, $207.25). Position is now ~8.1% of equity, at target, approaching the 10% max. Also the one genuine success: NVDA is the only name in this book where the ~$1,000 option budget actually buys a 30+DTE contract.
- **mu-oversold-swing (v6)** — Thesis and rule diverged measurably. The rule was written on 9/8 to enter a ~6.5% pullback around $950-960 with the $880 stop ~2 ATR away; it instead filled at $903.97, 2.7% (~0.7 ATR) above the stop, essentially on top of the strategy's own invalidation level ($887.61). The rule bought into the zone the document defines as thesis failure. Revision proposed.
- **tsm-swing-contest (v5)** — Two rules have been displaying "armed" while being permanently unreachable (`position_weight == 0` after the 9/10 fill), and the option leg is separately unexecutable. Add-on $415 / stop $404 is only 1.26 ATR of spacing, below the standard set in the NVDA v7 review. Revision proposed. Behaviorally the thesis held: lowest drawdown of the four semis names today.
- **amd-swing (v6)** — No action, correct. Closed $493.41; deep tier <$482 is now only -2.3% away, the closest it has been. Stop $438 is -11% away. Nothing off.
- **msft-diversifier-swing (v2)** — No action, correct, and the strategy earned its keep: the only green position on a sector shock day, the second such confirmation after 9/10. Watch item: its `buy_call $1200` leg at <$490 is probably unaffordable on a ~$505 stock too, but there is no evidence yet — I will not pre-emptively cut an untested rule.

**Lessons written**

1. *Gap-through-entry compresses the stop.* An entry condition with only an upper bound (`price < X`) fills far below X on a gap day, collapsing the designed entry-to-stop distance into noise. Documented with both of today's cases. Fix: bound the entry window on both sides for high-ATR names, or define stops as an ATR multiple of the actual fill.
2. *Option-leg affordability is structural.* Four skips now, all on >$400 stocks (TSM, AMD 9/10; MU 9/14). At ~2% of equity per position, only NVDA-priced names are reachable. The right response is deleting dead call legs, not raising budgets past the discipline.
3. Recorded the sector-shock day itself: four semis positions correlated near 1.0; the diversification that worked came from the non-semis leg, not from holding four different chip names.

**Proposals (both queued, neither applied — your approval only)**

- **#59 mu-oversold-swing v7** — entry window bounded (`price > 915 and price < 960`); stop $880 → $866 (~0.6 ATR below the 8/24 low, so a wick doesn't liquidate; ~1.1 ATR from the actual fill, worst case ~-$147); dead `buy_call $800` + `close_options` removed.
- **#60 tsm-swing-contest v6** — dead `position_weight == 0` rules resolved, unaffordable `entry-call` and its two paired `close_options` rules deleted (no TSM option position exists, so removal carries no risk); add-on becomes a bounded window `$406 < price < $414`; stop $404 → $398.

**Not proposed, and why:** NVDA has the same compressed-spacing defect, and it is the most consequential one in the book. I did not queue a fix because `propose_strategy_revision` rejects option actions on anything other than an `auto` strategy, and nvda-momentum-swing is `confirm` with two live $220 calls — any revision I could submit would have to strip `entry-call` *and* both `close_options` rules, leaving the open contracts with no automated exit. That is worse than the current state. If you want NVDA's add-on bounded and its stop re-spaced, the options question has to be decided first (keep the calls and accept manual management, or move the strategy to `auto`). I cannot change that boundary or the authorization level myself.

**Day summary（2026-09-11）**
账户权益 $95,276，现金 $79,819（约 83.8%），四个持仓全部延续 9/10 的自动成交状态，今日无成交、无观察记录、Pending 队列为空。今日是 9/10 板块级获利了结后的反弹修复日：AMD 收 $516.13（+2.49%，区间 $501.35-521.06，量 1,830 万股）、TSM 收 $433.19（+1.21%）、MSFT 收 $495.63（+0.65%）、NVDA 收 $218.29（-0.03%，量 8,329 万股为近月最低）。宏观层面 8 月 CPI 于今日公布，核心 CPI 环比 +0.3%（高于预期 +0.2%）、整体 +0.4%、同比维持 3.4%，利率期货对 9 月中议息会议的加息概率定价约 85%——这与 9/10 偏热的 PPI 是同一条线索：半导体的回撤是利率/估值驱动，不是 AI 需求叙事受损。四仓合计浮盈约 -$43，基本打平；NVDA 是唯一浮亏仓（-$164，-3.8%）。

**Per-strategy check**
- nvda-momentum-swing（v7，8% 目标，实际 4.6%）：收 $218.29，完全符合 v7 thesis 描述的 $207-234 区间震荡。加仓档 <$214（-2.0%）、期权腿 <$212（-2.9%）、止损 <$204（-6.5%）、止盈 >$248（+13.6%），全部未触及，今日"无动作"是正确结果。v7 修复的 $10（~1.4 ATR）档距在今日 $218.15-222.00 的窄幅波动中未受考验。需留意：entry-initial 仍处 triggered，自动修订不会 re-arm，若要重新建仓须在策略页手动操作。
- tsm-swing-contest（v5，8% 目标，实际 4.8%）：收 $433.19，持仓 10.5133 股 @ $428.03，浮盈 +$56（+1.2%）。今日 ATR 实测继续低于其他仓（区间仅 $428.90-435.37，1.5%），"最低波动锚"定位成立。加仓 <$415（-4.2%）、止盈 >$460（+6.2%）、止损 <$404（-6.7%）均未触及。entry-call 的 $900 腿已于 9/10 因"无可负担合约"跳过且条件（position_weight == 0）现已永久不可达——同 MSFT 的缺陷，但暂不重复提案。
- amd-swing（v6，4% 目标，实际 2.1%）：收 $516.13，持仓 3.9612 股 @ $504.90，浮盈 +$44（+2.5%），是今日最强仓。v6 于 9/9 把门槛从 $475-485 抬到 $505/$482 的重锚决策连续两日被验证：9/10 成交、9/11 即反弹 +2.5%。剩余深档 <$482（-6.6%）已随反弹拉远，止盈 >$560（+8.5%）、止损 <$438（-15.1%）。止损距市价过远是单档小仓（$2,000）的必然结果，可接受。
- msft-diversifier-swing（v1，8% 目标，实际 4.7%）：收 $495.63，浮盈 +$21。9/8 建仓以来四个交易日收盘落在 $491.65-495.63 的极窄带内，且 9/10 板块回落日仅 +0.16%——低波动非半导体分散器的核心假设已被实测确认。唯一缺陷是规则工程而非价格：entry-call 已死（见下）。
- mu-oversold-swing（v6，4% 目标，零仓位）：MU 收 $975.26（-0.22%，区间 $967.38-993.99）。entry-dip 门槛 <$960 距市价仅 -1.6%，是全书最接近触发的规则——本周 MU 已两次逼近（9/10 低 $973.50、9/11 低 $967.38）。止损 <$880、止盈 >$1,050 结构合理。零仓位在此是设计结果，不是失败；但需注意 MU 的 FQ4 财报约在 9 月底，若在财报前一两日触发建仓，等于用硬规则承担事件风险——这一点我未能在本轮核实确切日期（搜索未返回可靠结果），列为明确的未解事项。
- crm-fade-swing：已归档，无规则，无动作。

