Glasshouse Fund
Decision journal

AI fund decision · August 14, 2026

Maintain a neutral, risk-aware posture. Hold core winners without chasing them, do not add to lagging positions, avoid weak watchlist momentum names, and retain the elevated cash reserve because concentration risk remains the portfolio's central vulnerability.

Outlook: NEUTRALPortfolio $1,031,801Cash 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 (17 outperform, 17 underperform); 0 became trades. These are the fund's calibration record.

beat SPYlag SPYAAPL: underperform at 67%AMD: underperform at 73%AMZN: outperform at 61%APLD: outperform at 55%ASML: outperform at 66%AVGO: outperform at 58%BE: underperform at 56%CAT: underperform at 74%CORZ: underperform at 66%CRWV: outperform at 63%GLW: underperform at 78%GOOGL: underperform at 64%HD: underperform at 58%INTC: underperform at 59%IREN: outperform at 64%JNJ: outperform at 57%JPM: outperform at 68%MA: outperform at 67%META: underperform at 70%MSFT: outperform at 71%MU: outperform at 57%NVDA: outperform at 65%ORCL: outperform at 69%PG: underperform at 56%QQQ: underperform at 53%SNDK: underperform at 71%SPY: underperform at 50%TSLA: underperform at 70%TSM: outperform at 58%UNH: underperform at 68%V: outperform at 65%WMT: outperform at 59%XOM: outperform at 66%^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
GLWUnderperform SPY78%A 16.95% 30-day decline and disclosed customer, financing, execution and litigation risks support continued caution versus SPY.
CATUnderperform SPY74%CAT's 10.10% 30-day decline is a clear momentum deficit versus SPY despite a small five-day recovery.
AMDUnderperform SPY73%AMD's 13.42% 30-day decline signals materially weaker momentum than SPY despite a modest recent stabilization.
MSFTOutperform SPY71%MSFT's 29.03% 30-day return is exceptional versus SPY, although the strength is extended and warrants no chase.
SNDKUnderperform SPY71%The 20.24% 30-day decline remains a severe momentum failure despite the sharp five-day rebound.
METAUnderperform SPY70%META's 11.09% 30-day decline is a significant momentum deficit despite its small positive five-day return.
TSLAUnderperform SPY70%TSLA's 16.63% 30-day decline outweighs its recent five-day bounce and reflects unresolved relative weakness.
ORCLOutperform SPY69%ORCL has strong positive 5-day and 30-day momentum, indicating sustained relative leadership.
JPMOutperform SPY68%JPM's positive five-day momentum and 7.92% 30-day gain demonstrate broad relative strength.
UNHUnderperform SPY68%UNH is below cost and down 6.02% over 30 days, leaving its trend materially behind SPY despite possible cost-trend support.
AAPLUnderperform SPY67%Negative 5-day and 30-day momentum, plus valuation concerns in coverage, leave AAPL likely to trail SPY near term.
MAOutperform SPY67%MA's positive returns over both horizons support continued relative strength in payment networks.
ASMLOutperform SPY66%ASML's positive 5-day and 30-day returns indicate semiconductor-equipment leadership relative to SPY.
CORZUnderperform SPY66%A 15.79% 30-day decline and elevated volatility outweigh the recent bounce and data-center headline support.
XOMOutperform SPY66%XOM's 14.21% 30-day gain and positive five-day momentum demonstrate strong relative performance.
NVDAOutperform SPY65%Positive returns over both periods and supportive AI-infrastructure news support continued relative strength.
VOutperform SPY65%Positive five-day and 30-day returns support continued relative strength in transaction-growth exposure.
GOOGLUnderperform SPY64%Negative 5-day and 30-day returns leave GOOGL's near-term trend below SPY despite potential valuation support.
IRENOutperform SPY64%Strong positive returns over both five and 30 days indicate improving relative momentum, with high volatility risk.
CRWVOutperform SPY63%Strong 5-day and 30-day gains and reported AI-infrastructure demand support relative strength, albeit with high volatility.
^VIXUnderperform SPY62%Declining volatility and a record equity market favor lower VIX levels relative to SPY's constructive trend.
AMZNOutperform SPY61%Its 8.07% 30-day gain remains well ahead of SPY, though the recent pullback lowers conviction.
INTCUnderperform SPY59%A positive five-day move does not yet offset INTL's negative 4.81% 30-day return versus SPY.
WMTOutperform SPY59%WMT's positive returns over both time frames support a modest defensive relative-strength outlook.
AVGOOutperform SPY58%A positive 4.46% 30-day return supports modest outperformance potential despite recent five-day weakness.
HDUnderperform SPY58%HD's negative five-day return and essentially flat 30-day performance lag the rising market.
TSMOutperform SPY58%Positive five-day momentum and a nearly flat 30-day result support a tentative improvement relative to SPY.
JNJOutperform SPY57%Positive five-day and 30-day returns plus potential talc-litigation resolution support modest defensive outperformance.
MUOutperform SPY57%Strong five-day momentum suggests a recovery, but the negative 30-day return limits confidence.
BEUnderperform SPY56%The large five-day rebound follows a negative 30-day trend, making durability of relative strength uncertain.
PGUnderperform SPY56%Slightly negative returns over both horizons imply muted momentum relative to SPY.
APLDOutperform SPY55%Positive five-day momentum and a relatively contained 30-day decline support a tentative relative-strength recovery view.
QQQUnderperform SPY53%QQQ's modest 0.90% 30-day gain trails SPY's 2.99%, though the confidence edge is low.
SPYUnderperform SPY50%As the benchmark itself, SPY has no meaningful relative edge; this is a neutral placeholder required for full coverage.

