Glasshouse Fund
Decision journal

AI fund decision · August 24, 2026

Maintain all current holdings, execute no purchases or sales, and retain the 25.8% cash reserve. The portfolio has enough exposure to proven leaders, while concentration risk and broad pockets of relative weakness argue against chasing strength or averaging down.

Outlook: NEUTRALPortfolio $1,038,101Cash 25.8%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. 33 calls (14 outperform, 19 underperform); 0 became trades. These are the fund's calibration record.

beat SPYlag SPYAAPL: underperform at 67%AMD: underperform at 74%AMZN: outperform at 55%APLD: underperform at 76%ASML: outperform at 56%AVGO: underperform at 67%BE: underperform at 72%CAT: underperform at 68%CORZ: underperform at 84%CRWV: underperform at 62%GLW: underperform at 64%GOOGL: underperform at 60%HD: underperform at 53%INTC: underperform at 78%IREN: underperform at 58%JNJ: outperform at 72%JPM: outperform at 57%MA: outperform at 70%META: underperform at 75%MSFT: outperform at 71%MU: outperform at 54%NVDA: outperform at 56%ORCL: outperform at 59%PG: underperform at 57%QQQ: underperform at 54%SNDK: underperform at 65%SPY: outperform at 50%TSLA: outperform at 55%TSM: outperform at 55%UNH: underperform at 72%V: outperform at 69%WMT: underperform at 73%XOM: outperform at 66%50%60%70%80%90%100%
Each tick is one call, placed by stated confidence. Hover for the name.
All 33 calls, with reasoning
SymbolCallConf.Why
CORZUnderperform SPY84%CORZ has fallen 20.5% over 30 days and 11.5% over five days, showing persistent downside momentum.
INTCUnderperform SPY78%INTC is down 13.0% over five days and 7.2% over 30 days amid chip-sector weakness.
APLDUnderperform SPY76%APLD's 12.7% five-day decline and negative 30-day return indicate severe near-term momentum deterioration.
METAUnderperform SPY75%META's 14.9% 30-day decline and negative five-day momentum indicate pronounced relative weakness.
AMDUnderperform SPY74%AMD is down 6.0% over 30 days and 6.5% over five days amid broad semiconductor weakness.
WMTUnderperform SPY73%WMT's 9.1% five-day and 7.4% 30-day declines show substantial negative momentum following earnings-related weakness.
BEUnderperform SPY72%A 13.2% five-day decline signals high volatility and poor short-term momentum despite a positive 30-day return.
JNJOutperform SPY72%JNJ's 8.6% 30-day gain and 3.0% five-day gain demonstrate sustained relative strength versus SPY.
UNHUnderperform SPY72%UNH is down 7.5% over 30 days and remains below cost, indicating unfavorable relative momentum despite favorable earnings commentary.
MSFTOutperform SPY71%MSFT's 20.3% 30-day gain and positive five-day return remain substantially stronger than SPY, despite valuation concerns.
MAOutperform SPY70%MA is up 6.1% over 30 days and 3.3% over five days, displaying clear payments-sector relative strength.
VOutperform SPY69%V has positive five-day and 30-day momentum, with a 3.4% five-day gain providing strong recent confirmation.
CATUnderperform SPY68%CAT is down 6.1% over five days and 4.2% over 30 days, materially trailing SPY.
AAPLUnderperform SPY67%Negative 30-day momentum of 5.2% versus SPY's gain and concentrated portfolio ownership favor caution despite a positive five-day rebound.
AVGOUnderperform SPY67%Negative five-day and 30-day returns alongside the current semiconductor selloff suggest weaker near-term relative performance.
XOMOutperform SPY66%XOM's 12.0% 30-day gain and positive five-day return show strong relative strength, though geopolitical oil headlines add volatility.
SNDKUnderperform SPY65%SNDK's strong 30-day performance is being challenged by a 10.7% five-day selloff and negative memory-chip news.
GLWUnderperform SPY64%GLW's negative five-day and 30-day returns, alongside disclosed demand, margin and customer-financing risks, support caution.
CRWVUnderperform SPY62%Strong 30-day performance is offset by a 17.1% five-day decline, making the near-term risk/reward unfavorable versus SPY.
GOOGLUnderperform SPY60%GOOGL's 2.0% 30-day decline trails SPY, and Waymo optionality is longer-term rather than a confirmed near-term catalyst.
ORCLOutperform SPY59%ORCL's 20.7% 30-day gain supports relative strength, though AI-spending, leverage and execution concerns constrain confidence.
IRENUnderperform SPY58%Positive 30-day momentum is outweighed by a 6.7% five-day decline and elevated AI-infrastructure volatility.
JPMOutperform SPY57%JPM's 3.8% 30-day return slightly exceeds SPY, although the recent five-day decline limits conviction.
PGUnderperform SPY57%PG's small five-day gain does not offset its 2.3% 30-day decline relative to SPY.
ASMLOutperform SPY56%ASML retains a positive 30-day return despite a sharp chip-led five-day pullback, supporting only a low-confidence relative-strength view.
NVDAOutperform SPY56%NVDA's 5.6% 30-day gain exceeds SPY, but a 4.6% five-day pullback and earnings-event risk limit conviction.
AMZNOutperform SPY55%Positive 30-day momentum roughly matches SPY, and cloud/AI optionality supports a modest relative edge despite a negative five-day return.
TSLAOutperform SPY55%TSLA's 6.9% five-day rebound supports a tentative relative-strength view, but its negative 30-day return limits conviction.
TSMOutperform SPY55%TSM's 4.1% 30-day gain modestly exceeds SPY, although the five-day semiconductor decline creates near-term risk.
MUOutperform SPY54%MU's 11.7% 30-day gain is strong, but its 4.4% five-day decline amid chip weakness keeps confidence low.
QQQUnderperform SPY54%QQQ's 2.5% 30-day gain trails SPY and recent technology weakness is a near-term headwind.
HDUnderperform SPY53%HD has only modest 30-day gains and slightly negative five-day momentum, leaving it less compelling than SPY.
SPYOutperform SPY50%SPY is the comparison benchmark; expected tracking versus itself is neutral, so confidence is intentionally 0.50.