**Lessons**
写入 lesson 层一条"死规则与期权预算审计"：(1) 以 `position_weight == 0` 为条件的期权腿，一旦同条件股票腿成交即永久不可达，但策略页仍显示 armed，制造"还会执行"的错觉——NVDA v7 已修正，MSFT v1 仍存此缺陷，TSM v5 同样中招；(2) $900 的期权预算对 $400-500 价位标的普遍买不到 30+DTE 合约，9/10 的 TSM 与 AMD 两腿均因此跳过，应提高到约 $1,000-1,200 或直接删除规则，不要留着假装会执行。另写入 MSFT 当日结论（低波动分散假设被实测确认 + 死规则记录）。今日最重要的正面确认：AMD v6 的"重锚而非死等深档"是本书里第一个连续两日被价格验证的修订。

**Proposals**
提交 #55：msft-diversifier-swing v1 -> v2，只改两处——entry-call 从不可达的 `position_weight == 0 and price < 495` 改为纯回调条件 `price < 490 and position_weight < 0.10`（$490 位于 9/8-9/11 实测区间 $486-499 内），期权预算 $900 -> $1,200（约权益 1.26%，仍在 2% 纪律内）；配对的 take-profit-options(>520) 与 stop-loss-options(<465) 已存在且不变。价格档位（加仓 <$478、止盈 >$520、止损 <$465）、目标权重、基本面 thesis 全部不动，因为低波动假设已被实测确认，无重锚理由。等待你在 Pending 页决定——我无权自行应用。未提交的两项（TSM v5 同类死规则、MU 财报前建仓的事件风险）留待下一轮，避免同一天堆叠多个提案。

**Day summary (2026-09-10)**

A sector-wide profit-taking session in semis, on the heels of a hot August PPI print and Middle-East tension (external reporting; group-wide, not name-specific). The book went from two positions to four in one day, and every semis leg closed red:

- AMD −3.36% (o 508.70 / h 516.33 / l 502.92 / c 503.60, vol 15.6M — below the 22-28M of 9/8-9/9)
- NVDA −2.37% (c 218.36, vol 98.8M — below the 120-300M post-earnings pace)
- TSM −1.63% (c 428.28, vol 7.0M — the lightest in a month)
- MSFT +0.16% — the only green line

Two fills, both rule-driven, both hard rules on `auto` strategies, both at the sentinel's discretion (I placed nothing and can place nothing):
- TSM 09:41 ET, 10.5133 sh @ $428.03 ($4,500), rule `entry-initial` (<$430)
- AMD 12:11 ET, 3.9612 sh @ $504.90 ($2,000), rule `entry-step` (<$505)

Both accompanying $900 call legs were **skipped: "no affordable option contract."**

Equity $95,140, cash $79,819 (~84%), total unrealized −$179 (NVDA −$166 is nearly all of it). Pending queue was empty at open. Note that AMD filled at $504.90 and closed $503.60 — the entry is essentially at the day's second-lowest print zone, so today's −3.36% day-change on the position is a marked-to-open artifact, not a −3.36% loss (actual: −$5).

**Per-strategy check**

- **tsm-swing-contest (v5)** — Working as written, and this was the cleanest execution of the week. v5 deliberately waited for a ~2% pullback instead of chasing the 9/8 close of $439; it got $428.03 two sessions later and closed the day at $428.28, i.e. flat to cost. Weight 4.7% vs 8% target. Remaining ladder: add-on <$415 (−3.1%), stop <$404 (below the printed 8/24 low $405.15), TP >$460. The thesis claim that TSM is the lowest-vol name in the book held today: it fell least of the three semis. No change proposed.
- **amd-swing (v6)** — Working as written, and v6 vindicated the re-anchoring decision. v5 (written 9/8) required a pullback to $475-485, which was 7.5% below market and would never have filled; v6 moved the trigger to <$505 on 9/9 and it filled the next day. Weight 2.1% vs 4% target; deep tranche <$482 is only −4.3% away, which is reachable for a name with ~3% daily range. Stop $438 sits under the printed 9/3 low of $440.50 — correctly placed. No change proposed.
- **msft-diversifier-swing** — Did exactly the job it was added for: it was the only position that didn't fall on a semis-wide down day. Weight 4.7% vs 8% target; add-on <$478 is −2.8% away. One dead rule: `entry-call` is conditioned on `position_weight == 0`, which the 9/8 stock leg permanently consumed, so it will stay armed forever and never fire. Noted in memory; not worth a revision on its own since MSFT's low IV is one of the two places a ~$900 budget could plausibly buy a contract — flagging it for the user rather than churning the file.
- **nvda-momentum-swing (v6)** — **Measurably off. Revision proposed.** Two defects: (1) `add-on-dip` at <$210 and `stop-loss` at <$208 are $2 apart against a ~14-day ATR of ~$7.10 — any session reaching the add-on has near-even odds of also tripping the stop, so the design buys $3,500 and liquidates it. (2) The thesis explicitly says the stop is "anchored under the 8/24 swing low, not a round number," but $208 is a round number and sits *above* the printed 8/24 low of $207.25. That is the precise failure v3 was written to fix (the old <217 stop was taken out by a single $216.81 wick on 9/1; NVDA closed $217.44 that day and later ran to $234). Today's close of $218.36 puts price ~5% from the add-on, so this is not urgent — but it should be fixed before it matters, not after.
- **mu-oversold-swing (v2)** — Zero position, all rules armed, correct behavior. MU $977.41 vs the <$960 trigger: only −1.8% away, the closest of any armed entry, and MU fell hard today with the group. Worth flagging: when it does trigger, the paired `buy_call $800` will almost certainly be skipped as unaffordable (see lessons) — the stock leg will go on alone.
- **crm-fade-swing** — Archived, no rules, nothing to review.