The debate

Bull, bear and risk cases: bull 0.78 · bear 0.89 · risk 0.98
bull case · conviction 0.78

Buy or hold the portfolio with emphasis on the holdings demonstrating clear relative strength. The market backdrop is constructive: the S&P 500 gained 2.99% over 30 days, reached a record above 7,800, and benefited from cooler-than-expected inflation and falling oil prices. Several portfolio leaders substantially outperformed that benchmark, especially MSFT, AMZN, JPM, MA, NVDA, and V, providing evidence that exposure to profitable technology, digital payments, financials, and AI remains rewarded. The portfolio also retains approximately $267,697 in cash, allowing selective additions rather than forced buying. The case is strongest for maintaining winners and using weakness in high-quality laggards only selectively, while recognizing that AAPL, GOOGL, HD, and UNH are currently showing weaker momentum.

  • MSFT is the standout momentum leader: its 30-day return of 29.0% dramatically exceeds the S&P 500's 3.0%, despite a modest 0.6% five-day pullback. Recent coverage also references a post-earnings valuation review and the emerging Windows Surface Phone, providing potential catalysts beyond the existing trend.
  • AMZN gained 8.1% over 30 days, more than twice the benchmark return, even after a 3.4% five-day decline. That combination supports a pullback-within-an-uptrend interpretation rather than a broken trend.
  • JPM and MA are demonstrating broad-based relative strength: JPM rose 7.9% and MA 7.7% over 30 days, while both also advanced over the past five days. V is similarly strong, up 4.9% over 30 days and 1.0% over five days. The simultaneous strength across banks and payment networks supports continued exposure to financial and transaction-growth themes.
  • NVDA is up 6.8% over 30 days and 0.6% over five days. News that Nvidia rallied alongside Supermicro provides a near-term AI-infrastructure catalyst, while the stock's positive relative performance supports holding rather than reducing solely on short-term volatility.
  • JNJ is a defensive positive contributor, gaining 2.0% over 30 days versus the market's 3.0%, with a 1.1% five-day rise. Its proposed $5.5 billion talc settlement was reported as a possible turning point and helped the stock advance after hours, creating a potentially important litigation-overhang catalyst, although the settlement's ultimate value remains uncertain.
  • META's five-day gain of 0.5% and current price above its average cost support holding, even though its 30-day momentum is weak at -11.1%. The evidence favors patience rather than adding aggressively until relative strength improves.
  • The market trend itself is supportive: the S&P 500 is at a record, has posted roughly 3% monthly gains, and was positioned for a third consecutive weekly gain. VIX also declined 2.7% over 30 days, indicating that the constructive market environment has not yet been accompanied by a volatility surge.
  • AAPL and GOOGL are the clearest momentum drags, down 3.1% and 3.0% over 30 days, while HD is essentially flat and UNH is down 6.0%. These positions argue for selectivity, not for abandoning the broader portfolio: the strongest evidence supports reallocating incremental capital toward relative-strength leaders rather than averaging indiscriminately into laggards.
  • Portfolio-level alignment is favorable: MSFT, AMZN, JPM, MA, NVDA, and V are all currently above their stated average costs, while JNJ, META, and HD are also above cost. This provides a profitable base from which to hold winners and manage weaker positions without being forced into broad selling.
bear case · conviction 0.89

The portfolio should be treated defensively: several holdings are losing money or materially lagging a benchmark that gained 2.99% over the last 30 days, while multiple candidates show sharp downside momentum. The clearest evidence of deterioration is concentrated in UNH, GOOGL, AAPL, and META; adding to these or pursuing similarly weak candidates risks reinforcing relative underperformance.

