Glasshouse Fund
Decision journal

AI fund decision · August 26, 2026

Maintain a neutral, low-turnover stance. Hold concentrated core positions without adding, retain the 25.9% cash reserve, and require broader relative-strength confirmation before deploying capital.

Outlook: NEUTRALPortfolio $1,034,758Cash 25.9%Grounding: ok

Trades

No trades this day. The fund held.

Market calls

A directional call (beat or lag the S&P 500 over the horizon) on every researched name, whether or not the fund traded it. 34 calls (10 outperform, 24 underperform); 0 became trades. These are the fund's calibration record.

beat SPYlag SPYAAPL: underperform at 64%AMD: underperform at 64%AMZN: outperform at 57%APLD: underperform at 57%ASML: underperform at 55%AVGO: underperform at 61%BE: outperform at 55%CAT: underperform at 61%CORZ: underperform at 72%CRWV: outperform at 55%GLW: underperform at 57%GOOGL: underperform at 54%HD: underperform at 55%INTC: underperform at 70%IREN: underperform at 54%JNJ: outperform at 62%JPM: underperform at 53%MA: outperform at 68%META: underperform at 68%MSFT: outperform at 69%MU: underperform at 55%NVDA: underperform at 56%ORCL: outperform at 64%PG: underperform at 54%QQQ: underperform at 52%SNDK: underperform at 65%SPY: outperform at 50%TSLA: underperform at 60%TSM: underperform at 55%UNH: underperform at 66%V: outperform at 66%WMT: underperform at 61%XOM: outperform at 55%^VIX: underperform at 57%50%60%70%80%90%100%
Each tick is one call, placed by stated confidence. Hover for the name.
All 34 calls, with reasoning
SymbolCallConf.Why
CORZUnderperform SPY72%A 23.67% 30-day loss and negative five-day performance indicate severe ongoing weakness.
INTCUnderperform SPY70%A 14.75% 30-day decline and 5.73% five-day loss show persistent relative weakness.
MSFTOutperform SPY69%A 26.21% 30-day gain and positive five-day momentum make MSFT the strongest current relative-strength holding.
MAOutperform SPY68%Strong positive five-day momentum and 12.67% 30-day performance demonstrate clear relative strength.
METAUnderperform SPY68%Its 9.11% 30-day decline materially lags SPY despite the recent five-day rebound.
UNHUnderperform SPY66%The 8.05% 30-day decline and price below cost outweigh the recent five-day rebound.
VOutperform SPY66%Strong five-day momentum and an 8.89% 30-day gain support continuing relative strength.
SNDKUnderperform SPY65%Large negative five-day and 30-day returns indicate a weak and volatile trend.
AAPLUnderperform SPY64%Negative 5-day and 30-day momentum versus a positive SPY trend outweighs product-launch optimism.
AMDUnderperform SPY64%Its 13.24% 30-day decline signals materially weaker relative momentum despite a positive five-day bounce.
ORCLOutperform SPY64%Its 15.03% 30-day gain and positive short-term momentum support relative outperformance, despite valuation risk.
JNJOutperform SPY62%Positive five-day action and a 7.38% 30-day gain provide defensive relative strength.
AVGOUnderperform SPY61%The 10.10% 30-day decline and negative five-day return indicate weak relative trend.
CATUnderperform SPY61%Its 8.77% 30-day decline materially trails SPY despite a modestly stable recent week.
WMTUnderperform SPY61%A 7.58% five-day loss and negative 30-day performance show meaningful relative deterioration.
TSLAUnderperform SPY60%Its 6.35% 30-day decline trails SPY and lacks a sufficiently positive near-term reversal signal.
AMZNOutperform SPY57%Its 6.62% 30-day gain exceeds SPY, though the recent five-day pullback limits conviction.
APLDUnderperform SPY57%A 4.09% 30-day loss trails SPY and the short-term rebound is insufficient confirmation.
GLWUnderperform SPY57%Negative five-day and 30-day returns trail SPY amid execution and demand-related risks.
^VIXUnderperform SPY57%The VIX remains well below its level of 30 days ago despite a recent rise, consistent with limited sustained upside in the volatility index.
NVDAUnderperform SPY56%Near-flat 30-day performance and a negative five-day return trail SPY amid elevated earnings expectations.
ASMLUnderperform SPY55%Negative 30-day performance and flat recent momentum lag the broad market.
BEOutperform SPY55%Strong five-day momentum with a nearly flat 30-day result offers a tentative relative-strength setup.
CRWVOutperform SPY55%Positive 30-day performance supports a modestly constructive view despite recent weakness.
HDUnderperform SPY55%The earnings beat supports stability, but its 1.94% 30-day gain trails SPY and recent momentum is negative.
MUUnderperform SPY55%Negative 30-day performance and flat recent trading lag the benchmark.
TSMUnderperform SPY55%Slightly negative 30-day performance trails SPY despite positive five-day momentum.
XOMOutperform SPY55%Its 4.68% 30-day return modestly exceeds SPY, although recent five-day momentum is negative.
GOOGLUnderperform SPY54%Positive momentum is still below SPY's 30-day gain and the position remains below cost.
IRENUnderperform SPY54%Its modest positive 30-day return remains below SPY and recent momentum is negative.
PGUnderperform SPY54%The defensive profile is constructive, but its negative 30-day return trails the positive benchmark.
JPMUnderperform SPY53%Positive momentum is modestly below SPY's 30-day return, producing only a low-conviction cautious view.
QQQUnderperform SPY52%Its 0.76% 30-day gain is well below SPY, reflecting narrow and uneven technology leadership.
SPYOutperform SPY50%As the required benchmark reference, SPY is expected to track itself; the forced directional classification is economically neutral.