**Lessons written to memory**

1. `option-leg-budget-mismatch` — The recurring "no affordable option contract" skip is now explained, not just observed. A $900-1,000 budget cannot buy one 30+DTE, 5%-OTM contract on a $400-500 underlying: rough 30-day premium ≈ 0.4·S·IV·√(T/365) puts AMD near $2,600/contract, TSM near $2,000, MU (at $977) near $5,600 — all above the ~2%-of-equity (~$1,900) discipline cap. Practical rule: only NVDA (~$218) and low-IV MSFT are candidates for a ~$1,000 option leg; on $400+ high-IV names, write stock-only rules instead of legs that are guaranteed to skip. This supersedes the older "options tooling is down / server error" note in memory — today's skips are an economic constraint, not a broken tool, which is a different problem with a different fix.
2. `option-leg-mutually-exclusive-condition` — Pairing a stock leg and an option leg on the same `position_weight == 0` condition means whichever executes first kills the other permanently (MSFT 9/8). Option legs should carry independent price conditions.
3. `re-anchor-thresholds-works` — Four of five re-anchored entries (MSFT <$495, NVDA <$228, TSM <$430, AMD <$505) filled within 1-2 days, and every historical threshold set 5%+ below market (TSM v3's <$395/<$375, AMD v5's <$482, MU's <$960) never filled. Starter thresholds belong 1-2% below market; 4-7% is for deep tranches. Recorded with the honest counter-cost: because all four filled inside two days, the whole book was established straight into a sector pullback — fill-rate was bought at the price of entry-timing diversification, so staged/time-based tranching still matters.
4. Position/state notes: `AMD-2026-09-10`, `TSM-2026-09-10`, and `book-state-2026-09-10` (~16% invested, 84% cash is the designed outcome — the five target weights sum to only 24%, so cash is not idle by mistake).

**Proposals**

- Queued **#54, nvda-momentum-swing v7** (needs your approval on the Pending page; I cannot apply it): `add-on-dip` <$210 → **<$214** (into the real 8/24-8/26 base of $208.48/$213.05/$209.66 closes); `stop-loss` <$208 → **<$204** (below the printed 8/24 low with ~0.4 ATR slack, ~3.2 ATR under the $226.61 entry). Spacing becomes ~$10 (~1.4 ATR), matching how TSM (<415/<404) and MSFT (<478/<465) are already built. Also rewrites the unreachable `entry-call` from `position_weight == 0` to a pure dip condition (<$212), keeping its existing paired `close_options` profit-target and stop. Worst case if both legs fill: ~$8,000 notional, stop ≈ −6% off a blended ~$220 cost ≈ −$480, ~0.5% of equity.
- No revision proposed for TSM, AMD, MSFT, or MU. All four behaved as their theses specify, and the option-leg budget issue is better handled as a standing design rule for future drafts than by rewriting four files today.
- Judgment call you may want to overrule: I did *not* propose stripping the now-dead option legs from amd-swing and tsm-swing-contest. Both are marked `triggered` and cannot fire again, so removing them is cosmetic — but it does leave two strategy files describing option exposure the book will never hold.

**Day summary (2026-09-09, paper)**
Quiet session: no trades, no observations, empty pending queue at start. Equity $95,251, cash $86,319 (~90.6% cash, 9.4% invested). Both positions drifted slightly lower — NVDA closed $223.67 (-0.91% on the day, low $223.46, volume 75.2M, the lightest in a month) against a $226.61 cost, and MSFT closed $491.65 (-0.47%, volume 11.9M) against $492.98. Combined open P&L about -$68, i.e. noise. Meanwhile the three zero-position names all moved *away* from their entry gates rather than toward them: AMD +3.0% to $521.09 (high $526.79), MU +2.75% to $1,027.77 (high $1,042.40), TSM -0.78% to $435.56. The semis complex was strong and our two holdings were the two laggards — the book was positioned defensively into an up day for the sector it is supposedly long.

**Per-strategy check**
- **nvda-momentum-swing (v6, filled 9/8 @ $226.61)** — behaving as written. Thesis says NVDA is basing in a $208-234 range post-earnings; today's $223.46-226.18 range on the month's lightest volume is exactly that digestion, not breakdown. Entry done at 4.7% weight vs 8% target; add-on <$210 sits -6.3% away, TP >$248 +10.9%, stop <$208 -7.0%. Nothing to fix.
- **msft-diversifier-swing (filled 9/8 @ $492.98)** — doing its job. Three sessions of $489.80-495.19 while AMD moved 14% confirms the low-volatility diversifier premise (day moves -1.15%, -0.47%). Add-on <$478 (-2.8%), TP >$520 (+5.8%), stop <$465 (-5.4%). All levels sane relative to market. No change.
- **tsm-swing-contest (v5, zero position)** — closest to action: entry <$430 is only -1.3% below the $435.56 close, well inside TSM's ~2% daily range, so a fill is plausible any session. The v5 stop widening to $404 (under the printed 8/24 low $405.15) still looks correct given the 9/3 stop at $409.17 that reversed to close $417.02 the same day. No change.
- **mu-oversold-swing (v2, zero position)** — the awkward one. Entry <$960 is -6.6% below market while TP >$1,050 is only +2.2% above, and MU already printed $1,042.40 intraday today. The gap is not yet inverted so I am not proposing a level change, but this is a live example of the position-guard hole below: if price tags $1,050 with zero shares, a "sell all" could consume the profit target before we ever own the stock. Worth watching daily.
- **amd-swing (v5, zero position)** — measurably broken, revision filed. v5 was drafted on the 9/8 spike day and anchored entry to the pre-spike $475-485 zone; AMD then closed $505.74 and $521.09. Entry <$482 now sits 7.5% *below* market while TP >$525 sits 0.8% *above* — the strategy cannot buy but is one session from firing sell-all on zero shares. Separately, the stop at $445 sits above the printed 9/3 low of $440.50, i.e. inside normal noise, repeating the exact mistake the thesis claims to have fixed.
- **crm-fade-swing** — archived, no rules, nothing to review.

**Lessons written**
1. `level-sanity-check-when-drafting`: before submitting any strategy, verify stop < entry < take-profit *and* that market price sits between entry and TP. Anchoring entries to pre-spike levels on a violent re-rating day assumes the news move fully round-trips. This is the fourth pullback-only stall (TSM Aug, NVDA earnings gap, MU >$1,000, AMD now) but the first that was self-inflicted at drafting time rather than caused by the market running away.
2. `exit-rules-need-position-guard`: every take-profit/stop-loss in this book is price-only, with no `position_weight > 0` guard. During a zero-position stretch a price spike through the TP line risks consuming the exit rule before entry ever fills. MU is in that danger zone right now. Add guards on the next revision of each strategy.
3. Updated the account-state profile entry, including the structural point that the five target weights sum to only 32%, so ~68% cash is the blueprint's design, not an execution failure — and that option legs still cannot place because the options server has been erroring since contest week, so the book must be judged as stock-only.

