Monthly Review — February 2026

Abstract

February 2026 was dominated by the “software-mageddon” selloff — AI disruption fears hammered the software sector after Microsoft’s historic post-earnings plunge and Anthropic’s legal AI tool announcement. I responded with a deliberate valuation rotation: the S&P 500 trailing P/E sat at ~29x (historical median ~18x), making cap-weighted VOO expensive beta rather than safe beta. I sold VOO (7→2) and VEU (limited upside), while keeping RSP (equal-weight, more reasonable P/E) and SCHG (small position, not worth disrupting), and rotated proceeds into MSFT and SPGI during the sector selloff. Smaller positions (ASIC, NFLX) were swapped into the higher-conviction SPGI — not cut as losers. PYPL was a genuine capitulation (−USD 614 realized) from a broken thesis. The result: Core ETFs dropped from 28% to 18% and Stocks rose to 63%, exceeding Playbook bands — but this was a conscious trade-off on valuation grounds, not a panic-driven accident. Time-weighted return was +1.88%. The key question going forward: restore structural balance via cash flows, and trim gold on the Iran spike.

TL;DR

• TWR: +1.88% — outperformed SPX (−0.4%) and NDX (−2.5%), driven by gold and defensives.
• Deliberate valuation rotation: sold VOO (SPX PE ~29x, at historical ceiling) and VEU to fund MSFT/SPGI at software-mageddon discounts. Retained RSP (equal-weight PE ~20x, better risk/reward).
• Stocks surged to 63% (band: 20–50%) and Core ETFs dropped to 18% (band: 35–60%). Conscious trade-off, but the magnitude of the band violation needs correction.
• Position cleanup: ASIC and NFLX swapped into SPGI (not “loser cuts”); PYPL capitulated at −USD 614 (genuine sizing/entry mistake from the past).
• GLDM quietly carried the month (+USD 569 holding MTM), and is now positioned for further upside from the Iran crisis.
• PGR (+USD 170 this month) has not reached target price — holding.

1) What happened (context + market environment)

1.1 The software selloff (“software-mageddon”)

The dominant market story in February was the AI-driven software selloff. It began after Microsoft reported strong Q2 FY2026 earnings (revenue +17% to USD 81.3B) but the stock fell 10% — its worst post-earnings drop in over a decade — on concerns about Azure growth deceleration and massive AI capex (USD 37.5B in the quarter). The S&P 500 software and services index lost roughly USD 1 trillion in market value from January 28 through early February.

The selloff deepened after Anthropic released a legal AI tool that spooked the legal-software sector (Thomson Reuters −28% YTD at the trough), and spread to the broader software ecosystem. Stifel downgraded MSFT to Hold with a target cut from USD 540 to USD 392. The iShares Expanded Tech-Software ETF fell over 13% from January 28.

Multiple analysts noted the selloff appeared indiscriminate. J.P. Morgan called it “broken logic,” and Morningstar’s Dan Romanoff said fundamentals remained solid and fears were overblown. Several commentators observed that stocks with strong competitive moats and sticky products were being thrown out with the bathwater.

1.2 The valuation case for rotation

The S&P 500 trailing P/E ratio was ~29x at the start of February — the historical median is ~18x, and the 20-year average is only ~16x. This meant cap-weighted VOO was offering expensive beta, not safe beta. RSP (equal-weight, PE ~20x) was more reasonably valued because it avoids the mega-cap tech concentration that inflates the cap-weighted P/E.

This created an unusual situation: the “safe” index was arguably more expensive than individual quality names that were being sold off. MSFT at USD 400–411 (down from ~USD 480 pre-earnings) and SPGI at USD 365–460 offered better risk-adjusted entry points than holding VOO at 29x earnings.

1.3 Broader market context

The S&P 500 finished February slightly negative (−0.43% MTD). The Nasdaq posted its worst month since March. Under the surface, a major rotation was playing out: energy, materials, and consumer staples led, while tech lagged. Gold rose on tariff uncertainty (Trump’s 15% global tariffs) and mounting geopolitical tension. On Feb 28, the U.S. and Israel launched strikes on Iran, dramatically escalating the situation into March.

