Starting Index
transparent model intelligence
Account
Evidence

Backtested performance with the work shown.

The live model family is published forward from its public inception. This page shows the research backtest behind it: point-in-time constituents, SPY comparison, validation sweeps, and the same annual return view used by the research workflow to inspect behavior.

Start Free AccessSee Method
Backtest WindowMay 1, 1996-Jun 24, 2026

Research data only. Subscriber allocations are published separately. The live platform record and profile record are tracked forward from publication records.

Research Equity Curve
Core, Balanced, and Growth versus SPY. Annual bars follow the selected model
Core
Balanced
Growth
SPY
$0$5.5M$11M$16M$22M200020052010201520202025Growth of $10,000 ยท actual scaleCoreBalancedGrowthAnnual return ยท Balanced+239%-239%1997 Balanced +24.4%, excess -6.7%1997 SPY +31.1%, excess -6.7%1998 Balanced +31.9%, excess +5.6%1998 SPY +26.4%, excess +5.6%1999 Balanced +91.4%, excess +72.0%1999 SPY +19.4%, excess +72.0%2000 Balanced +97.1%, excess +106.9%2000 SPY -9.8%, excess +106.9%2001 Balanced +3.9%, excess +15.2%2001 SPY -11.3%, excess +15.2%2002 Balanced +4.0%, excess +27.6%2002 SPY -23.6%, excess +27.6%2003 Balanced +59.0%, excess +36.8%2003 SPY +22.2%, excess +36.8%2004 Balanced +34.9%, excess +26.2%2004 SPY +8.7%, excess +26.2%2005 Balanced +9.9%, excess +6.4%2005 SPY +3.5%, excess +6.4%2006 Balanced -3.2%, excess -15.0%2006 SPY +11.8%, excess -15.0%2007 Balanced +24.8%, excess +21.4%2007 SPY +3.4%, excess +21.4%2008 Balanced -17.7%, excess +20.0%2008 SPY -37.7%, excess +20.0%2009 Balanced +54.5%, excess +34.6%2009 SPY +19.9%, excess +34.6%2010 Balanced +0.3%, excess -10.6%2010 SPY +11.0%, excess -10.6%2011 Balanced -5.3%, excess -4.1%2011 SPY -1.2%, excess -4.1%2012 Balanced +29.1%, excess +17.4%2012 SPY +11.7%, excess +17.4%2013 Balanced +48.6%, excess +22.1%2013 SPY +26.4%, excess +22.1%2014 Balanced +21.4%, excess +9.1%2014 SPY +12.4%, excess +9.1%2015 Balanced +3.0%, excess +3.8%2015 SPY -0.8%, excess +3.8%2016 Balanced +30.0%, excess +18.8%2016 SPY +11.2%, excess +18.8%2017 Balanced +21.6%, excess +3.2%2017 SPY +18.5%, excess +3.2%2018 Balanced -26.6%, excess -19.6%2018 SPY -7.0%, excess -19.6%2019 Balanced +22.1%, excess -6.5%2019 SPY +28.7%, excess -6.5%2020 Balanced +22.2%, excess +7.1%2020 SPY +15.1%, excess +7.1%2021 Balanced +12.8%, excess -16.0%2021 SPY +28.8%, excess -16.0%2022 Balanced +12.9%, excess +32.8%2022 SPY -19.9%, excess +32.8%2023 Balanced +27.3%, excess +2.5%2023 SPY +24.8%, excess +2.5%2024 Balanced +26.3%, excess +2.3%2024 SPY +24.0%, excess +2.3%2025 Balanced +73.5%, excess +56.9%2025 SPY +16.6%, excess +56.9%2026 Balanced +168.3%, excess +161.0%2026 SPY +7.3%, excess +161.0%
Validation
Balanced evidence checks from the published profile-specific validation artifact
Parameter Tests99

Checks whether results still hold when the model settings are nudged nearby.

Walk-Forward Rows594

Tests the profile on later periods after each setup window, closer to real-world use.

Stress Windows495

Looks at difficult market periods so risk is visible, not hidden by long-term averages.

Parameter Median22.70%

The middle result across setting changes. Higher suggests the edge is less fragile.

Walk-Forward Median20.09%

The middle forward-test result. This helps judge whether performance persisted.

Stress Median-1.22%

Middle hard-market result versus SPY at -31.54%. Less negative means better defense.

Summary Read
Balanced quick read from the research record
Return Engine26.02% CAGR

Growth multiple: Balanced 1065.2x ยท SPY 11.2x.

Risk Shape12.42%

Difference between SPY's worst drawdown and Balanced's worst drawdown. Positive is better.

Sharpe Ratio0.92

Balanced's risk-adjusted return over the full research window. Higher is better.

Win Rate77%

Balanced outperformed SPY in 23 years out of 30 complete overlapping years.

Research record uses point-in-time S&P 500 membership at each decision date. 17.6% of days were handled defensively.

Research Standard
What the backtest is designed to protect against
  • Point-in-time S&P 500 membership, not today's constituents projected backward.
  • Decisions use data available at the rebalance date.
  • Delisted and replaced symbols are included when historical data is available.
  • SPY is used as the broad-market benchmark for the same dates.
  • The ranking formula is proprietary. The testing standard is intentionally visible.
Known Limits
The parts a serious investor should still care about
  • Backtests are historical research, not live account results.
  • Historical symbol coverage is strong, but not perfect: active membership coverage is 93.32%.
  • Drawdowns are real and can be emotionally difficult even when long-term returns are strong.
  • Future market structure can differ from the research period.
  • Live publication and audit records are the source of truth once a model period has been published.

Research evidence is provided for due diligence and context. It is not investment advice and does not guarantee future results.