  • UNH is both underwater versus average cost (-1.05%) and down 6.02% over 30 days, roughly 9 percentage points behind the S&P 500 benchmark.
  • GOOGL is also underwater (-1.77%) and down 3.03% over 30 days, underperforming the benchmark by about 6 percentage points.
  • AAPL is down 3.11% over 30 days despite being only modestly above its average cost, indicating recent weakness has materially eroded its position-level gain and represents roughly 6 percentage points of benchmark underperformance.
  • META has the most concerning held-stock momentum signal: down 11.09% over 30 days, approximately 14 percentage points behind the benchmark. Its current price remains above average cost, but the recent trend is sharply adverse.
  • HD is down 0.47% over 30 days while the benchmark rose 2.99%, another sign of relative stagnation even though the position remains above cost.
  • The candidate list contains several severe momentum failures: SNDK (-20.24%), GLW (-16.95%), TSLA (-16.63%), CORZ (-15.79%), AMD (-13.42%), and CAT (-10.10%). These are poor entry candidates absent evidence of stabilization.
  • GLW carries additional disclosed risks including customer losses, customer financing and receivables problems, technology change, product or component issues, litigation, regulatory developments, and the possibility that capital spending cannot be matched to demand. Combined with its 16.95% 30-day decline, the downside case is especially strong.
  • The evidence is not uniformly bearish across every holding, but the benchmark-relative weakness and concentration of negative momentum argue against treating recent price weakness as an isolated issue or aggressively adding to laggards.
risk case · conviction 0.98

The primary capital-risk exposure is concentration: AAPL and AMZN together represent 52.2% of invested assets, while the three largest positions. AAPL, AMZN, and JNJ, represent 70.7%. Sector exposure is also concentrated, with Information Technology, Consumer Discretionary, and Health Care comprising 90.1% of invested assets. Cash at 25.9% provides a substantial liquidity buffer, but the invested portion remains heavily dependent on a small number of holdings and sectors.

  • AAPL is the largest position at 27.0% of invested assets, creating the greatest single-name concentration risk.
  • AMZN is 25.2% of invested assets; combined with AAPL, the two positions account for more than half of invested assets.
  • JNJ is 18.5% of invested assets, so the top three positions total 70.7% of invested assets.
  • Information Technology is 40.3% and Consumer Discretionary is 31.1% of invested assets; together they account for 71.4%.
  • Health Care adds another 18.7%, leaving only 9.9% of invested assets outside the three largest sectors.
  • Cash is 25.9% of the portfolio and reduces overall invested exposure, but it does not diversify the risk within the invested sleeve.
  • No leverage or short exposure is shown; the largest identified risks are concentration and limited sector diversification rather than financing exposure.

Cash thesis

Cash of $267,697, or 25.94% of portfolio value, is intentionally retained because the invested sleeve is concentrated in AAPL, AMZN and JNJ and heavily exposed to technology, consumer discretionary and health care. Although the index trend is positive, several large holdings are lagging SPY and there is no tool-supported catalyst requiring immediate deployment. Cash provides downside protection and flexibility to add only after stronger confirmation or to manage a material deterioration in concentrated holdings.

Full commentary

Bear case response

Accept: AAPL, GOOGL, UNH, META and HD have materially lagged SPY over 30 days, so averaging down is not justified without stabilization. Accept: UNH and GOOGL being below cost reinforces the need for patience, not automatic loss realization, because the current positions are small for UNH and moderate for GOOGL. Accept: META's -11.09% 30-day return is the sharpest adverse momentum signal among held names; it should be held but not increased until relative strength improves. Accept: severe 30-day declines in SNDK, GLW, TSLA, CORZ, AMD and CAT make them unsuitable new entries today; GLW's disclosed operational, customer, financing, litigation and regulatory risks add to that caution. Partially reject: broad defensiveness does not require selling all risk assets because MSFT, AMZN, NVDA, JPM, MA, V and JNJ have positive relative momentum or defensive characteristics and are above cost. Accept the rebuttal that AMZN and MSFT should not be chased after strong runs, particularly given AMZN's 3.41% five-day decline and existing concentration. The final action is therefore no new purchases and no fresh trims today: retain winners, monitor laggards, and preserve cash.

Market summary

The broad backdrop is constructive: SPY gained 2.99% over 30 days, the S&P 500 reached a record above 7,800, inflation news was favorable, and VIX fell to 14.48. However, leadership is uneven. MSFT, AMZN, NVDA, JPM, MA, V, JNJ and XOM show relative strength, while AAPL, GOOGL, META, UNH and HD lag the index. The appropriate stance is to retain established winners, avoid chasing extended momentum, and wait for clearer confirmation before deploying additional capital.

Portfolio assessment

Portfolio value is $1.032 million with 25.94% in cash. The portfolio has meaningful gains in MSFT, AMZN, NVDA, JPM, V and MA, but invested assets remain highly concentrated: AAPL, AMZN and JNJ are the dominant exposures. AAPL and AMZN have recently weakened, while AAPL was already trimmed on 2026-08-12. In line with risk lessons emphasizing selective trimming, avoidance of repeated trading, and preservation of liquidity, no additional transaction is warranted today.

Risk assessment

The primary risk is concentration rather than broad market stress. AAPL and AMZN together comprise more than half of invested assets, and AAPL, AMZN and JNJ comprise roughly 70.7% of invested assets. Technology and consumer-discretionary exposure could decline together if AI/cloud leadership or consumer demand weakens. AAPL, GOOGL, META, HD and UNH have trailed SPY over 30 days, while MSFT's exceptional recent gain also creates pullback risk. The cash reserve mitigates total-portfolio drawdown risk but does not diversify the invested sleeve.

← August 13, 2026