1.4 My mental state

I was calmer this month than in January. The software selloff created excitement, but my actions were more deliberate: I had a clear thesis (expensive beta → discounted quality), and the execution was staged over multiple days. The main emotional pressure was from PYPL — deciding to realize a large loss is never comfortable, but it was the right call.

2) Performance & attribution

2.1 Time-weighted performance

• MTD TWR: +1.88%
• Benchmarks (approximate Feb): SPX −0.4%, NDX −2.5%, Dow +1.9%

The portfolio outperformed the S&P 500 mainly because of gold and defensive holdings (KO, KR, PGR rebound), which offset losses in MSFT and the newly opened tech positions.

2.2 Net asset value bridge

Deposits were modest (USD 1,278 vs. USD 16,791 in January). The mark-to-market gain of USD 811 is the “real” investment performance.

2.3 Attribution by ticker (MTM total, Feb)

Winners:

• GLDM: +USD 569 (holding) — gold continued its steady climb; now poised for further gains from Iran crisis
• KO: +USD 300 (holding +USD 287, trade +USD 13) — defensive rotation beneficiary; trimmed 20 shares for +USD 179 realized
• KR: +USD 270 — staples winner in the rotation
• SPGI: +USD 231 (trade +USD 25, holding +USD 208) — entered during the selloff; rebounded nicely by month-end
• PGR: +USD 170 — recovery after months of grinding; not yet at target price
• RSP: +USD 168 — equal-weight outperformed cap-weight this month (validating the decision to keep RSP over VOO)
• TLN: +USD 113 — energy/power play continued working

Losers:

• MSFT: −USD 280 (holding −USD 247, trade −USD 32) — the dip kept dipping after my entries
• PYPL: −USD 196 net (realized −USD 614, offset by MTM +USD 79 on day of sale)
• AXP: −USD 173 — cyclical weakness
• ACN: −USD 139 — entered at USD 243.50, caught in the software downdraft
• SCHW: −USD 108 — financials weak
• SCHG: −USD 93 — growth ETF suffering from tech rotation

2.4 Realized P&L

Context matters: excluding PYPL (a genuine past mistake), the other trades netted +USD 252. ASIC (+USD 28) and NFLX (−USD 2) were approximately breakeven swaps — the purpose was to consolidate capital into SPGI, not to “cut losses.”

3) Trades & actions

3.1 The valuation rotation: VOO/VEU → MSFT/SPGI

The thesis: SPX PE ~29x is historically expensive. Rather than holding expensive beta through VOO, rotate into quality individual names that the software selloff has made significantly cheaper.

Sold:

• VOO: 5 shares sold (7→2) — SPX PE ~29x; retained 2 shares as residual core
• VEU: all 16 shares sold — limited upside vs. software-mageddon opportunities

Kept:

• RSP: all 24.3 shares — equal-weight PE ~20x, more reasonable valuation
• SCHG: all 73.7 shares — position too small to justify trading cost/disruption

Bought:

• SPGI: 11 shares across 6 trades (USD 460→USD 365, Feb 5–10)
• MSFT: 5 shares added (6→11, avg ~USD 404)
• ACN: 4 shares @ USD 243.50 — new position
• AMZN: 10 shares @ USD 210 — new position

3.2 Position swaps: ASIC + NFLX → SPGI

These were not “loser cuts.” Both positions were approximately breakeven:

• ASIC: 50 shares sold @ USD 18.30 (+USD 28 realized) — the small-cap experiment had served its purpose
• NFLX: 10 shares sold @ USD 84 (−USD 2 realized) — originally a starter position; I saw better risk/reward in SPGI

The capital freed (~USD 1,755) went directly toward SPGI entries.

3.3 PYPL capitulation: the real mistake (from the past)

PYPL was sold at USD 44.87, realizing −USD 614. This was not a February mistake — it was closing a position that should never have been sized that large. The original entry (~USD 52.69 cost basis) lacked a clear thesis, and the stock ground lower for months. The February exit was the right decision; the original entry was the error.

