N/A - N/A Pick - Aiplexity

Rugged mountain peaks reflected in the calm water of a high alpine lake

Research brief

Sep 28, 2026 Afternoon

04:40 PM UTC

N/A N/A N/A

Final allocation: Keep cash (N/A)

After the stock-selection N/A, a separate review compared S&P 500 exposure (VOO; SPY is an equivalent market proxy) with Robinhood Gold cash over 28 days. Gold cash APY: 3.6%, checked 2026-09-28T18:20:04.927567+00:00; matched cash return: 0.2717%. Current base APY held constant; existing Gold subscription; pre-tax eligible settled cash, no margin debit or promotional boost.

Estimated net index return: +0.4500%; forecast uncertainty: 0.8000 percentage points; conservative excess over cash: -0.6217%.

Keep eligible settled cash earning Robinhood Gold interest; do not buy VOO. The official rate snapshot at 2026-09-28T18:20:04.927567+00:00 is 3.6% APY, corroborated by Robinhood's current support page. Holding that base APY constant, the required pre-tax 28-day cash return is 100*((1+3.6/100)^(28/365)-1) = +0.2716779741%; the existing Gold subscription fee is sunk, and this assumes eligible settled cash, no margin debit and no promotional boost. Robinhood says APY can change, so this is a conditional cash forecast, not a guaranteed rate (Robinhood Gold cash rate). VOO's +0.45% estimated net buy-and-hold return falls to -0.35% after the positive 0.80-point uncertainty deduction, 0.621678 points below cash. Moreover, the indicative VOO quote is not a verified executable bid/ask, and evidence does not support a complete 60-minute-average/120-minute-confirmation stop, 15%-hard-stop/10-minute-confirmation, 1.8-times-T1 profit-barrier scenario with a defensible conditional stop-fill price and exit probabilities; no conservative executed policy return above cash can be certified. The audited single-stock N/A remains separate and is not treated as a bearish market signal. Supplied positions show no active public VOO/SPY lot, so a duplicate-index-lot prohibition is not invoked; nevertheless five Technology positions total about 6.34 of 16.44 supplied slots, and Vanguard's June 30 VOO sector snapshot was 38.01% Information Technology. Adding VOO would increase overlapping technology/AI factor exposure despite broad issuer diversification (Vanguard VOO portfolio). No order was placed.

Separate 28-calendar-day VOO review at the specified 2026-09-28 14:20:04 EDT decision time, ending 2026-10-26 at the corresponding time. The fresh Yahoo Finance feed showed VOO $705.126 at 14:20:00 EDT, above its completed September 25 SMA20 $704.06 and SMA50 $700.25; this is indicative market evidence, not an executable bid/ask or a guaranteed fill (VOO quote). The saved stock report's 13:49 EDT SPY +0.25% net forecast is earlier analyst work, not today's VOO quote or an index-buy authorization. My uncalibrated 28-day buy-and-hold analyst estimate is +0.60% VOO price return, +0.00% eligible distribution, less 0.15 percentage points estimated round-trip spread/slippage, yielding +0.45% net before uncertainty. This modest positive price judgment balances FactSet's prior rise in 2026 S&P 500 EPS estimates (+6.1% from June 30 to August 31) and VOO's above-average trading trend against a roughly 24.8x quote-reported P/E, the September 16 Fed rate increase to 3.75%-4.00%, and the 10-year Treasury yield rising from 4.79% on September 1 to 5.17% on September 25 (FactSet EPS revisions, Fed statement, Treasury yields). This is a causal judgment for exactly 28 days, not an annual historical return scaled down or a source-published forward return. The search-indexed Vanguard 2026 schedule lists September 28 as VOO's ex-date and its next ordinary ex-date in December, corroborated by Wall Street Horizon; direct retrieval of Vanguard's PDF returned 404, so the date is not independently verified from a live issuer PDF. Buying during the September 28 session is too late for that distribution, so no in-window dividend is added. The 0.03% annual fund expense ratio is already reflected in market/NAV returns and is not subtracted again; 0.15 points is trading friction, not a fund-fee charge (Vanguard fund details). A separate positive 0.80-percentage-point analyst forecast-uncertainty deduction reflects the unresolved transmission of EPS revisions into prices and the scheduled September 30 PCE, October 2 jobs, and October 14 CPI releases under recently higher rates; it is not ATR, VIX, a one-sigma move, or an additional modeled downside outcome (BEA schedule, BLS October schedule). Thus uncertainty-adjusted buy-and-hold VOO is -0.35%, below cash. Sensitivity: with the same friction and uncertainty, VOO needs more than +1.221678% price appreciation in 28 days to beat cash; +1.50% would produce +0.55% adjusted and reverse the comparison, while +1.00% would produce +0.05% adjusted and still lose. The strongest contrary evidence is sustained positive EPS revisions and the index's above-SMA trend; those make a rapid upside move plausible, but do not establish its 28-day probability. The prior Surf search human challenge recorded in stock-report.md was not retried; independent native web/issuer pages were used, and their direct URLs appear here rather than project citation IDs.

