3. Two-Strategy Investment Framework

Framework revision (May 2026): Account assignment was updated for tax optimization — IRA aggressive momentum, Brokerage quality buy-and-hold. See journal entry 2026-W20 | May 17, 2026 — Strategic Framework Revision. Pipeline mechanics below remain accurate; execution policy follows the revised subsections.

3.1 IRA — Aggressive Momentum Strategy (Revised May 2026)¶

Revised rationale — tax optimization: The IRA’s tax-deferred status makes it the OPTIMAL account for aggressive momentum investing.

• No capital gains tax on any trade

• 100% efficient compounding — every gain reinvested

• No wash sale rule — sell and rebuy freely

• Free rotation into strongest IBD signals weekly

• Higher beta acceptable — volatility has no tax penalty

Original rationale was partially wrong: Original logic: protect retirement capital = low beta Revised logic: losses in IRA give no tax deduction anyway gains compound 100% tax-free higher beta = higher compounding potential 8% sell rule provides the real protection

Revised parameters: Universe: IBD 50 + BC20 (highest momentum names) Composite: 90+ preferred RS Rating: 85+ minimum Beta limit: 2.0 (removed 1.5 cap) Acc/Dis: B+ minimum Sell rule: 8% — apply actively, rotate freely Hold period: Days to weeks — follow IBD signals Goal: Maximum total return, tax-free compounding

Best names (examples): NVDA, MRVL, SITM, AMD, SIMO — leaders

Implementation (pipeline)¶

• Inputs: Big Cap 20 and Big Cap 50 IBD exports are merged; each ticker keeps the best (lowest) IBD_Rank and max Composite_Rating; Source_List records BC20 / IBD50 / both.

• RS filter: RS_Rating ≥ 83 (RS_LEADER_FLOOR).

• Primary ordering: Stocks on both lists (Multi_List) sort before others; then IBD_Rank ascending (IBD composite rank as the main quality sort).

• Risk-adjusted score: Risk_Adj_Score = Composite_Rating × Beta_Penalty (see §4).

• Quality spread: WACC_Spread = Annual_ROE/100 − WACC_Stock when ROE is available (IRA watchlist and IRA reports).

IRA Quality Universe — Berkshire Hathaway Cross-Reference¶

During strong uptrends IBD lists fill with high-beta momentum names that are inappropriate for the IRA quality mandate. In these periods use the Berkshire Hathaway 13F holdings as a supplementary universe for IRA candidates.

Why Berkshire works as an IRA filter: Berkshire’s portfolio represents Buffett’s decades of identifying businesses with durable competitive advantages, consistent earnings, low debt, and pricing power — exactly the profile the IRA quality strategy seeks. All 42 holdings are profitable, institutionally validated, and liquid.

The combined methodology: Buffett validates WHAT to own (quality filter) IBD confirms WHEN to buy (momentum/timing filter) Beta Penalty ensures appropriate IRA risk level

IRA-qualified Berkshire names (beta under 1.0): KO (0.58) — Consumer staples, dividend aristocrat KR (0.50) — Grocery, non-cyclical, very defensive UNH (0.60) — Healthcare, consistent growth CB (0.60) — Insurance, Buffett’s core competency VRSN (0.85) — Internet monopoly, pure recurring revenue CVX (0.85) — Energy, dividend aristocrat HEI (0.85) — Aerospace components, niche monopolies V (0.92) — Payment network, near-zero beta sensitivity DPZ (0.70) — Consumer discretionary, franchise model LAMR (0.95) — Outdoor advertising, real estate characteristics

Process for using Berkshire universe: Step 1: Check IBD Digital for Composite and RS ratings Need Composite 75+ and RS 70+ minimum Step 2: Check current price vs buy point on chart Must be in base or within 5% of buy point Step 3: Confirm volume — same two-step rule applies Step 4: Verify beta under 1.0 on IBD Stock Checkup Step 5: Check earnings date — not within 2 weeks

Update frequency: Quarterly — Berkshire files 13F within 45 days of each quarter end. Next filing: ~May 15, 2026 (Q1 2026 data).

Source: SEC 13F filings via gurufocus.com or sec.gov

Market Environment Caveat — Critical¶

Berkshire names score LOWER on IBD during strong uptrends because consumer staples and defensive names underperform momentum/growth stocks — their RS ratings drop.

IBD ratings for Berkshire names improve DURING:

• Market corrections (IBD 0%-40% exposure)

• Sector rotation to defensives

• Bear markets or consolidation periods

Best use of Berkshire universe: UPTREND (current): Research and watchlist only Build familiarity with the names Set price alerts at buy points

CORRECTION: Active buying opportunity RS ratings recover as tech sells off Defensive names become IBD leaders Deploy IRA cash here aggressively

Think of the Berkshire list as your CORRECTION SHOPPING LIST. When IBD downgrades to 0%-20% exposure and the market pulls back — these are the names to buy for the IRA while waiting for the next confirmed uptrend.

Example timing: W1-W6 of last correction (Feb-Apr 2026): KO, UNH, CB would have held value RS ratings would have been 70-85 range Acc/Dis would have been A or B (institutions rotating in) IBD would have flagged them as relative leaders

3.2 Brokerage — Quality Buy and Hold (Revised May 2026)¶

Revised rationale — minimize tax drag: The Brokerage account is subject to capital gains taxes.