**Proposals**
- Queued revision **#53: amd-swing v5 → v6** (awaiting your approval on the Pending page; I cannot apply it myself). Re-anchors entry into two tranches keeping total capital unchanged at $4,000: $2,000 starter at <$505 (a ~3% give-back into the 9/8 body, matching your habit of a small first tranche) plus $2,000 at the original <$482 deep level. Take-profit raised $525 → $560 so it is no longer below market; stop moved $445 → $438, just under the printed 9/3 swing low and ~2.5x the ~$17 ATR below the starter. Option leg trimmed to a single $900 dte>=30 otm=5% call with close_options on both exits. Fundamental thesis and invalidation unchanged. No revision proposed for the other four — MSFT and NVDA are tracking their theses, TSM v5's levels are already correct, and MU's mismatch is not yet severe enough to justify chasing levels.

**Day summary (2026-09-08)**
The post-contest book went live. Two entry rules fired and filled: NVDA 19.858 sh @ $226.61 and MSFT 9.128 sh @ $492.98, ~$4,500 each (~4.7% weight apiece against 8% targets). Equity $95,315.37, cash $86,319.45 — still ~90.5% cash, 9.4% invested. Combined open P&L is essentially zero (MSFT +$10, NVDA -$14). Both names faded intraday: NVDA opened $233.21 and closed $225.73 (-2.01%, low $224.85, 112.6M shares); MSFT closed $493.95 (-1.15%) on light 15.7M volume. The MU take-profit queued on the 09-07 holiday filled at the open at $1,028.62 — 1.8% *better* than its $1,010 trigger, which retires my earlier suspicion that the sentinel had failed to execute it on 09-04.

The one real defect: both paired call legs (MSFT $900, NVDA $1,000, dte>=30) errored with "options MCP server failed to start … no such file or directory: 'uvx'". The stock legs filled, the option legs did not. Actual exposure is stock-only.

**Per-strategy check**
- **nvda-momentum-swing (v6)** — thesis honored. It called for entry in the low/mid part of the $208-234 post-earnings base; the fill at $226.61 is mid-range, arguably the upper half of it. Stop $208 sits just under the 8/24 low ($207.25), TP $248 requires a genuine range breakout. Today's -2% fade left the 8/24 low untouched, so no invalidation. add-on-dip <$210 armed but -7% away. Position is half of target weight.
- **msft-diversifier-swing** — did exactly its job: gave the book a non-semis leg on the day semis wobbled (NVDA -2.0% vs MSFT -1.15%, and MSFT's whole day range was 1.0% vs NVDA's 3.8%), which is the low-volatility premise the strategy was written on. add-on <$478 is -3.2% away, the nearest rung anywhere in the book. TP $520, stop $465.
- **tsm-swing-contest (v5)** — zero position; entry <$430 vs $439.01 (-2.0% away). Correct behavior for a pullback-only anchor, and the widened $404 stop (under the printed $405.15 low) directly fixes the 9/3 failure where the old $410 stop was clipped at $409.17 on a day TSM closed $417.02.
- **amd-swing (v5)** — zero position; entry <$482 vs $505.74 (-4.7% away). The strategy explicitly refused to chase the +6.5% Anthropic/Saudi-AI spike, and AMD held $505.74 today rather than retracing. Missing the continuation is the accepted cost of not chasing; the take-profit logged "no position" twice, which is cosmetic noise, not an error.
- **mu-oversold-swing (v2)** — zero position after the profitable exit; entry <$960 vs $1,000.26 (-4.0% away). Clean round-trip closed: the satellite's only job this cycle was done at $1,028.62.

**Lessons** (written to memory)
1. Option legs are currently inoperable environment-side; real exposure is stock-only, and a paired entry-call can never retry once its stock leg fills because the condition requires `position_weight == 0`. The failure is benign — no bad fills, no phantom risk — but I should not assume an options budget is funded.
2. Rules triggering while the market is closed fill at the next open, not at the trigger price. MU proved this favorably (+1.8%); it will hurt symmetrically on a gap-down.
3. A 1-2% same-day fade on a swing entry sized ~4.7% of equity is noise, not evidence the entry logic is wrong. Stops are deliberately 7-8% away because contest week proved tight stops get clipped by wicks. These entries get judged on their 1-2 week horizon.
4. Structural watch item: three of five strategies are pullback-gated 2-4.7% below market, and the two that filled sit at half their target weight. This is the same configuration that froze TSM at 5.4% for weeks. Not yet a divergence — but if AMD/MU/TSM stay unfilled for ~5+ sessions while the semis uptrend continues, the honest fix is re-anchored near-market rungs, not more waiting.

**Proposals**
None. I deliberately filed no strategy revision today. The stops and option-DTE changes made this morning have had exactly one session of live data, every rule that fired did what its thesis said, and the only malfunction (options tooling) is an environment failure that no YAML edit can repair — the unreachable entry-call rules cost nothing when they error. Proposing changes now would be reacting to a single red close, which is precisely the reflex that produced last week's chase-then-stopped-out pattern. I've recorded the weight-shortfall trigger in memory instead and will propose re-anchoring only if the market refuses to come to us.

Day summary
2026-09-07 was a US market holiday (Labor Day), so there was no trading session and no new price data — every quote I pulled (MU $1,016.59, NVDA $230.36, AMD $477.57) is still the 09-04 close. The only event was the sentinel firing mu-oversold-swing's take-profit rule (price > 1010 -> sell all) at 13:51 UTC against that stale close, submitting a sell of 2.717439509 MU. As a DAY market order on a non-trading day it is queued and can only fill at the 09-08 open; status is still "submitted, fill pending," so the +$262.53 (+10.5%) shown on the position is a mark, not a realized result. The paired take-profit-options rule correctly skipped (no option positions open). Account: equity $95,286.74, cash $92,524.22, MU $2,762.52 the sole position — roughly 97% cash. Pending queue empty.

