Bivar Capital — Quantitative Research

Momentum FactorSystematic strategy construction and out-of-sample analysis

A long-only momentum strategy on US large-cap equities with a profitability filter and market regime overlay. 254 months of out-of-sample data. Quarterly TTM fundamentals with availability lag. Full drawdown, rolling return, and crisis period analysis.

I. Strategy Specification

Universe

US domestic common stocks, market cap > $5B

Signal

9-month total return, skip most recent month (9-1)

Profitability filter

EBIT > 0 (trailing twelve months, 2-month availability lag)

Selection

Top 20 by signal, descending

Weighting

Equal weight (5% per position)

Rebalance

Last trading day of each month

Regime filter

S&P 500 close vs 200-day simple moving average

Regime rule

Above MA200 → fully invested / Below → 100% cash

Benchmark

S&P 500 Total Return

Period

January 2005 – March 2026 (254 months)

II. Summary Statistics

+16.5%

CAGR

+8.1%

Alpha

0.76

Sharpe

1.08

Sortino

-33.1%

Max Drawdown

17.9%

Volatility

Strategy	CAGR	Alpha	Sharpe	Sortino	MaxDD	Vol	Win	$100K →
12-1 Raw	+13.1%	+4.7%	0.37	0.50	-69.4%	27.3%	12/22	$1.36M
9-1 Raw	+14.2%	+5.8%	0.42	0.56	-65.0%	26.8%	13/22	$1.66M
9-1 + MA200	+15.1%	+6.7%	0.59	0.77	-47.9%	20.5%	13/22	$1.97M
9-1 + MA200 + EBIT	+16.5%	+8.1%	0.76	1.08	-33.1%	17.9%	14/22	$2.56M
9-1 + MA200 + EBIT (Q)	+15.6%	+7.2%	0.69	1.03	-35.6%	18.3%	12/22	$2.16M
9-1 + MA200 + EBIT (A)	+8.7%	+0.3%	0.34	0.48	-48.9%	16.5%	10/22	$581,494
S&P 500	+8.4%	Benchmark

III. Annual Returns

Year	Strategy	S&P 500	Alpha	Cumulative	S&P Cumul.
2005	+25.8%	+8.4%	+17.4%	$125,782	$108,365
2006	+5.7%	+12.4%	-6.7%	$132,937	$121,754
2007	+33.0%	-4.2%	+37.2%	$176,835	$116,701
2008	+0.0%	-40.1%	+40.1%	$176,835	$69,915
2009	+26.2%	+30.0%	-3.9%	$223,085	$90,908
2010	+25.5%	+19.8%	+5.7%	$279,977	$108,876
2011	-9.2%	+2.0%	-11.2%	$254,219	$111,102
2012	+26.2%	+14.1%	+12.1%	$320,877	$126,822
2013	+42.1%	+19.0%	+23.1%	$456,075	$150,905
2014	+13.2%	+11.9%	+1.3%	$516,161	$168,885
2015	+3.5%	-2.7%	+6.2%	$534,089	$164,250
2016	+18.1%	+17.5%	+0.6%	$630,659	$192,917
2017	+37.4%	+23.9%	+13.5%	$866,593	$239,049
2018	-8.4%	-4.2%	-4.1%	$794,070	$228,915
2019	+8.5%	+19.3%	-10.8%	$861,445	$273,055
2020	+44.2%	+15.2%	+29.1%	$1,242,454	$314,428
2021	+3.0%	+21.6%	-18.6%	$1,279,362	$382,262
2022	-11.3%	-9.7%	-1.5%	$1,135,263	$345,103
2023	+1.3%	+18.9%	-17.5%	$1,150,357	$410,207
2024	+73.3%	+24.7%	+48.7%	$1,993,702	$511,359
2025	+23.4%	+14.9%	+8.5%	$2,460,007	$587,421
2026	+4.0%	-5.9%	+9.9%	$2,557,284	$552,670

