W Analytics track record

How the Net Bias call has performed

Net Bias has run live in the W Analytics Terminal since June 2, 2026. For the five years before that, we replayed every session back to June 1, 2021 as of that morning’s open. Compared with simply holding the S&P 500, its biggest advantage has been risk: more return for each unit of risk taken, and a much shallower worst drop.

Five years walk-forward tested. Live since June 2, 2026. Long-only, no look-ahead, before trading costs. How to read this record.

Total return, June 2021 – September 2026

+91.1% vs +82.8% for the S&P 500

$10,000 grew to $19,108

Return per unit of risk (Sharpe ratio)

1.05 vs 0.79 for the S&P 500

32% higher

Worst drawdown

−15.2% vs −26.7% for the S&P 500

43% shallower

June 2021 – September 2026 ($10,000 grew to $19,108 vs $18,283 for the S&P 500), with 22% lower volatility and a higher annualized return (13.0% vs 12.0%).

Growth of $10,000 and drawdowns: Net Bias strategy vs. S&P 500, June 2021 to September 2026 Five-year walk-forward replay, live since June 2, 2026. Net Bias grew $10,000 to $19,108 with a worst drawdown of -15.2%. Buying and holding the S&P 500 grew it to $18,283 with a worst drawdown of -26.7%. Live$8k$10k$12k$14k$16k$18k$20k0%-10%-20%-30%20222023202420252026 Growth of $10,000 Drawdown from prior high S&P 500 worst -26.7%Net Bias worst -15.2%
Net Bias strategy: $19,108 S&P 500 buy-and-hold: $18,283 Weekly closes, starting from $10,000

Live since June 2, 2026

80 sessions, June 2 – September 25, 2026Net BiasS&P 500
Return+3.3%+1.5%
Worst drawdown-4.2%-4.2%
Average exposure1.12x1.00x

These are the Net Bias signals the terminal published in real time, applied to the S&P 500 index. Four months is a short stretch, so treat it as a check that the live signal behaves like the replay, not as a forecast.

The full record

Over the full period, Net Bias earned 32% more return per unit of risk than the index, and its worst drop was 43% shallower, while still finishing slightly ahead.

June 2021 – September 2026Net BiasS&P 500
$10,000 grew to$19,108$18,283
Total return+91.1%+82.8%
Annualized return13.0%12.0%
Annualized volatility12.4%15.9%
Return per unit of risk (Sharpe, 0% cash rate)1.050.79
Worst drawdown-15.2% Mar 2023-26.7% Oct 2022

Year by year

YearNet BiasS&P 500Difference
2021 (from Jun 1)+12.2%+13.3%-1.2 pts
2022-14.1%-19.4%+5.3 pts
2023+19.8%+23.1%-3.3 pts
2024+20.9%+24.4%-3.5 pts
2025+21.5%+16.5%+5.0 pts
2026 (to Sep 25)+12.7%+12.1%+0.6 pts

Net Bias trailed the index in 2021, 2023 and 2024. Those were steady up years, when holding a full position paid best. It lost less in 2022’s decline and led in 2025 and 2026 so far.

How the record is built

  • As of each open. The Net Bias score and stance are computed only from data available before that session’s open, then held until the next open. Nothing from later in the day or later in history is used.
  • Exposure, not direction. The stance sets a position between 0x and 1.6x the S&P 500, adjusted for trend and volatility conditions. There are no short positions.
  • Returns before costs. Results are measured open to open on the S&P 500 index from a $10,000 start, with no trading costs, slippage, financing for exposure above 1x, dividends or taxes.
  • Replay, then live. Sessions through June 1, 2026 are a walk-forward replay of the same rules. Since June 2, 2026, each session uses the signal the terminal published that morning.
  • Every session counts. No days, trades or years are removed, including the ones where the strategy lagged.

What the record shows

The edge has come from sizing, not from predicting tomorrow’s direction. Average exposure was 0.78x the index, and the strategy sat fully in cash for 214 sessions (16%). A bearish stance means “take less risk,” not “the market will fall tomorrow.” The S&P 500 still rose on more than half of bearish-stance days, and the benefit came from being smaller when risk was elevated.

See today’s Net Bias read, stance and exposure in the terminal.

Start 30-day free trial

How to read this record

This record combines two kinds of evidence. The first is a five-year walk-forward replay. It runs the same rules session by session, using only what was known at each morning’s open, so each of the 1,256 sessions is scored the way it would have been called at the time. Nothing is left out. The second is a live track record: since June 2, 2026, the terminal has published the Net Bias call every morning, and those results are shown separately above so you can compare the two.

The replay is a simulation built after the fact, so the rules had the benefit of hindsight in their design. Both the replay and the live results apply the Net Bias call to the S&P 500 index rather than to a brokerage account. They are shown before trading costs, slippage, financing, dividends and taxes.

Past results, whether simulated or live, don’t guarantee future performance. The W Analytics Terminal is a research and decision-support tool, not personalized investment advice. Markets involve risk of loss.