Per-strategy check
- mu-oversold-swing (v5, auto): Thesis was an oversold bounce from the $928-950 band toward the 8/17 high; that played out — MU broke out 09-04 (+6.10%, high $1,017.77, volume 35.1M, heaviest in weeks) on the sold-out AI/HBM narrative, and the $1,010 target was reached. But the exit mechanics are visibly off: the live quote already cleared $1,010 during the 09-04 session without the rule acting, and the rule instead fired on a closed-market holiday off the stale close, guaranteeing execution at Tuesday's open rather than at the target. Direction correct, timing surrendered to the next gap. Contest window (08-28 to 09-04) has also expired, so the thesis itself is now out of date.
- nvda-momentum-swing (v5, auto): No position, no trades. Entry bands ($226-232 for $3,000; $220-226 for $2,000) contain the last price of $230.36, but entry-momentum and entry-call are already consumed/triggered from the contest week, so nothing can act. The strategy is effectively retired-in-place while its stated thesis window has closed.
- amd-swing (auto): No position. Last price $477.57 sits inside the original $470-485 entry band, but entry rules are consumed; only exit rules remain armed with nothing to exit. Same dead-strategy state.
- tsm-swing-contest (v4, auto): Flat since the 09-03 stop at $409.17 (which was followed by a $417.02 close the same day — the noise-width stop problem already recorded). No re-entry rule exists, so it cannot participate in the semis rally.
- crm-fade-swing (auto): No option position; the put was salvaged 09-02 at $1.07. Remaining rules are all close_options with nothing to close. The short-software leg no longer exists, so the semis/software pair is unpaired.

Lessons
1. Auto hard rules evaluate against the last available quote, not a live one. A price threshold can therefore fire up to about a session and a half late — including on a closed market — and the resulting DAY market order executes at the next open. On a name whose daily range is 3-6% (MU), "sell all above $1,010" in practice means "sell at whatever Tuesday's open gives." Thresholds should be levels you would still accept one session later, and an exit is never done until filled_at/filled_avg_price exist. (I tried to persist this to the lesson layer four times; each write was rejected for exceeding the 500-character limit, so it is recorded here in the transcript and I will retry with a shorter entry next pass.)
2. Structural, and larger than any single rule: all five active strategies are one-shot contest specs whose window ended 09-04. Four of five have consumed entries and no re-entry path, so the book can only decay toward cash — which is exactly what happened, and today's MU sale completes it. Once that fill lands the account is ~100% cash with zero monitored exposure, while the user's standing blueprint (NVDA 20% / AAPL 15% / TSLA 12% / TSM 10% / MU 8%) is entirely unfunded. The gap is now authorization design, not price levels.
3. MU remains the week's only winner and confirms the user's framing: high-beta satellite where patience, not tight stops, paid. TSM/NVDA/AMD all stopped out on intraday noise and then rallied; MU was left alone and gained 10.5%.

Proposals
None filed today. I deliberately did not queue a revision: with the market closed there is no new evidence, the MU order is still unfilled so I cannot measure the actual slippage the fix should target, and mechanically patching a strategy whose contest thesis has expired would be cosmetic. The right next step is a decision from the user, not an edit from me — either archive these five contest specs and draft post-contest strategies aligned to the long-term blueprint, or explicitly re-authorize them with re-entry paths. I can draft either once you tell me which, and I'd rather do it after seeing the 09-08 MU fill price.

**Day summary — 2026-09-04 (final contest session)**
Equity $95,280.79, cash $92,524.22 (~97% uninvested). No trades and no observations today. The only holding is MU 2.7174 sh @ $919.98, now $1,016.59 (+6.10% on the day, +10.5% on cost, +$256). The semis complex ripped across the board: MU $1,016.59 (+6.10%, high $1,017.77, volume 34.7M — heaviest in weeks, on the sold-out AI/HBM memory narrative), AMD $477.57 (from a $462.00 open, +4.7%), TSM $428.52 (+2.8%), NVDA $230.36, CRM $259.29. The week's shape is now unambiguous: the direction call was right and the mechanics gave the money back. Options overlays bought 08-28 (5-DTE, 2-4% OTM) were near-total write-offs of roughly -$4.8k, which is effectively the entire -$4.7k drawdown, while the four stock legs together finished about flat; and noise-width stops sold every name within a session or two of the price recovering above the exit.

**Per-strategy check**
- **mu-oversold-swing (v5)** — Thesis validated, and the single best decision of the week: the stop stayed at $875 (a real invalidation, not noise), the position was never shaken out, and the longer-dated/smaller option shape kept overlay damage contained. One thing is measurably off and needs verification, not a rule change: `take-profit` (price > 1010 → sell all) and `take-profit-options` (price > 1010 → close_options) both read `armed` while MU quoted $1,016.59 and printed a $1,017.77 high — the level cleared and no fill appeared in the journal. Either the sentinel had not evaluated after the breakout, or the order path failed as it did on 09-01. This is the same execution-reliability defect the v5 thesis already recorded, so it is a monitoring item for the next pass; if MU opens back under $1,010 the +10.5% target is simply forfeited.
- **nvda-momentum-swing (v5)** — Zero position. The v5 diagnosis was right on both counts: the $217 stop was noise-width, and the widened $211 stop plus the new $220-226 re-entry are the correct shape. But NVDA closed $230.36, inside `entry-momentum`'s own $226-232 band, and nothing acted — the entry is marked `triggered` and one-shot entries never re-arm, while the new re-entry band sits below the market. The strategy is structurally sidelined.
- **amd-swing (v4)** — Zero position, stopped out near $455 on 09-03. AMD spent the entire 09-04 session inside `entry-dip`'s $470-485 band and closed $477.57, ~+4.9% above the stop price, with nothing firing. Same one-shot-entry lockout; unlike NVDA it has no re-entry rule at all.
- **tsm-swing-contest (v4)** — Zero position, stopped out at $409.17 on 09-03 in a session that closed $417.02. TSM then traded 09-04 from $421.88 to $428.52, entirely inside `entry-dip`'s $424-435 band, with nothing firing. The "relatively more stable member of the complex" premise failed all week and the $410 stop was inside ordinary daily range.
- **crm-fade-swing (v4)** — No position; the put was salvaged 09-02 at $1.07. CRM closed $259.29, i.e. the fade never came and the software short leg of the pair was simply wrong this week — the pair's long side (semis) worked and the hedge cost money on both legs of its own logic.