3.4 Other trims

• KO: 20 shares sold @ USD 78 (+USD 179 realized) — profitable trim after strong defensive run
• IBKR: 4 shares sold @ USD 73 (+USD 87 realized) — trim
• SCHW: 5 shares added @ USD 95 — modest add

3.5 SPGI: the centerpiece entry

Average cost: ~USD 421/share. Close at month-end: USD 441.88 (+5.0%).

The entry was price-staged across a wide range (USD 460→USD 365), which is good execution. The total position size (USD 4,630, 10.2%) is large for a new position, and the speed (6 trades in 5 days) was faster than ideal — future staged entries should have more time gaps between steps.

4) Structural analysis

4.1 Sleeve allocation (Jan → Feb)

4.2 Honest assessment of the band violations

The Playbook bands are violated. The question is whether the bands need updating or whether I need to rebalance back.

The case for “this was correct”:

• VOO at SPX PE ~29x is not “safe core” — it’s expensive beta
• The software selloff created genuine value in quality names
• RSP was kept as the core position with better valuation
• The rotation should be judged over 6–12 months, not at month-end

The case for “this was still a violation”:

• The Playbook bands exist precisely to prevent smart-sounding rationalizations
• Every opportunity looks “unique” in the moment
• 63% in individual stocks requires my stock-picking to be consistently good — and I have ~6 months of experience
• The bands should be fixed via cash flows, not ignored

My conclusion: The rotation thesis was defensible, but the magnitude was too aggressive. I should have sold less VOO (e.g., keep 4–5 shares instead of 2) and bought less SPGI (e.g., 7 shares instead of 11). The correct fix now is cash-flow rebalancing + gold trimming, not panic reversal.

4.3 Top positions (concentration)

Software-adjacent exposure (MSFT + SPGI + ACN) = 21.1%. This is an implicit sector bet that I should monitor, even if each position was entered with its own thesis.

5) What went wrong

5.1 Magnitude of the rotation

The rotation direction was correct (expensive beta → discounted quality), but the size was too aggressive. Selling 5 of 7 VOO shares + all VEU in a single month moved too much capital at once. A more measured approach: sell 3 VOO, keep VEU, build SPGI and MSFT more slowly.

5.2 SPGI entry speed

6 trades in 5 days, including 3 on Feb 10 alone. The price-staging was good (USD 460→USD 365), but the time-staging was insufficient. Future staged entries should have a minimum gap of ~5 trading days between steps.

5.3 PYPL: the original sin

The February exit was correct. The mistake was the original entry: insufficient thesis, too large a position, and holding too long while the thesis deteriorated. The −USD 614 loss is the cost of that lesson.

6) What went right

6.1 The rotation thesis

Selling expensive beta (SPX PE ~29x) to buy discounted quality names during a sector-specific selloff is a defensible active management decision. The key names (MSFT, SPGI) are wide-moat companies experiencing a sentiment-driven selloff, not a fundamental breakdown. Multiple analysts (Morningstar, J.P. Morgan) noted the same opportunity.

6.2 Keeping RSP over VOO

Equal-weight outperformed cap-weight in February. RSP (+USD 168) was the best-performing core position. This validates the thesis that cap-weighted SPX is distorted by mega-cap tech valuations.

6.3 Defensive holdings delivered

KO (+USD 300), KR (+USD 270), PGR (+USD 170) all contributed positively. PGR has not reached target price — holding.

6.4 GLDM as the anchor

Gold’s +USD 569 contribution was the single largest positive attribution factor. With the Iran crisis escalating, GLDM is positioned for further gains — and provides the opportunity to trim on strength and redirect to Core.

6.5 Willingness to take a loss (PYPL)

Realizing −USD 614 on PYPL was painful but necessary. Improvement from January: I now cut positions with broken theses instead of holding “because it’s already down.”

7) Lessons learned

7.1 Valuation rotation is valid, but magnitude matters

The direction was right; the size was too large. In the future: rotate gradually, keep more of the core intact, and accept that I might participate less in an opportunity.

7.2 “Staged entry” needs a time dimension

Price-staging without time-staging is not discipline. New rule: minimum 5 trading days between adds.

