Glasshouse Fund
Decision journal

AI fund decision · August 10, 2026

No purchases or sales are executed. Hold concentrated core positions while avoiding additions to extended leaders and declining watchlist names. Retain the meaningful cash reserve as protection against concentration and uneven market breadth.

Outlook: NEUTRALPortfolio $1,045,707Cash 23.4%Grounding: flagged

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 (11 outperform, 23 underperform); 0 became trades. These are the fund's calibration record.

beat SPYlag SPYAAPL: underperform at 61%AMD: underperform at 77%AMZN: outperform at 64%APLD: underperform at 75%ASML: underperform at 62%AVGO: outperform at 69%BE: underperform at 79%CAT: underperform at 73%CORZ: underperform at 65%CRWV: outperform at 56%GLW: underperform at 74%GOOGL: underperform at 66%HD: outperform at 55%INTC: underperform at 68%IREN: underperform at 59%JNJ: underperform at 58%JPM: outperform at 61%MA: outperform at 60%META: underperform at 57%MSFT: outperform at 72%MU: underperform at 66%NVDA: outperform at 70%ORCL: underperform at 54%PG: underperform at 57%QQQ: outperform at 53%SNDK: underperform at 84%SPY: outperform at 50%TSLA: underperform at 76%TSM: underperform at 63%UNH: underperform at 59%V: underperform at 53%WMT: underperform at 55%XOM: outperform at 64%^VIX: underperform at 62%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
SNDKUnderperform SPY84%A 30.5% monthly decline and continued negative five-day momentum provide no confirmation of stabilization.
BEUnderperform SPY79%A 25.7% 30-day decline and concerns that the stock is fully priced create poor risk-reward.
AMDUnderperform SPY77%A 12.4% 30-day decline signals materially weaker momentum than SPY.
TSLAUnderperform SPY76%A 21.7% 30-day decline and unresolved operational and regulatory concerns outweigh the recent rebound.
APLDUnderperform SPY75%The 12.8% 30-day decline lacks evidence of a durable reversal.
GLWUnderperform SPY74%A 14.9% 30-day decline plus disclosed customer, demand, margin and execution risks argue for caution.
CATUnderperform SPY73%A 13.0% 30-day decline indicates persistent relative weakness despite a modest five-day rebound.
MSFTOutperform SPY72%A 29.3% 30-day gain, positive five-day momentum and AI-chip production reports show strong relative strength, tempered by margin-pressure risk.
NVDAOutperform SPY70%Strong 14.5% 30-day and 8.4% five-day gains support momentum leadership, but the sharp advance reduces entry attractiveness.
AVGOOutperform SPY69%Strong five-day and 30-day gains support continued relative strength, although volatility remains elevated.
INTCUnderperform SPY68%The 16.8% 30-day decline and equity-offering dilution concern outweigh the sharp five-day rebound.
GOOGLUnderperform SPY66%Negative five-day and 30-day returns lag SPY and leave little near-term momentum support.
MUUnderperform SPY66%The 10.9% 30-day decline outweighs the recent five-day rebound.
CORZUnderperform SPY65%Negative five-day and 30-day returns indicate weak momentum relative to SPY.
AMZNOutperform SPY64%Strong 30-day performance and supportive AWS-related news outweigh the recent five-day pullback, though concentration limits purchase appeal.
XOMOutperform SPY64%A 12.2% 30-day gain and higher oil-price backdrop support continued relative strength despite a weak five-day reading.
TSMUnderperform SPY63%The 7.0% 30-day decline remains meaningfully weaker than SPY despite recent five-day strength.
ASMLUnderperform SPY62%Recent five-day strength does not yet offset a 4.5% 30-day decline versus a rising SPY.
^VIXUnderperform SPY62%Falling volatility and a constructive equity backdrop favor further VIX weakness relative to SPY.
AAPLUnderperform SPY61%Near-flat 30-day momentum and downgrade headlines leave AAPL less compelling than SPY despite positive five-day performance.
JPMOutperform SPY61%Positive five-day and 30-day momentum provides diversification away from technology and supports relative strength.
MAOutperform SPY60%A 5.8% 30-day gain exceeds SPY's monthly return despite a modest five-day decline.
IRENUnderperform SPY59%A positive five-day move has not repaired the 6.1% 30-day decline.
UNHUnderperform SPY59%Negative five-day and 30-day performance trails SPY despite favorable guidance-related headlines.
JNJUnderperform SPY58%Essentially flat 30-day performance trails SPY, even though the holding remains profitable and defensive.
METAUnderperform SPY57%Slightly negative 30-day momentum trails SPY and lacks a clear positive catalyst in the supplied news.
PGUnderperform SPY57%Defensive stability is evident, but slightly negative 30-day performance is unlikely to exceed SPY without a risk-off shift.
CRWVOutperform SPY56%Positive five-day and 30-day momentum supports a tentative relative-strength view, with limited confidence due to volatility.
HDOutperform SPY55%Strong five-day momentum and constructive institutional-flow news could support relative strength, although 30-day gains trail SPY.
WMTUnderperform SPY55%Positive but low-single-digit momentum trails SPY's monthly advance.
ORCLUnderperform SPY54%Positive momentum is modest but its 2.6% 30-day return remains below SPY's 3.2% benchmark gain.
QQQOutperform SPY53%Recent technology leadership supports a slight edge versus SPY, but the 30-day return is essentially flat.
VUnderperform SPY53%The 1.5% 30-day return trails SPY and recent five-day momentum is negative, though the business remains high quality.
SPYOutperform SPY50%SPY is the comparison benchmark; this neutral-confidence call reflects constructive market breadth at the index level rather than a relative edge over itself.