The debate

Bull, bear and risk cases: bull 0.72 · bear 0.87 · risk 0.97
bull case · conviction 0.72

Buy or hold selectively: the portfolio has several clear relative-strength leaders, while the broader market remains constructive. The S&P 500 is up 3.5% over 30 days, volatility has fallen 18.9%, and multiple holdings are outperforming that benchmark, most notably MSFT, JNJ, MA, NVDA, and JPM. Recent weakness in technology and chip stocks appears concentrated rather than evidence of a broad market breakdown, creating an opportunity to maintain exposure to companies with strong intermediate momentum and identifiable catalysts.

  • MSFT is the strongest momentum holding: up 20.3% over 30 days versus 3.5% for the S&P 500, with an additional 0.8% gain over five days. Reported catalysts include its data-center build and an upcoming Xbox gaming roster, although the stock is described as trading near fair value after a 66% run.
  • JNJ combines strong momentum with potential company-specific catalysts: it is up 8.6% over 30 days and 3.0% over five days, materially outperforming the benchmark. Recent coverage highlights why the stock is up and an $85 billion 'consolation prize' for shareholders.
  • MA is another high-quality relative-strength leader, up 6.1% over 30 days and 3.3% over five days. V is also gaining over both periods, up 3.1% over 30 days and 3.4% over five days, supporting continued exposure to payments.
  • NVDA is up 5.6% over 30 days despite a 4.6% five-day pullback and a market session in which chip stocks weighed on the indices. The upcoming earnings period is an explicit near-term catalyst, and the intermediate trend remains stronger than the S&P 500.
  • JPM is up 3.8% over 30 days, slightly ahead of the benchmark, while AMZN is approximately in line at 3.5%. These provide additional exposure to areas that have held up during the recent market softness.
  • GOOGL's short-term momentum is weak, but the constructive long-term evidence includes Waymo expanding to more domestic cities, entering international markets, and scaling operations. That offers a credible strategic catalyst even though the stock is down 2.0% over 30 days.
  • AAPL's evidence is mixed: a recent headline cites a fresh record and HSBC sees another 10% upside, but the measured 30-day return is negative 5.2%. That supports holding or monitoring rather than making it the primary add.
  • The market backdrop is not uniformly risk-off: the S&P 500 is up 3.5% over 30 days and the VIX is down 18.9% over the same period. The recent chip-led decline therefore looks more like sector rotation or consolidation than confirmation of a broad trend failure.
bear case · conviction 0.87

The portfolio is vulnerable to a broad momentum deterioration that the headline S&P 500 gain masks. Several holdings are already underperforming sharply over 30 days or trading below cost, while multiple candidate names show even more severe weakness. The strongest bear case is that recent gains in many positions may be backward-looking, with leadership weakening across technology, communications, healthcare, retail, and industrial exposure.

  • AAPL is a meaningful holding and has fallen 5.2% over 30 days despite remaining above its average cost, indicating deteriorating near-term momentum beneath a still-positive unrealized return.
  • UNH is both below cost (-2.2%) and down 7.5% over 30 days, providing direct evidence of active weakness in an existing position.
  • GOOGL and META are below cost, down 0.8% and 1.6% respectively; their 30-day declines are materially worse at 2.0% and 14.9%. META is especially concerning because the loss is accompanied by the sharpest weakness among the current holdings.
  • The portfolio's technology and growth exposure is exposed to a weakening group: AAPL, GOOGL, and META are all losing momentum, while candidate semiconductor names AMD (-6.0%), AVGO (-2.6%), and INTC (-7.2%) are also deteriorating.
  • Several candidates show severe downside momentum rather than merely flat performance: CORZ is down 20.5% over 30 days, META is down 14.9%, and INTC and WMT are down more than 7%. Buying into these names would risk treating falling prices as opportunities without evidence of stabilization.
  • Weakness is not confined to one sector: CAT is down 4.2%, PG 2.3%, WMT 7.4%, and UNH 7.5% over 30 days. That breadth raises the risk that the problem is portfolio-wide momentum deterioration rather than an isolated company issue.
  • The S&P 500 rose 3.5% over the same 30-day period, so the negative returns in AAPL, GOOGL, META, and UNH represent relative weakness versus the benchmark. Underperformance during a rising index is a warning sign for downside if the broader market stops supporting prices.
  • GLW's cautionary filing evidence explicitly identifies risks including inability to match capital spending with customer demand, margin pressure, technology change, customer financing problems, loss of significant customers, litigation, regulatory developments, and product-performance issues. Combined with GLW's 2.1% 30-day decline, this makes it a weak candidate for additional exposure.
  • The decline in the VIX does not validate the laggards; it may instead leave the portfolio exposed if currently calm conditions reverse while several holdings and candidates already exhibit negative momentum.
risk case · conviction 0.97