The following stock-selection report is retained as the prior research record.

Summary

Stock decision: N/A. Do not initiate a new single-stock position on the 13:49 EDT research freeze of September 28, 2026. This is not a broad-market crash call: the nine fully modeled, actionable candidates in the 25-symbol audit all have nonpositive 28-day net absolute expected return after forecast uncertainty. The other 16 rows identify unsupported essential forecast evidence rather than inventing returns. The recovery round is complete as recorded in the audit. finance.yahoofinance.yahoo
Closest apparent edges did not survive re-underwriting. NVDA's revised analyst estimate is +0.2430% net before a 1.50-percentage-point uncertainty deduction, or −1.2570% adjusted absolute. ZYME's +2.0563% net less 2.90 points is −0.8437%; CMG's +1.5417% less 1.60 points is −0.0583%. These are decider scenario estimates and arithmetic, not measured future returns or calibrated exit frequencies. finance.yahoofinance.yahoofinance.yahoo
Benchmark: the decider estimates +0.25% net SPY over the same 28 calendar days, assuming +0.40% price change, no eligible in-window ex-date dividend and 0.15% friction. SPY is a comparison, not the veto: every modeled stock fails the absolute gate before SPY is subtracted. finance.yahoossga
Allocation is not decided here. The subsequent, separate review must compare a matched-horizon, uncertainty-adjusted VOO/index outcome with compounded interest at a then-verified Robinhood Gold APY. N/A for stock selection is neither an automatic cash allocation nor a VOO recommendation.

Why Now

The current decision window follows two genuinely new issuer disclosures. NVIDIA announced a further $150 billion repurchase authorization on September 28, leaving $235 billion expected to be available through fiscal 2028; it did not report a mandated September–October repurchase schedule or a new hardware order. Its Q2 filing showed $19.7 billion of purchases and $99.3 billion of authority remaining on July 26, with repurchases subject to management discretion. On a simple capacity reconciliation, $235 billion minus the $150 billion increment implies approximately $85 billion of previously available authority immediately before the announcement, absent another simultaneous change. This inference weakens the case that the authorization itself causes incremental four-week buying. investor.nvid

Zymeworks disclosed new combined 2026 revenue guidance of $278–292 million and adjusted EBITDA guidance of $114–128 million on September 28. The underlying Theravance transaction was first announced June 29 and closed September 23; 75% of U.S. YUPELRI profit-share cash is assigned to service the $350 million OMERS note until repayment. The split between that encumbered U.S. stream and ex-U.S. royalties, together with post-close unrestricted cash, was not established in the supplied evidence. The new guide is a real catalyst, but not a directly measurable 28-day equity-cash increment. globenewswireir.zymeworksir.zymeworks

At the synchronized 13:49 EDT snapshot, NVDA was $230.245, ZYME $27.17, CMG $32.675, and SPY $767.38. These are research quotes, not a promise of later executable prices; other audit rows have their individually recorded, earlier same-session timestamps. NVDA remained below available regular-phase VWAP $231.1103, while ZYME was above its available phase VWAP $26.7489 but below its completed SMA20 $27.691. The VWAP data are feed-limited, not exchange-certified event VWAP. finance.yahoofinance.yahoofinance.yahoo

Business And Market Context

NVIDIA's hardware platform, cash generation and liquidity keep it worth monitoring, but the September buyback announcement concerns capital-return capacity, not near-term operating demand. Its August Q2 report and Q3 guidance predate today's event; the September 9 Australia capacity plan and September 28 agent-safety software announcement do not quantify funded chip orders inside this 28-day window. The portfolio already has supplied active exposure to ANET, DELL, HPE and SMCI, so another AI-infrastructure long would also add theme concentration; the supplied table, not a broker reconciliation, is the position basis. investor.nviinvestor.nviglobenewswir