The debate

Bull, bear and risk cases: bull 0.76 · bear 0.86 · risk 0.98
bull case · conviction 0.76

Buy or hold selectively, with emphasis on the portfolio's strongest relative-strength leaders rather than broad-based accumulation. The S&P 500 gained 3.6% over 30 days, while several holdings materially outperformed: MSFT rose 26.2%, MA 12.7%, V 8.9%, JNJ 7.4%, and AMZN 6.6%. This leadership is supported by identifiable catalysts, including Microsoft's data-center build and reported AI revenue-sharing discussions, Mastercard and Visa's strong recent momentum, Home Depot's Q2 beat and reaffirmed guidance, Apple's new AI-enabled Mac launches, and Alphabet's Waymo expansion. The case is strongest for holding winners and adding incrementally to demonstrated leaders, while treating recent laggards such as AAPL, META, NVDA, and UNH more cautiously.

  • MSFT is the clearest momentum leader: its 30-day gain of 26.2% is roughly 22.6 percentage points above the S&P 500, and its five-day momentum remains positive at 1.7%. News points to continued data-center investment and potential AI-related revenue-sharing discussions.
  • MA and V show both momentum and relative strength: MA gained 12.7% over 30 days and 4.5% over five days, while V gained 8.9% and 5.1%, respectively. Both substantially outperformed the benchmark over the month, supporting continued exposure to payments growth.
  • JNJ combines defensive characteristics with strong price action: it gained 7.4% over 30 days versus the S&P 500's 3.6% and was also positive over five days. Recent coverage includes a reported $85 billion shareholder 'consolation prize,' providing a potential corporate catalyst, although the evidence is not detailed enough to quantify it.
  • AMZN is outperforming the benchmark over 30 days, up 6.6%, despite a modest five-day pullback. Its reported discussions with China's Moonshot AI over the Kimi K3 model provide an additional AI-related catalyst beyond its existing portfolio exposure.
  • HD's 30-day return of 1.9% is not market-leading, but it is positive and the company recently beat Q2 expectations, reaffirmed guidance, and indicated tariff refunds helped maintain value. That combination supports holding rather than selling into a modest pullback.
  • GOOGL has improving short-term momentum, up 0.65% over five days and 1.4% over 30 days, while Alphabet's constructive fund memory highlights Waymo's expansion into more cities and international markets as a longer-term optionality catalyst.
  • The bull case should remain selective: AAPL is down 4.8% over 30 days, META is down 9.1%, NVDA is up only 0.5%, and UNH is down 8.0%. Those positions lack comparable current momentum, even though Apple has new AI Mac launches and the market is awaiting Nvidia earnings.
  • Risk conditions are not fully benign: the VIX rose 3.8% over five days, and market coverage cites sticky inflation and upcoming inflation data. This argues for incremental buying and concentration in relative-strength names, not an aggressive all-at-once deployment.
bear case · conviction 0.86

The strongest bearish case is that several holdings and candidates are materially underperforming a rising S&P 500, with weakness concentrated in large-cap technology and healthcare names. AAPL, META, and UNH show notably negative 30-day momentum, while UNH and GOOGL are already below cost. The candidate set contains even sharper deterioration, including CORZ, INTC, AMD, AVGO, and CAT. This creates risk that recent weakness reflects broader loss of leadership rather than isolated noise, while the available evidence provides no confirmation of a near-term reversal.

  • The S&P 500 gained 3.56% over 30 days, but AAPL fell 4.82%, META fell 9.11%, and UNH fell 8.05%, indicating substantial relative underperformance among current holdings.
  • UNH is both below its average cost by 0.86% and down 8.05% over 30 days, making it the clearest current holding with simultaneous realized weakness and deteriorating momentum.
  • GOOGL is below its average cost by 2.99%, adding downside evidence even though its 30-day momentum figure is not provided.
  • AAPL's 30-day decline of 4.82% is particularly important because it is one of the larger listed holdings by shares and is moving against a positive benchmark.
  • META's 9.11% 30-day decline is severe relative to the benchmark, while its current price is only modestly above average cost, leaving limited evidence of durable strength.
  • The candidate list shows broad weakness across technology and cyclical names: INTC fell 14.75%, AMD 13.24%, AVGO 10.10%, CAT 8.77%, and CORZ 23.67% over 30 days.
  • Additional candidate weakness is visible in SNDK at -7.41%, TSLA at -6.35%, GLW at -4.48%, APLD at -4.09%, and ASML at -3.08%, suggesting the deterioration is not confined to one name.
  • The available Corning risk disclosures identify numerous ways the downside could persist, including inability to pace capital spending to customer demand, margin-improvement execution risk, technology change, product or component issues, loss of significant customers, litigation, regulatory developments, and customer financing or payment problems.
  • Several current holdings lack positive downside-confirming momentum evidence in the supplied data, so the portfolio's apparent resilience outside the highlighted names should not be treated as evidence that the broader risk has cleared.
risk case · conviction 0.98