Capital risk is dominated by concentration rather than broad diversification: AAPL, AMZN, and JNJ comprise 71.4% of invested assets, while Information Technology and Consumer Discretionary together represent 70.7% of invested assets. The 25.8% cash allocation reduces immediate market exposure but does not offset the portfolio’s reliance on a few holdings and sectors.

  • AAPL is the largest position at 27.3% of invested assets, followed by AMZN at 24.9% and JNJ at 19.2%. A material adverse move in any one of these holdings could significantly affect portfolio capital.
  • The top three positions total 71.4% of invested assets, leaving relatively little invested exposure outside them.
  • Information Technology is 40.0% and Consumer Discretionary is 30.7% of invested assets; their combined 70.7% indicates substantial sector concentration.
  • Health Care adds another 19.4% of invested assets, primarily through JNJ at 19.2%, creating both single-name and sector concentration.
  • The remaining sectors are small: Communication Services is 7.7% and Financials 2.2%, limiting diversification across the displayed sector exposures.
  • Cash is 25.8% of the portfolio, providing a substantial uninvested buffer and reducing total portfolio exposure, but the invested portion remains highly concentrated.

Cash thesis

Cash is 25.8% of portfolio value. This is intentional rather than idle: it offsets concentrated single-name and technology/consumer-discretionary exposure, avoids averaging down into laggards such as AAPL, META, UNH and weaker semiconductor names, and preserves flexibility for a clearer market pullback or confirmed company-specific catalyst. With SPY positive but breadth mixed and chip stocks under pressure, the opportunity cost of cash is currently acceptable.

Full commentary

Bear case response

Accept: AAPL's 5.2% 30-day decline versus SPY's 3.5% gain is meaningful relative weakness, especially given its portfolio size; no additional purchase is appropriate. Accept: UNH is below cost and down 7.5% over 30 days, while META is below cost and down 14.9%; both warrant monitoring and no averaging down. Accept: GOOGL's negative 30-day return and below-cost position weaken the near-term case despite strategic Waymo optionality. Accept: semiconductor and related watchlist weakness in AMD, AVGO, INTC, CORZ and other names argues against buying the dip before stabilization, particularly with NVDA earnings as a binary event. Accept in part: weakness outside technology, including CAT, PG and WMT, indicates that breadth is uneven; however, SPY remains positive over 30 days and low VIX levels do not yet establish a broad market breakdown. Reject the implication that broad liquidation is required today: JNJ, MSFT, MA, V, JPM and XOM show relative strength, while existing 25.8% cash already materially limits market exposure. Accept: concentration in AAPL, AMZN and JNJ is the central portfolio vulnerability. The appropriate response is to refrain from new concentration, retain cash, and reassess on confirmed deterioration rather than execute reactive sales after recent AAPL trade churn.

Market summary

The S&P 500 is up 3.5% over 30 days and the VIX is down 18.9%, which supports a constructive broad-market backdrop. However, the current session is risk-off in semiconductors and several portfolio holdings are materially lagging SPY over 30 days. Leadership is narrow: MSFT, JNJ, MA, V and XOM show stronger momentum, while AAPL, META, UNH and several chip-related watchlist names show weaker trends. This supports patience rather than deploying cash into either extended leaders or falling names.

Portfolio assessment

No transactions are warranted today. The portfolio is above $1.0M and retains meaningful unrealized gains in JNJ, AMZN, MSFT, MA, V and JPM, but remains concentrated in AAPL, AMZN and JNJ. AAPL and AMZN together represent roughly 38.7% of total portfolio value, while JNJ adds about 14.3%. Existing cash and prior reductions in volatile technology exposure are sufficient near-term risk controls; another immediate AAPL adjustment would risk repeating recent buy/trim churn. Core holdings should be retained while momentum and upcoming semiconductor earnings are monitored.

Risk assessment

The principal risk is concentration, not a lack of market exposure. AAPL, AMZN and JNJ account for approximately 71% of invested assets, and technology plus consumer-discretionary exposure remains dominant. AAPL's 30-day decline, META's 14.9% 30-day decline, UNH's 7.5% decline and uneven semiconductor momentum create downside risk if market leadership narrows further. MSFT's strong gain and JNJ's momentum are positive, but both are already meaningful positions and should not be chased. Cash reduces total-market sensitivity but cannot eliminate adverse moves in the largest holdings.

← August 21, 2026 August 25, 2026 →