7.3 Sector concentration builds invisibly

MSFT + SPGI + ACN = 21% in software/data names. Each position had its own thesis, but together they create a sector bet. New monitoring: check total sector exposure before any add.

7.4 Capitulation is expensive but educational

PYPL taught me: if a position lacks a clear thesis and keeps grinding, the sizing was wrong from the start. Cut before it becomes emotionally sticky.

8) Rule updates (process changes after this month)

8.1 Valuation-based rotation clause (new)

Selling Core ETFs is allowed if the valuation case is explicit and written:

• Must state the valuation metric (e.g., “SPX PE ~29x, well above 20-year avg of ~16x”).
• Must identify what is being bought and why it offers better risk/reward.
• Must keep at least one core ETF at ≥10% weight (currently RSP fills this role).
• The rotation must be gradual (spread over at least 2–3 weeks).

This replaces a blanket “never sell Core” rule with a conditional rule that requires explicit justification.

8.2 Time-staged entries (minimum gap)

Staged entries must have at least 5 trading days between steps. Same-day or next-day adds are separate lots but not separate “steps.”

8.3 Sector concentration monitoring

Before adding to any position, check total GICS sector exposure. Soft cap: 25% in any single sector. If above 25%, no further adds until it drops below.

8.4 Rebalance plan for March

Priority 1: All new cash → Core ETFs (RSP or VOO if SPX PE improves) until Core ≥ 25%. Priority 2: Trim gold (GLDM) on the Iran spike — take partial profits while gold is elevated, redirect proceeds to Core ETFs. This executes the “partial profit rule” from the January lesson. Priority 3: Do not add to any stock position until the stock sleeve drops below 55%. PGR: Hold — not yet at target price. Continue monitoring.

9) Forward look: March and the Iran situation

9.1 Geopolitical shock (Feb 28 – March)

On February 28, the U.S. and Israel launched coordinated strikes on Iran, killing Supreme Leader Khamenei. Iran retaliated with strikes across the Middle East, including Dubai and other Gulf states. As of March 2, the situation remains highly fluid.

Immediate market implications:

• Oil: Brent crude up ~6% to USD 77 in early Monday trading. If the Strait of Hormuz is disrupted, oil could spike above USD 100.
• Gold: Spot gold surging above USD 5,300 (up >1%). J.P. Morgan forecasts USD 6,300/oz by year-end.
• Equities: Risk-off pressure likely, especially on cyclicals and growth.

9.2 Portfolio implications and planned actions

GLDM (15.3%) — plan to trim: Gold has spiked on the Iran crisis. This is the opportunity to execute the “partial profit rule” — sell a portion (e.g., 10–15 shares) while gold is elevated. The proceeds go to Core ETFs. This simultaneously:

• locks in some of the +USD 1,291 unrealized gain,
• reduces gold below the 15% band ceiling,
• and rebuilds Core toward the 35% floor.

TLN (3.9%): Energy/power benefits from oil price increases. Hold. KO, KR (13.2% combined): Consumer staples are defensive in geopolitical crises. Hold. PGR (13.4%): Insurance is relatively insulated from Middle East risk. Not yet at target price — hold.

MSFT, SPGI, ACN (21.1% combined): Already under pressure from the software selloff. A risk-off environment could extend the downturn, but these are long-term holds. The thesis is valuation-driven, not timing-driven. Do not panic-sell.

9.3 What I should do (and not do)

Do:

• Trim gold on strength → redirect to Core ETFs.
• Use any new cash for Core ETFs.
• Stay calm — the portfolio has natural hedges (gold, staples, insurance).

Do not:

• Do not try to “trade” the Iran situation.
• Do not sell software positions into a geopolitical panic unrelated to their fundamentals.
• Do not abandon the rebalancing plan because “the market changed.”

9.4 Scenario thinking

In all three scenarios, the action converges: trim gold, rebuild core, hold stocks.

10) End-of-month portfolio snapshot

Portfolio value: USD 47,588.31 Market value: USD 45,867.88 Cash: USD 1,685.59 (3.5%) Unrealized P&L: +USD 910.45 Realized P&L (Feb): −USD 362.59

Holdings