**Lessons**
Written to memory: (1) `one-shot-entry-decays-book-to-cash` — an entry gated on `position_weight == 0` fires once, stays `triggered`, and never re-arms, so after a stop-out the strategy is retired and the book ratchets one-way into cash; 09-04 is the clean proof, with AMD and TSM sitting inside their own entry bands all session at zero weight. Any auto strategy carrying a stop must carry an explicit re-entry, or the stop is an exit from the market rather than a risk control. (2) MU stock note updated with the $1,016.59 breakout, the earnings date (~09-30), the stale "<$1,000 abandon swing" line, and the un-executed take-profit to verify. The week-level post-mortem entry was repeatedly rejected by the 500-character memory limit; its substance is preserved here and in the two entries that did land — I will re-write it compactly on the next pass.

**Proposals**
None queued. I deliberately did not propose revisions today: the contest window closes with this session, so re-arming entries or re-anchoring stops on amd-swing / tsm-swing-contest / nvda-momentum-swing would commit fresh capital under a mandate that has expired, and the honest fix is a new post-contest mandate rather than another patch to four one-week strategies. The one live decision — whether MU's un-filled $1,010 take-profit should be honoured, re-anchored higher into the breakout, or converted to a trailing exit — is yours; MU is the only position and the only strategy still doing what it was written to do. Note the boundary: I have no order tool in this pass and no way to apply anything myself.

**Day summary (2026-09-03)**
One trade fired today: tsm-swing-contest's stop-loss sold all 4.6819 TSM at $409.17 (rule: price < $410). The paired stop-loss-options rule was skipped — no option positions remained to close, since the $435 calls had already been salvaged on 9/01. The irony is measurable: TSM's intraday low was $407.82 and it *closed the same session at $417.02*, so the stop sold roughly 1.9% below the day's close on a day the whole complex rebounded (NVDA $228.45, MU $958.16, AMD $456.16, CRM $264.42). Account: equity $95,120.79 (-4.9% vs the $100k start), cash $92,524.28 — about 97% uninvested. The only surviving position is MU 2.717 sh @ $919.98, +$96.52 (+0.22% today). Pending queue was empty at session start.

**Per-strategy check**
- *tsm-swing-contest (v4)*: Rule executed exactly as written; the rule itself was wrong for the volatility. Stop sat ~1.2% under a consolidation range on a name with 2%+ daily ranges, so an intraday probe became a realised loss. No re-entry is reachable — entry band $424-435 is ~1.7-4.3% above the $417 close, so this strategy is now retired-by-mechanics with zero position.
- *nvda-momentum-swing (v4)*: Same failure, one day earlier. Stopped out near $217 on a 9/01 intraday probe; NVDA closed that very session at $217.44 (above the stop) and is now $228.45, ~5% above the exit. Worse, NVDA traded all of today *inside* its own entry band ($226-232) with position_weight zero and nothing fired — the one-shot entry was already consumed.
- *amd-swing (v?)*: Stopped out at <$455; AMD closed $456.16 today after a $440.50 washout low. Third instance of the identical pattern. Entry band $470-485 is now ~3-6% above market: unreachable.
- *mu-oversold-swing (v5)*: The one strategy behaving as designed. Equity leg held, +0.22% today, MU $958 sits comfortably between the $875 stop and the $1,010 take-profit. Its v5 fix (dte>=21, $800 budget) is the shape the others should have adopted.
- *crm-fade-swing (v4)*: salvage-decay correctly closed the puts on 9/02 at $1.07 after the fade never began. Vindicated today — CRM rose to $264.42, only 1.4% below the $268 stop. Booking residual premium beat surrendering it. Entry-put band ($245-250) is far below market, so the leg stays flat, which is correct.

**Lessons**
Two written to curated memory. First: tick-based stops placed 1-3% under a range on high-volatility semis systematically harvest the day's low — four stop-outs this week, not one of which was followed by further downside. Short-hold stops need close-confirmation or ≥1.5x ATR width. Second, and larger: a strategy with a one-shot entry and a stop is a lottery ticket in a chop market. After each stop-out no rule could re-enter, so the book decayed one-way into ~97% cash and did not participate in today's broad rebound. The dominant cost this week was non-deployment, not loss-taking — the four expired 5-DTE calls (~-$4.8k) explain essentially the entire drawdown while the four stock legs netted roughly flat.

**Proposals**
Queued one revision: nvda-momentum-swing v5 (#32). It re-anchors the stop from $217 to $211 (below the 9/01 whipsaw low of $215.10, still above the 8/24 washout low of $207.25), adds a single smaller lower-band re-entry rule ($220-226, $2,000) so a stop-out no longer retires the strategy, and reshapes the option leg to dte>=21 / $1,000 / otm=4% with the decay exit loosened to $218 to match the longer tenor. It requires your approval on the Pending page — I cannot apply it. I deliberately did *not* queue three near-identical proposals for amd-swing and tsm-swing-contest: the contest week ends tomorrow, so approving this may buy only one session, and the same two fixes apply verbatim to those strategies if you want them carried into next week. Say the word and I'll draft them.

**Day summary (2026-09-02, contest day 4)**
Equity $95,146 (-4.85% vs the $100k start), cash $90,609 (~95% uninvested). Only one action fired today: the CRM put leg (5x $252.50P, 9/4 expiry) was closed at $1.07 by the v4 `salvage-decay` rule at 13:48 ET, recovering ~$535 of the $1,395 premium instead of letting it decay to zero. Underlyings were broadly higher: NVDA $224.41 (+3.2% from 9/1's $217.44 close, intraday high $227.95), MU $956.08 (+2.43%), TSM $415.57 (+0.38%), AMD $457.06 (-0.55%), CRM $257.19 (-0.36%). Remaining book is two stock legs only — MU 2.717 sh @ $919.98 (+$94) and TSM 4.682 sh @ $427.18 (-$56). No option positions remain open anywhere; the pending queue is empty.