IV. Rolling Return Analysis

A. Rolling 3-Year Returns (annualized)

+15.0%

Mean

-3.0%

Worst

99%

% Positive

77%

% Beats S&P

End Month	Strategy	S&P 500	Alpha
2008-06	+18.0%	+2.4%	+15.6%
2008-12	+15.6%	-10.2%	+25.9%
2009-06	+12.5%	-10.2%	+22.8%
2009-12	+21.0%	-7.7%	+28.7%
2010-06	+17.7%	-11.8%	+29.5%
2010-12	+16.3%	-5.0%	+21.3%
2011-06	+16.4%	+1.0%	+15.4%
2011-12	+12.9%	+11.7%	+1.2%
2012-06	+14.6%	+14.0%	+0.6%
2012-12	+9.0%	+8.5%	+0.5%
2013-06	+14.2%	+15.9%	-1.8%
2013-12	+18.8%	+13.7%	+5.1%
2014-06	+21.0%	+14.1%	+6.9%
2014-12	+25.2%	+17.9%	+7.3%
2015-06	+28.5%	+14.8%	+13.7%
2015-12	+21.7%	+12.7%	+9.0%
2016-06	+15.5%	+9.3%	+6.2%
2016-12	+10.2%	+6.6%	+3.6%
2017-06	+10.6%	+7.3%	+3.3%
2017-12	+17.7%	+9.1%	+8.6%
2018-06	+16.8%	+9.6%	+7.2%
2018-12	+14.1%	+7.0%	+7.1%
2019-06	+13.6%	+11.9%	+1.7%
2019-12	+10.5%	+13.0%	-2.5%
2020-06	+8.6%	+8.6%	+0.1%
2020-12	+12.9%	+12.0%	+0.9%
2021-06	+16.0%	+16.5%	-0.5%
2021-12	+22.4%	+23.9%	-1.5%
2022-06	+15.7%	+8.8%	+7.0%
2022-12	+10.5%	+5.9%	+4.5%
2023-06	+10.3%	+12.8%	-2.5%
2023-12	-2.4%	+8.3%	-10.6%
2024-06	+3.8%	+8.3%	-4.5%
2024-12	+8.0%	+7.3%	+0.7%
2025-06	+11.6%	+17.9%	-6.4%
2025-12	+21.4%	+21.3%	+0.2%

B. Rolling 5-Year Returns (annualized)

+15.3%

Mean

+2.7%

Worst

100%

% Positive

88%

% Beats S&P

End Month	Strategy	S&P 500	Alpha
2010-06	+17.7%	-2.9%	+20.5%
2010-12	+19.5%	+0.1%	+19.3%
2011-06	+17.6%	+0.8%	+16.8%
2011-12	+14.5%	-2.4%	+16.9%
2012-06	+12.3%	-2.0%	+14.2%
2012-12	+10.9%	-0.6%	+11.5%
2013-06	+15.4%	+4.6%	+10.7%
2013-12	+21.4%	+15.4%	+6.0%
2014-06	+22.8%	+16.4%	+6.5%
2014-12	+16.9%	+13.0%	+3.8%
2015-06	+18.4%	+14.9%	+3.5%
2015-12	+13.9%	+10.2%	+3.7%
2016-06	+14.8%	+9.7%	+5.1%
2016-12	+19.7%	+12.2%	+7.4%
2017-06	+20.3%	+12.2%	+8.0%
2017-12	+22.4%	+13.4%	+9.0%
2018-06	+20.0%	+11.1%	+8.9%
2018-12	+11.2%	+6.3%	+5.0%
2019-06	+10.6%	+8.5%	+2.1%
2019-12	+11.0%	+9.4%	+1.6%
2020-06	+8.7%	+8.5%	+0.2%
2020-12	+17.0%	+12.9%	+4.0%
2021-06	+20.3%	+15.4%	+4.9%
2021-12	+18.5%	+16.3%	+2.1%
2022-06	+13.6%	+9.3%	+4.3%
2022-12	+6.9%	+7.5%	-0.6%
2023-06	+5.1%	+10.4%	-5.3%
2023-12	+6.5%	+13.7%	-7.2%
2024-06	+13.9%	+13.2%	+0.8%
2024-12	+16.8%	+12.7%	+4.1%
2025-06	+15.4%	+14.9%	+0.6%
2025-12	+11.7%	+12.8%	-1.1%

No 5-year window has ever produced a negative return. The worst 5-year annualized return was +2.7%. The strategy beat the S&P 500 in 88% of rolling 5-year periods.

V. Crisis Period Analysis

Crisis	Period	Strategy	S&P 500	Alpha	MA200 Cash
GFC (full)	2007-07–2009-03	+19.3%	-40.0%	+59.3%	17/21 mo
GFC (peak-trough)	2007-10–2009-02	-2.9%	-48.5%	+45.6%	16/17 mo
Flash Crash	2010-04–2010-08	-9.7%	-3.8%	-5.8%	4/5 mo
Debt Ceiling	2011-05–2011-10	-9.6%	-7.3%	-2.3%	3/6 mo
China/Oil	2015-06–2016-02	-5.4%	-0.2%	-5.3%	5/9 mo
Q4 2018	2018-09–2018-12	-19.2%	-7.2%	-12.0%	3/4 mo
COVID (3mo)	2020-01–2020-03	-4.9%	-9.7%	+4.8%	2/3 mo
COVID (6mo)	2020-01–2020-06	+7.5%	+1.4%	+6.1%	3/6 mo
Mom Crash 2021	2021-01–2021-12	+3.0%	+21.6%	-18.6%	No
2022 Bear	2022-01–2022-10	-1.0%	-9.6%	+8.6%	8/10 mo
2025 Tariffs	2025-01–2025-04	-14.4%	-2.1%	-12.3%	2/4 mo

The MA200 filter excels in prolonged downtrends (GFC, 2022 bear) but cannot protect against factor rotations where the broad market rises while momentum underperforms (2021). Fast crashes (Q4 2018) occur before the filter triggers. These are known structural limitations.