The largest capital risk is concentrated exposure rather than cash scarcity: AAPL, AMZN, and JNJ represent 71.2% of invested assets, while Information Technology and Consumer Discretionary together account for 70.9% of the portfolio. A material adverse move in any of these large positions or in either dominant sector could disproportionately affect portfolio capital. Cash at 25.9% provides a substantial buffer, but the invested portion remains highly concentrated.

  • AAPL is the largest position at 27.5% of invested assets, creating the biggest single-name concentration risk.
  • AMZN is 24.6%; combined with AAPL, the two positions comprise 52.1% of invested assets.
  • JNJ adds another 19.1%, bringing the top three positions to 71.2% of invested assets.
  • Information Technology is 40.4% and Consumer Discretionary is 30.5%; together they comprise 70.9% of the portfolio, indicating a pronounced sector tilt.
  • The portfolio has limited diversification outside the dominant exposures: Health Care is 19.3%, Communication Services 7.6%, and Financials 2.2%.
  • Cash is 25.9% of the portfolio, which reduces immediate exposure of total capital to equity-specific losses, but 74.1% remains invested and concentrated.
  • The evidence supports a structural concentration-risk assessment, but does not provide volatility, correlation, or drawdown data to quantify the magnitude of potential losses.

Cash thesis

Cash is $267,697, or 25.9% of portfolio value. This reserve is intentional rather than idle: it offsets concentrated exposure to AAPL, AMZN, JNJ and AI/cloud-linked equities, preserves flexibility around inflation data and Nvidia-related volatility, and avoids forcing new purchases into either extended leaders or lagging names. The opportunity cost is acceptable until market breadth improves or a diversification-oriented entry becomes more compelling.

Full commentary

Bear case response

Accept: AAPL (-4.82% over 30 days), META (-9.11%) and UNH (-8.05%) have materially underperformed a rising S&P 500, so no new capital should be allocated to them today. Accept: UNH and GOOGL are below average cost, which warrants monitoring, although a small unrealized loss alone is not a sufficient reason to sell. Partly reject: GOOGL is weak relative to SPY, but the supplied data do show positive 5-day (+0.65%) and 30-day (+1.42%) returns, so its weakness is less severe than the bear case implies. Accept: sharp declines in watchlist technology and cyclical names, including CORZ, INTC, AMD, AVGO and CAT, indicate weak breadth and argue against broad risk-on buying. Accept: the available catalysts for several holdings are not quantified earnings evidence, so they do not justify chasing winners. Reject broad liquidation: MSFT, MA, V, JNJ and AMZN retain better relative strength than the benchmark or constructive fundamentals, and the portfolio already has a substantial cash buffer. The appropriate response is to hold core positions, avoid adds to laggards and extended leaders, and wait for confirmation rather than make repeated tactical trades.

Market summary

The S&P 500 is up 3.56% over 30 days, but leadership is narrow and uneven. MSFT, MA, V, JNJ and AMZN show relative strength, while AAPL, META and UNH materially lag the benchmark. Sticky-inflation headlines, an uptick in the VIX over five days, and Nvidia earnings/inflation-event uncertainty argue against deploying cash aggressively or chasing recent winners.

Portfolio assessment

The portfolio is up modestly from its $1 million starting value but remains structurally concentrated: AAPL, AMZN and JNJ are approximately 52.9% of total portfolio value and about 71.2% of invested assets. AAPL has negative 5-day and 30-day momentum, AMZN has pulled back over five days despite positive 30-day performance, and JNJ has become a large defensive winner. No transactions are warranted today: recent AAPL buy/sell activity should not be repeated without a clearer catalyst, and selling core positions solely on short-term momentum would risk overtrading.

Risk assessment

The principal risk is concentration, not a lack of liquidity. AAPL, AMZN and JNJ dominate invested capital, while technology and consumer-discretionary exposure create correlated downside if AI/cloud leadership or consumer demand weakens. AAPL, META, UNH and GOOGL have weaker relative performance or are near/below cost, while MSFT, MA and V have strong gains that could be vulnerable to mean reversion. Elevated cash moderates total-capital drawdown risk but does not eliminate the impact of the large holdings. Fund memory reinforces disciplined incremental action, avoidance of repeated turnover, and selective trimming only when deterioration persists.

← August 25, 2026