• Short-term gains (under 12 months): ordinary income 22-32%

• Long-term gains (12+ months): preferential rate 15-20%

• Every short-term trade costs 22-32 cents per dollar

• Disposition effect here is especially costly (Barber-Odean)

Optimal strategy: Buy quality names with strong fundamentals Hold minimum 12 months for long-term gains rate Sell only on fundamental thesis breaks Normal volatility: hold through

Revised parameters: Universe: BC20 only (large cap quality) Composite: 90+ preferred RS Rating: 80+ minimum Beta limit: Under 1.0 preferred Acc/Dis: B+ minimum Hold period: 12+ months minimum Goal: Minimize tax events, long-term gains rate

Best names (examples): AGX, COKE, BKV, SNEX — low beta quality

Illustrative Brokerage hold policy: Quality leaders already owned under prior momentum logic are typically held toward a long-term capital-gains mark unless the thesis breaks. Live share counts and account balances are omitted from this public book.

Note: Revised policy is hold to long-term mark unless thesis breaks.

Implementation (pipeline)¶

• Input: Single RS Master export (same column map as other IBD screens) — data source for watchlist generation.

• RS filter: RS_Rating ≥ 83.

• Leader Score: Leader_Score = (RS × 0.6) + (EPS × 0.2) + (ACC_Numeric × 4) (see §4).

• Primary ordering: Positive weekly price movers first (Price_Change_Pct > 0), then Risk_Adj_Leader_Score descending.

• Risk-adjusted score: Risk_Adj_Leader_Score = Leader_Score × Beta_Penalty.

• Analyzer status: LEADER vs RISK flags in reports; execution follows buy-and-hold policy above, not weekly rotation.

3.3 Aggressive account — High-Beta Momentum + Margin¶

Account type: Individual Brokerage (taxable)
Status: Paper trading W20-W21 → live W22+
Capital: $8,000 initial |$1,500 minimum reserve

Correction from original plan:
Originally planned as Roth IRA. Roth IRA requires earned income (W-2/self-employment) as contribution source. Contribution-source constraints can force an Individual Brokerage opening instead of a Roth; treat account type as a tax/legal constraint, not a preference.

Strategy:

• Universe: RS Leaders + IBD 50 (highest momentum)

• Beta target: 2.0+ (intentional high beta)

• Beta penalty: NOT applied — beta is the feature not bug

• Sell rule: 8% — applied actively

• Hold target: 12+ months where possible (long-term rate)

• Margin: Phase 1 paper → Phase 2 real after 4-6 weeks

Position sizing — $8,000 capital:

• Minimum reserve: $1,500 (always maintained)

• Maximum deployable: $6,500

• Position size: $2,000-2,500 per trade

• Max positions: 3 simultaneously

Margin plan for this account:

• Phase 1 (W20-W21): Paper margin tracking only

  • Track hypothetical margin positions

  • Verify returns clear 12.5% hurdle rate

  • Build discipline with zero real risk

• Phase 2 (W22+): Real margin deployment

  • First margin position: $1,000-2,000

  • One position at a time initially

  • Must clear hurdle rate before adding

  • This account PRECEDES margin in Brokerage account

Why this account precedes Brokerage margin:

• Smaller capital = smaller margin mistake cost

• High-beta names = faster feedback on thesis

• Establishes margin discipline before larger account

• Lessons here protect the $28,000+ Brokerage account

Target names (W20 entry signals):

• SNDK: score 81, 3wk recurring, storage AI thesis

• MRAM: score 81, 3wk recurring, defense contract

• AMD: score 81, 4wk recurring, AI GPU thesis

• SITM: score 78, high beta 2.91, semiconductor timing

Three-account strategy summary:

• IRA account: High-beta momentum, tax-deferred compounding

• Brokerage account: Quality buy-hold, long-term gains rate

• Aggressive account: High-beta + margin, maximum return target

3.4 Academic Foundation — Behavioral Finance¶

Barber and Odean (1999) “The Courage of Misguided Convictions” Financial Analysts Journal Vol. 55 No. 6 pp. 41-55 DOI: 10.2469/faj.v55.n6.2313

Two systematic investor errors this methodology prevents:

1. Disposition Effect: investors hold losers, sell winners

   • Response: 8% sell rule (no discretion on losers)

   • Response: Power trend rule (no premature winner sales)

2. Overconfidence: active traders earn lowest returns

   • Response: 10-point checklist (replaces conviction)

   • Response: Volume 1.40x+ requirement (prevents impulse)

   • Response: 60-second checklist — five steps before price chart (§7.2)

Publication target: Financial Analysts Journal (same journal as Barber-Odean citation above).

See also §15.4 Behavioral Finance Layer, §15.5 Modern Portfolio Theory Framework, and §7.2 Entry Decision Checklist for operational rules.

3.5 Cross-strategy conviction¶

• Definition: Tickers appearing in both IRA_Watchlist.csv and Brokerage_Watchlist.csv.

• Output: Cross_Strategy_Conviction.csv merges IRA columns (e.g. composite, IRA_Risk_Adj_Score) with Brokerage columns (Leader_Score, Brok_Risk_Adj_Score, etc.), sorted by conviction quality.

• Why it matters: Surviving both the large-cap composite filter and the RS/leader filter is a double confirmation—useful for sizing, options overlays, or a concentrated “conviction 9” sleeve in performance_tracker.py.