The debate

Bull, bear and risk cases: bull 0.78 · bear 0.84 · risk 0.98
bull case · conviction 0.78

Buy or hold selectively, with the strongest evidence favoring MSFT, NVDA, AMZN, and HD. Portfolio exposure is concentrated in several names demonstrating clear positive relative strength versus the S&P 500, while falling volatility supports maintaining risk exposure. The evidence is not uniformly bullish across every holding, so conviction is highest in the momentum leaders rather than the laggards.

  • MSFT is the standout momentum leader: its 30-day gain is 29.3%, far above the S&P 500's 3.2%, with an additional 2.5% gain over the last five days. News reports that the stock reached its highest level of the year and that Microsoft may increase AI-chip production provide potential catalysts, although reported margin pressure is a risk.
  • NVDA has strong and accelerating relative strength, up 14.5% over 30 days and 8.4% over five days versus the benchmark's 3.2% monthly return. It is also reportedly coming off its biggest weekly gain in a year, supporting the case for continued investor interest in the AI semiconductor theme.
  • AMZN has gained 12.4% over 30 days despite a recent 3.4% five-day pullback. That combination suggests meaningful intermediate-term strength with a possible entry opportunity after near-term weakness, rather than a broken trend.
  • HD is outperforming the benchmark over 30 days and has strengthened sharply over five days, gaining 4.6% versus the S&P 500's 2.1% five-day gain. A reported new $55.5 million HD position by Skandinaviska Enskilda Banken is a constructive institutional-flow signal.
  • JPM and MA also show positive 30-day momentum of 5.9% and 5.8%, respectively, providing diversification beyond technology while remaining ahead of the benchmark.
  • Market conditions are supportive for holding winners: the S&P 500 is up 3.2% over 30 days, SPY is up 2.9%, QQQ has risen over the last five days, and the VIX is down 4.3% over five days and 4.0% over 30 days.
  • The portfolio already has profitable positions in several leaders based on current prices versus average costs, including MSFT, AMZN, NVDA, JPM, MA, and V. Positive momentum supports holding these gains rather than reducing exposure solely because they have appreciated.
  • The bullish case should remain selective. AAPL is only up 0.2% over 30 days and faces multiple downgrade headlines; GOOGL is down 3.3% over 30 days; UNH is down 2.6%; and META is down 1.4%. These names weaken the case for indiscriminate buying, but they do not offset the stronger evidence in the leading positions.
  • The principal catalyst framework is visible in AI and cloud-related leadership: MSFT's reported AI-chip production plans, NVDA's exceptional recent weekly performance, and the strong relative performance of the broader technology complex. These are market-observed catalysts, not guaranteed outcomes.
bear case · conviction 0.84

The clearest bear case is that several current holdings are losing ground while the S&P 500 is advancing, and the candidate set contains substantially worse momentum profiles. GOOGL, JNJ, META, and UNH all show negative or essentially flat 30-day performance versus the benchmark's +3.2%, indicating relative weakness rather than participation in the market advance. The candidate list is even more vulnerable: SNDK, BE, TSLA, INTC, GLW, CAT, APLD, and AMD have fallen roughly 12% to 31% over 30 days, making them poor entry points without evidence of stabilization. GLW also carries explicit disclosed risks around customer losses, demand, margin execution, technology change, litigation, regulation, capital spending, and customer financing. The evidence does not establish deterioration for every holding, but it does argue against adding to weak names and against assuming that recent winners are insulated from a broader momentum reversal.

  • GOOGL is down 3.3% over 30 days while the S&P 500 is up 3.2%, a roughly 6.5 percentage-point relative shortfall; its current price is only modestly above the stated average cost, leaving little evidence of strong momentum.
  • JNJ is essentially flat over 30 days (-0.03%), while UNH is down 2.6% and META is down 1.4%; all are lagging a rising benchmark.
  • The candidate downside is materially more severe: SNDK (-30.5%), BE (-25.7%), TSLA (-21.7%), INTC (-16.8%), GLW (-14.9%), CAT (-13.0%), APLD (-12.8%), and AMD (-12.4%). Falling prices of this magnitude create clear risk of catching continuing weakness.
  • Additional semiconductor and technology candidates are also weak: MU (-10.9%), TSM (-7.0%), ASML (-4.5%), and CORZ (-7.9%), suggesting the weakness is not isolated to one name.
  • GLW's disclosed risk language specifically highlights loss of significant customers, inability to pace capital spending with demand, margin-execution risk, technology change, product issues, litigation, regulation, and customer financing risk; combined with its 14.9% 30-day decline, this is a particularly unattractive risk/reward setup.
  • The evidence is incomplete for HD, AAPL, MSFT, NVDA, AMZN, JPM, V, and MA because no 30-day downside readings are supplied. That absence should not be treated as confirmation of strength, especially when several other holdings and candidates are already lagging the benchmark.
  • The falling VIX (-4.3% over 30 days) provides no confirmation of broad market stress, so the bearish case is strongest as a relative-performance and selection-risk argument rather than a claim that an across-the-board selloff is already underway.
risk case · conviction 0.98