The broader market does not independently warrant a crash veto. At the freeze SPY was above its completed SMA20 $764.0263 and SMA50 $759.908, but equal-weight and small-cap breadth lagged and upcoming PCE, payroll and CPI data could alter the return distribution. The +0.25% SPY forecast is an uncalibrated decider estimate, not a source-reported forward return; the already-passed September ex-date is why it assumes no in-window distribution. finance.yahoofinance.yahofinance.yahohome.treasurssgabls.gov

Factor Stack

Factor	Source-reported observation	Decision implication — analyst assessment
Catalyst/news	NVIDIA's fresh $150bn authorization extends through FY28; its filing also shows substantial preexisting capacity and discretionary execution. investor.nvid	A fresh headline is not evidence of incremental 28-day buying or a barrier-exit frequency.
Growth, profitability and liquidity	NVIDIA reported Q2 operating cash flow of $24.077bn; ZYME published combined 2026 non-GAAP guidance while its U.S. profit share is partly assigned to debt service. investor.nviglobenewswire	Financial resilience supports monitoring, but does not close the catalyst-to-common-equity forecast bridge.
Valuation and repricing	KOD's DAYBREAK result met both one-year noninferiority endpoints, but the 13:04 quote was already about 172% above the prior close; NTRA's Japan commercialization was expected in 2027. ir.kodiakfinance.yahoinvestor.nat	Genuine news can be fully or excessively consumed before a new entry.
Momentum and entry	NVDA was above its completed SMA20/50 yet below available phase VWAP; ZYME was above phase VWAP but below its completed SMA20, with partial-session turnover below its 20-day full-session average. finance.yahoofinance.yahoo	Neither entry imperfection is a hard veto; both reduce confidence in travel beyond first resistance to the policy profit barrier.
Sentiment/social	Earlier same-day research found active NVDA discussion containing unverified China-purchase and buyback claims; ZYME social search found only a one-vote syndication post, and KOD's trial-result Reddit discussion had two votes. redditredditreddit	These checks show attention, not independent evidence of orders, adoption or 28-day returns.
EPS revisions	No verified near-term NVIDIA order/estimate uplift or ZYME consensus-revision bridge was obtained in the supplied research. globenewswire	Treat the missing revision transmission as model uncertainty, not as proof of a negative revision.

Investment Thesis

The thesis is to withhold an immediate single-stock buy until one candidate has a supported positive absolute policy return after uncertainty. This is a forecast-evidence decision, not a confidence threshold or a rule requiring every stock to beat SPY. The completed recovery reviewed 25 unique symbols: nine with fully specified scenarios and 16 whose essential target, stop-fill or exit-probability evidence was not supported. The discovery work included one additional 503-constituent S&P 500 screen and independent idea-search lanes after a challenged search service required human intervention; the challenged service was not retried. The screen's ordering was an idea source, not a return probability. finance.yahooglobenewswirsec.govca.investingglobenewswirmarketbeat

Policy mathematics — decider estimates, not observed outcomes. The fixed horizon is 28 calendar days from entry (to October 26 at the corresponding time). The authored stop is evaluated using a trailing 60-minute average held below its threshold for 120 minutes; a 15% hard-loss condition requires 10 minutes of confirmation. T1 is a four-week reference, not the full-position sale. The actual profit exit is at 1.8 times T1's percentage gain from entry. Early-exit proceeds are held in cash to the scheduled horizon; scenario returns allow at least 0.1% adverse slippage plus any additional stated friction. Gaps, thin liquidity and confirmation delays can make actual stop fills materially worse than the authored stop or even the conditional stop_exit_price estimate. These mechanics are policy assumptions, not guaranteed execution.