VI. Drawdown Analysis

All drawdowns exceeding −10%

Start	Trough	Recovery	Depth	Decline	Recovery
2005-02	2005-03	2005-05	-11.2%	1mo	2mo
2006-04	2006-07	2006-12	-12.6%	3mo	5mo
2011-04	2011-07	2012-08	-12.1%	3mo	13mo
2018-09	2019-08	2020-07	-21.6%	11mo	11mo
2021-11	2023-09	2024-02	-33.1%	22mo	5mo
2024-11	2025-02	2025-12	-19.0%	3mo	10mo

The worst drawdown (−33.1% for the deepest) was driven by the 2021 momentum crash followed by the 2022 bear market. The regime filter mitigated the market-wide decline but could not prevent the momentum-specific losses.

VII. Regime Filter Impact

Growth of $100,000 — with and without MA200

The MA200 filter reduces maximum drawdown from -65.0% to -33.1% while maintaining comparable CAGR. The cost is delayed re-entry after market bottoms, visible in the 2009 and 2023 underperformance streaks.

VIII. Rebalance Frequency Sensitivity

Frequency	CAGR	Alpha	Sharpe	MaxDD	Vol	$100K →
Monthly	+16.5%	+8.1%	0.76	-33.1%	17.9%	$2.56M
Quarterly	+15.6%	+7.2%	0.69	-35.6%	18.3%	$2.16M
Annual	+8.7%	+0.3%	0.34	-48.9%	16.5%	$581,494

Annual rebalance reduces CAGR by 7.9pp, eliminating nearly all alpha. Momentum is a fast-decaying signal. Monthly rebalance is required to capture the premium before it fades. Quarterly rebalance is an acceptable compromise with ~1pp CAGR drag and lower transaction costs.

IX. Rejected Modifications

Modification	Sharpe Δ	Reason
Inverse volatility weighting	−0.02	Underweights highest-momentum names
Momentum threshold > 5%	+0.00	Redundant in >$5B universe
ROIC > median filter	−0.06	Removes capital-intensive momentum winners; unstable across sub-periods
ROIC > 75th percentile	−0.04	Sharpe drops from 1.11 to 0.43 between sub-periods
Top 10 concentration	−0.06	Higher volatility without proportional return increase
Top 30 diversification	−0.03	Dilutes momentum signal
12-month lookback (with MA200)	−0.12	Stale signal, higher drawdowns
6-month lookback (with MA200)	−0.04	Signal too noisy for regime overlay

X. Known Risks and Limitations

Momentum crashes

Factor rotations where momentum underperforms the broad market (2021: +3% vs S&P +22%). The MA200 filter cannot protect against this — it only triggers on market-wide downtrends.

Sector concentration

Momentum naturally concentrates in leading sectors. In 2024–2025, this meant heavy AI/tech exposure. A sector shock would be amplified.

Transaction costs

Not modelled. Monthly rebalance of 20 large-cap stocks ≈ 5–10 trades/month. Estimated annual drag: 1–2%.

Taxes

Monthly rebalance generates short-term capital gains. After-tax CAGR may be 3–5pp lower.

Regime filter lag

MA200 is late at turning points. Fast crashes (Q4 2018) occur before the filter triggers. Recoveries (2009, 2023) see delayed re-entry.

Capacity

20 stocks >$5B investable up to ~$50M without significant market impact.

Recent outliers

2024 (+73%) is an extreme outlier. Do not extrapolate.

Survivorship

Sharadar includes 12,151 delisted stocks. Delisted names with incomplete price data may introduce residual bias.

XI. Methodology

Data source

Sharadar / Nasdaq Data Link

Fundamentals

Quarterly TTM (sf1_quarterly_ttm.parquet)

Availability lag

avail_ym field: 2 months after quarter end

Prices

Daily adjusted close incl. dividends

MA200

200 trading day simple moving average on S&P 500 close

Universe

15,456 tickers incl. 12,151 delisted

Benchmark

S&P 500 Total Return

Risk-free rate

3% (Sharpe, Sortino)

Delisted treatment

−100% return if no forward price

Constraints

Long only. No leverage. No derivatives.