**Per-strategy check**
- *crm-fade-swing (v4)* — The only strategy that acted, and it worked exactly as the 9/1 revision intended. The fade never began (higher closes 8/27 through 9/1), the puts sat in the old dead band where no rule could fire, and the new $254 salvage floor exited them promptly. Realized ~-$860 on the leg instead of the ~-$1,395 write-off that v3 guaranteed. `entry-put` (245–250) is correctly out of reach at $257, so no re-chase. Thesis invalidation ("fade fails to begin") has now fully played out; nothing further to do this week.
- *nvda-momentum-swing (v4)* — Stock leg was stopped out at $217.44 on 9/1; NVDA then closed $224.41 today, +3.2% above the stop. The $226 `salvage-options` floor and the raised $226 `entry-call` floor were never tested because the calls had already been swept at expiry. Rules are internally consistent now, but the episode is the clearest evidence that the $217 stop was inside the daily noise band.
- *amd-swing* — Stopped out at $454.19 on 9/1; AMD closed $459.61 that same session and $457.06 today. Same defect as NVDA: a stop ~3.5% away on a name with a ~3% average daily range. Flat, no rules live.
- *mu-oversold-swing* — The only leg still working. Stock +2.43% today, +3.9% on the position, and it is the one strategy whose stop ($875) sits ~1.5x the daily range away and therefore never fired. Its call was also the only option leg to lose less than half (dte>=21 rather than 2–7 DTE). `take-profit` at $1,010 is ~5.6% above spot and reachable; `salvage-options` is moot with no contract open.
- *tsm-swing-contest (v4)* — Stock leg -2.7% at $415.57, sitting ~1.3% above its $410 stop, i.e. inside one day's range — the same fragility that already cost NVDA and AMD. The v4 $420 salvage floor did its job on 9/1 ($0.48 recovered). Weakest of the four longs all week, which contradicts the thesis line calling TSM "the relatively more stable member of the complex."

**Lessons**
Written to curated memory: the contest week's entire drawdown is attributable to the option overlay, not to stock selection or direction. Five short-dated legs consumed ~$6.1k of premium and returned ~$1.4k (NVDA -97%, AMD -97%, TSM -84%, CRM -62%, MU -40%), while the five stock legs together lost roughly $150. The common flaw was buying 2–7 DTE out-of-the-money options immediately after a large move, when implied volatility was highest — theta and IV crush overwhelmed the thesis in every case, including the ones where the directional call was right. The single leg with dte>=21 (MU) lost least by a wide margin.
A second lesson I could not fit into memory this pass and will record next session: stops placed inside a single daily range manufacture losses. NVDA printed lows under $217 on three consecutive sessions before the stop finally executed at $217.44, then rallied 3.2% the next day; AMD sold at $454.19 and closed that very session at $459.61. MU, whose stop sat ~2x its daily range away, is the only stock leg still open and profitable. Future auto stops should require distance >= ~2x recent average daily range, or close-based confirmation.

**Proposals**
None queued today. The three defects I can measure — short-dated option sizing, stop distances inside the noise band, and the "TSM is the stable one" assumption — all concern strategies that are now flat or expire with the contest on 9/4, so a revision would be paperwork rather than risk control. Every dead-band option defect has already been patched (v4 across all five). The right place for these findings is the post-contest strategy design, and I will bring concrete parameter proposals (dte>=21, ~1% premium sizing, 2x-ADR stops) when the next cycle's documents are drafted rather than amending expiring ones now.

**Day summary — 2026-09-01 (contest week, day 3)**
Equity $95,498 (-4.5% from the $100k start), cash $90,074, so ~94% of the account is uninvested after today's forced exits. The broad AI-semis complex sold off together: NVDA -1.5% to $217.44, AMD -2.4% to $459.61, MU -2.6% to $933.44, TSM -0.3% to $414.01 — press coverage attributes it to a global bond selloff lifting yields, i.e. a valuation/rate move, not a change in AI demand. Two stop-losses executed as designed (NVDA at 13:40, AMD at 14:40), leaving MU stock, TSM stock and the CRM put as the only open positions.

The day's defining event was not directional, it was operational. Every `close_options` attempt failed from 13:40 to ~16:10 because the options order tool could not start (missing `uvx` binary). The four salvage-options rules that were added on 8/31 specifically to prevent premium write-offs all fired correctly and on time (MU<940, AMD<468, NVDA<226, TSM<420) and none of them could execute. Hourly retries at 14:10, 14:40, 15:10, 15:40 and 16:10 all errored. The contracts were finally liquidated at 16:18–16:19, three of them by the DTE<=1 expiry sweep rather than by strategy rule: NVDA 13x at $0.03, AMD 2x at $0.15, MU 1x at $6.30, TSM 4x at $0.48. Right rules, wrong outcome — the premium was already gone by the time the orders went through.

**Per-strategy check**
- **nvda-momentum-swing** — thesis (post-earnings breakout continuation) invalidated. NVDA gave back the entire 8/27 +8.7% gap over three sessions ($227.98 → $217.55 → $220.78 → $217.44) and broke the $217 stop; stock leg sold as designed. The $230 call ended at $0.03, a -97% write-off. Rules behaved correctly on the equity side; the option side was destroyed by tenor, then by the execution outage.
- **amd-swing** — thesis (stabilization/bounce) invalidated. AMD fell through $468 (salvage) and $455 (stop) in the same session; stock sold at $454.20, the $485 call closed at $0.15 (-98%). Both stops fired in the right order; the option exit again could not execute when it mattered.
- **mu-oversold-swing** — the only long still working. Stock leg is +1.3% at $933.44, comfortably above the $875 stop; the oversold-bounce thesis stands. But the call that was profitable on 8/31 (+$105) lost 40% in one -2.6% session and was closed by the expiry sweep at $6.30, not by the salvage rule. Note the live hazard: MU at $933 sits just under the $942–950 entry-call band, so the same short-dated leg can be re-bought within days.
- **tsm-swing-contest** — the only strategy whose option was closed by its own rule today (salvage-options at $0.48, -84% from $3.05), which is v4 working as intended, just late and cheap. Stock leg holds 4.68 shares at $427.18, now $414.01 (-3.1%), sitting only ~1% above the $410 stop. TSM has closed lower on four of the last five sessions; the "relatively more stable member of the complex" assumption remains unconfirmed.
- **crm-fade-swing** — the clearest measurable divergence. The fade has simply not begun: CRM closed higher every single session after the spike (252.05 → 256.00 → 257.54 → 258.16) and never once traded back toward entry. That is the strategy's own written invalidation, and no rule expresses it. The puts are at $1.90 vs $2.79 entry (-$445) with three sessions left, and CRM at $258 sits in a dead band where none of the exits (<245, <228, >268) can fire. Today the put lost -5.98% while the underlying rose +0.24% — theta is now the dominant term.

**Lessons** (both written to curated memory)
- `contest-option-overlay-verdict`: all four 5-DTE, 2–4% OTM calls were near-total write-offs; combined option damage of roughly -$4.8k accounts for essentially the whole -$4.5k account drawdown, while the four stock legs together finished about flat. Requiring an options leg on *every* strategy converted a flat equity book into a -4.5% book. If required again: one leg for the whole book, `dte>=21` not `>=5`, ~1% of equity.
- `option-exit-execution-fragility`: correct option exit rules are not the same as reliable option exits. Rules that must act inside a single session cannot be depended on in this environment; prefer equity expressions and longer tenors that survive an outage.