Candidate	Quote time / entry	T1; actual full-profit barrier	Estimated policy EV; uncertainty; adjusted absolute	Why it does not pass
NVDA	13:49 EDT / $230.245 finance.yahoo	$239; $246.004 (calculation, +6.8444% barrier)	+0.2430%; 1.50pp; −1.2570% (decider estimates/calculation)	Prior $85bn authority and no required near-term buyback-flow change; first resistance $233.21 and 52-week high $236.54 precede the barrier. investor.nvidfinance.yahoo
ZYME	13:49 EDT / $27.17 finance.yahoo	$29.50; $31.364 (calculation, +15.4361% barrier)	+2.0563%; 2.90pp; −0.8437% (decider estimates/calculation)	Barrier marginally exceeds the $31.34 52-week high; encumbered cash and thin participation make the forecast fragile. finance.yahooglobenewswire
CMG	13:49 EDT / $32.675 finance.yahoo	$35.40; $37.58 (calculation)	+1.5417%; 1.60pp; −0.0583% (decider estimates/calculation)	Intraday reversal lacks newly established chain-wide traffic or sales economics and remains below completed SMA20/50. ir.chipotlefinance.yahoo

For NVDA, the decider's 30% profit / 35% stop / 35% time outcome assumptions use a $222 authored stop but a worse $218.50 conditional stop exit, +0.5% conditional time exit and 0.20% friction. The cited quote supports the observed price and levels; the probabilities and conditional returns are uncalibrated analyst assumptions, not facts from that quote. A coherent downside test of 25/40/35, $218 stop exit, flat time exit and 0.25% friction produced −0.6662% net EV and a 0.9092pp minimum uncertainty consistency floor, below the declared 1.50pp deduction. For ZYME, a 20/37/43 adverse case with $24.60 stop exit, flat time exit and 0.35% friction gave −0.7626% EV and a 2.8189pp floor, covered once by its 2.90pp deduction. These stresses test sensitivity; they do not empirically calibrate exit odds. finance.yahoofinance.yahooinvestor.nvidglobenewswire

Against estimated SPY +0.25%, the calculated uncertainty-adjusted excess returns are NVDA −1.5070 percentage points, ZYME −1.0937 points, and CMG −0.3083 points. Negative excess is not the reason for N/A; their adjusted absolute returns above are already nonpositive. The full audit retains the separate T1 and first resistance, all actionable scenario inputs and estimates, and one stated hard-rejection reason for each unmodeled candidate.

Risks And Thesis Breakers

The principal risk of waiting is missing a rapid repricing if actual repurchase activity, new funded AI orders, or ZYME's retained economics prove much stronger than the present evidence. Conversely, buying solely on the NVIDIA authorization could mistake long-dated capacity for immediate demand; the historical Q2 buyback pace cannot establish September–October flow. For ZYME, neither its adjusted EBITDA guide nor the U.S. profit share alone resolves unrestricted cash, GAAP conversion or the royalty split. investor.nvidglobenewswire

Reconsideration tests — analyst thresholds, not standing buy orders: NVDA needs observable incremental repurchase execution versus prior authority or new funded orders/estimate revisions, followed by accepted trading above the available $231.11 phase VWAP and through $233.21/$236.54 obstacles with a newly frozen stop and scenario. ZYME needs a verified post-close unrestricted-cash and U.S./ex-U.S. royalty bridge, bid–ask depth, and a completed hold above $27.691 before rebuilding its profit-barrier odds. These are future research triggers, not a claim that a later entry is currently executable. finance.yahoofinance.yahooglobenewswire

The remaining material limitations are dynamic issuer-event calendars, macro outcomes, order-book depth, an independently reconciled broker-position list and feed-limited phase VWAP. A challenged search source remained unavailable; independent issuer, filing and native-source checks were used rather than interpreting its failure as no opportunity. The decider supplied no market_regime_evidence field; the audit uses the non-crash reason code and the cited SPY trend instead, so no crash-specific evidence field has been invented. The audit's research certification does not certify future prices, brokerage availability or forecast calibration. finance.yahoofinance.yahooinvestor.nvid

Investment Recommendation

N/A — no immediate single-stock buy at the 13:49 EDT research freeze. The authorized reason is no_positive_actionable_candidate_after_recovery: one additional generation round was completed and all nine actionable 28-day scenarios failed the expected net absolute return minus evidence-based uncertainty > 0 gate; the remaining 16 audit rows identify specific unsupported inputs. Do not translate the unfilled stock slot into an automatic cash or index order. The next allocation decision separately compares VOO against current verified Robinhood Gold cash yield over the same horizon. A new stock selection requires refreshed quotes, source-supported scenario probabilities and stop-exit estimates, and a newly positive absolute gate—not merely a low confidence label or SPY outperformance.