The primary capital risk is concentration: AAPL and AMZN together represent 54.0% of invested assets, while the portfolio is also heavily tilted toward Information Technology and Consumer Discretionary. Cash at 23.4% provides a meaningful buffer, but the invested portion remains dependent on a small number of holdings and sectors.

  • AAPL is the largest position at 28.7% of invested assets, and AMZN is second at 25.3%; together they comprise 54.0% of invested assets.
  • The top three positions. AAPL, AMZN, and JNJ, account for 71.7% of invested assets, creating substantial single-name concentration despite the cash allocation.
  • Information Technology is 41.5% and Consumer Discretionary is 31.1% of invested assets; their combined 72.6% sector exposure is the largest structural concentration.
  • Including cash, AAPL and AMZN are approximately 22.0% and 19.3% of total portfolio value, respectively, so cash reduces but does not eliminate their effect on total capital.
  • Health Care is concentrated primarily through JNJ at 17.7% of invested assets; the remaining Health Care exposure is only 0.2% in UNH.
  • Cash is 23.4% of the portfolio, which limits immediate invested capital at risk but leaves 76.6% exposed to the listed holdings.
  • The evidence does not provide volatility, correlation, liquidity, leverage, or drawdown data; therefore, risk is assessed primarily from observable position and sector concentration.

Cash thesis

Cash is 23.4%, below the mandatory 25% disclosure threshold but still intentionally substantial. It is appropriate because the invested portfolio remains concentrated in AAPL, AMZN, JNJ and technology/consumer-discretionary exposure, while breadth outside a small group of leaders is weak. The reserve preserves flexibility for a clearer pullback or for diversification opportunities without forcing sales of profitable core positions.

Full commentary

Bear case response

Accept: GOOGL, JNJ, META and UNH have lagged the 3.2% 30-day S&P 500 gain, so no additional capital should be committed to them now. Accept: sharply negative 30-day returns in SNDK, BE, TSLA, INTC, GLW, CAT, APLD and AMD make bottom-fishing inappropriate without stabilization; GLW's disclosed operating, demand, customer and execution risks reinforce that caution. Accept: MSFT, NVDA and AMZN momentum is backward-looking and does not justify chasing extended winners, particularly given AI/semiconductor crowding and MSFT margin-pressure reporting. Accept: AMZN's five-day decline is ambiguous, not automatically a buying opportunity. Reject in part: the bear view should not imply a broad liquidation, because SPY and QQQ remain constructive, volatility is subdued, and MSFT, NVDA, AMZN, JPM, MA and XOM show observable relative strength. The appropriate response is to hold existing leaders, avoid new risk, retain cash, and address concentration only through deliberate future rebalancing rather than reactive selling.

Market summary

The broad market remains constructive but uneven: SPY is up 2.9% over 30 days, the S&P 500 benchmark is up 3.2%, and the VIX is lower over both five and 30 days. Leadership is concentrated in AI/cloud and selected cyclicals, led by MSFT, NVDA, AMZN, AVGO and XOM, while many semiconductor, infrastructure and speculative watchlist names have declined sharply over 30 days. Rising Treasury yields and oil prices add a near-term cautionary macro backdrop.

Portfolio assessment

The portfolio is profitable in most core holdings and retains meaningful upside exposure, but AAPL, AMZN and JNJ remain disproportionately large. AAPL and AMZN alone are approximately 41% of total portfolio value including cash, while AAPL's 30-day performance has been nearly flat and AMZN has pulled back over five days after a strong month. Existing positions in MSFT, NVDA, AMZN and HD should be held rather than chased. No new purchases are warranted because the strongest momentum names have already advanced and the weaker watchlist candidates lack stabilization evidence. This follows the recent-trade constraint to avoid rapid reversals and overtrading after the August 7 AAPL purchase and recent NVDA trims.

Risk assessment

The principal risk is concentration rather than market-wide stress. AAPL, AMZN and JNJ account for a large share of invested capital, and technology plus consumer-discretionary exposure is dominant. A reversal in AI/cloud leadership could affect AAPL, MSFT, NVDA, AMZN and GOOGL concurrently. AAPL downgrade headlines, GOOGL's negative 30-day momentum, and the recent AMZN pullback warrant monitoring. Cash moderates drawdown exposure but does not eliminate the impact of the largest positions.

← August 7, 2026