**Proposals** (queued, awaiting your approval — I cannot apply either)
- **#30 crm-fade-swing v4** — adds `salvage-decay` (close puts while CRM > 254) to recover roughly $950 of the $1,395 premium instead of letting it decay to zero, and tightens the entry cap from $262 to $250 so a new put is only opened once the fade has actually started. A $4 buffer separates the bands to prevent churn. No loss tolerance widened.
- **#31 mu-oversold-swing v5** — leaves every price level and the equity thesis untouched, and reshapes only the option leg: `dte>=21` instead of `>=5`, budget $800 (~0.8% of equity) instead of $1,500, otm 5% to hold a comparable delta. Rationale notes the same change should be applied to tsm-swing-contest, amd-swing and nvda-momentum-swing if you accept the reasoning; I stopped at MU rather than filing four near-identical proposals for you to read.

I did not propose anything for nvda-momentum-swing or amd-swing today: both are now flat with their stops correctly executed, and re-tuning a strategy the same day it stopped out would be fitting to one bad session rather than to a measured defect.

**Day summary**
2026-08-31, contest week day 2. No trades fired and no observations were logged — every armed rule sat inside its no-action band all session. Equity closed at $96,975.61 against the $100,000 contest-week start (-3.0%), with cash unchanged at $84,395.01 (~87% of equity still uninvested).

The important number is the split. Summed across the four equity legs, open P&L is **-$29** (AMD -$3, MU +$105, NVDA -$75, TSM -$56). Summed across the five option legs, open P&L is **-$2,994** (NVDA 230C -$1,235, TSM 435C -$956, AMD 485C -$678, CRM 252.5P -$230, MU 965C +$105). Essentially 100% of the drawdown is the option overlay, not the stock selection. Three of four semis longs rose today (MU +2.77%, NVDA +1.36%, AMD +1.10%; TSM -0.60%) and the option book still bled, because every contract is 5-DTE and out of the money.

**Per-strategy check**
- **nvda-momentum-swing (v3)** — Equity leg fine: NVDA closed $220.50 (+1.36%) after a $216.21 low, i.e. it probed below its own $217 stop intraday for the second session running (8/28 low was $216.81) and recovered both times; stock leg -2.6%. The v3 fix (adding a $224 entry floor) correctly kept it from buying the crash. The 13 $230 calls expiring 9/2 are at $0.13 vs $1.08 (-88%) and need ~+4.3% in two sessions just to reach the strike. Thesis intact, overlay broken.
- **tsm-swing-contest (v3)** — The weak link. TSM ~$415, the only contest long to fall today, sitting ~1.2% above its $410 stop; stock leg -2.9%. The thesis's premise that TSM is "the relatively more stable member of the complex" is not being borne out this week — it is the worst performer of the four. The $435 calls expiring 9/4 are -78% and need ~+4.8% to reach the strike.
- **amd-swing (v3)** — Healthiest equity leg: AMD $470.72 (+1.10%), stock leg -0.1%, comfortably above the $455 stop. The $485 calls expiring 9/2 are -66% and need ~+3% in two sessions.
- **mu-oversold-swing (v3)** — Best strategy in the book and the oversold-bounce thesis is confirmed: MU closed $958.73 (+2.77%), stock leg +4.0%, and the $965 call is the only profitable option in the account (+$105, now 0.7% OTM). But that gain is unprotected — see below.
- **crm-fade-swing (v3)** — CRM $257.53, still inside the entry band and still refusing to fade; puts -16.5%. The v3 fixes (the $262 entry ceiling that killed the re-entry loop, and the exit-window at <$245) were correct and remain the right structure. The fade thesis is unconfirmed, not yet invalidated ($268 is 4.1% away). This one has the same expiry problem as the others — between $245 and $268 no rule can fire before the 9/4 expiry — but I ran out of tool calls before proposing a fix; it is the first item for tomorrow.

**Lessons**
1. **The overlay's exits were keyed to the equity thesis, not to the contract.** Every option leg was given close_options rules at the same prices as the stock take-profit and stop-loss. Those levels are 6-12% away from spot, which is a reasonable swing band for shares and an unreachable one for a 5-DTE option. The result is a dead zone across the entire realistic price range where no rule can ever fire, so the designed outcome for each contract is a silent 100% premium write-off — no trigger, no journal entry, no decision. That is not a bad forecast; it is a structural hole in the rule set, and it is the whole of this week's loss.
2. **It cuts against winners too, which is what makes it structural rather than just unlucky.** MU's call is profitable right now and its default path is still decay to zero, because $1,010 will not print in two sessions. A rule set that can only bank a gain by luck is mis-specified even when the thesis is right.
3. **The stops appear to be evaluated on the sentinel's pass, not intraday.** NVDA traded below $217 on two consecutive sessions without stop-loss firing, and recovered both times. This week that helped. I should not assume a price-level stop protects against an intraday spike, and should stop describing these as if they do.
4. **Stock selection is not the problem; position sizing of premium is.** -$29 on ~$9,470 of equity exposure versus -$2,994 on ~$5,745 of premium. Sizing premium at ~$1,500 per name across four names put ~6% of equity into instruments that can go to zero in five days — inside the letter of the 2%-per-position guidance per trade, but concentrated on one factor (near-dated semis vol) in aggregate.

*Note: I exhausted my tool budget on the four revisions and did not write these to curated memory. Lessons 1-4, especially the premium-concentration point, should be committed on the next pass.*

**Proposals**
Four revisions queued for approval on the Pending page; none is applied automatically, and none widens a stop or loosens a loss tolerance:
- **#26 nvda-momentum-swing v4** — add `salvage-options` (close calls below $226) to recover residual premium instead of letting it expire worthless; raise entry-call floor $224→$226 so the salvage and entry bands cannot overlap into a churn loop.
- **#27 tsm-swing-contest v4** — add `salvage-options` below $420, with a $4 buffer to the $424 entry floor; thesis text updated to record that the "more stable member" premise is failing. Equity stop left at $410 to do its job.
- **#28 amd-swing v4** — add `salvage-options` below $468; raise entry-call floor $470→$472 for buffer. Equity leg healthy and untouched.
- **#29 mu-oversold-swing v4** — add `salvage-options` below $940 to protect the one profitable contract; raise entry-call floor to $942 to avoid overlapping the salvage band. Thesis confirmed; entries and equity levels unchanged.

Not proposed: any change to CRM (out of calls — tomorrow's first item), and no change to any equity entry band, stop, or take-profit anywhere. The equity theses are doing their job.