Strategies · Classic rules

Trend filter: the 200-day line

The oldest rule in the book: own stocks while they trade above their 200-day average, and step into Treasuries when they fall below it. It gives up some upside in exchange for sidestepping most long bear markets.

Pre-registered Sep 26, 2026, before it was tested. Results computed Sep 26, 2026.

Open in the builder

Hypothetical backtest; not investment advice; paper trading only.Past results, real or simulated, do not predict future ones. ARCANE places no live orders from this page.

The rule

Trades every trading day

While the S&P 500 (SPY) closes above its 200-day moving average, hold it; otherwise hold 7–10 year Treasuries (IEF). Checked daily.

Where the numbers came from. The 200-day line is the convention (Faber, 2007, and Composer’s community); nothing was tuned.

Hold what is below
If the price of SPY is greater than the 200-day moving average of SPY
then: Hold SPY
Otherwise
Hold IEF

As text

symphony "Trend filter: the 200-day line"
description "Stocks above the 200-day average, Treasuries below it."
rebalance daily
benchmark SPY
weight equal
  if current-price(SPY) > moving-average-price(SPY, 200)
    asset SPY
  else
    asset IEF

How it would have done

Oct 2, 2003 to Sep 25, 2026, against SPY

The whole history

149 trades

A year
+8.2%
Annualized return
In all
+509%
Oct 2, 2003 to Sep 25, 2026
Sharpe
0.71
Return per unit of risk
Worst fall
−31%
Max drawdown
Swings
12%
Volatility, a year
Beta
0.33
To the benchmark

Each decision reads closing prices up to that day and trades at the next day’s close. It trades toward its targets every day. Costs: 5 bp of slippage on every trade and 0.3 bp of fees on sales. Uninvested cash earns nothing. Hypothetical: no real orders were placed.

$0$5$10$15200420082012201620202024Held out

Growth of $1
SPY, held
Held-out period

−40%−20%0%

What it held: SPY 81% of the time; IEF 19% of the time.

In sample and the sealed holdout

The holdout was set before the preset was tested, and opened once

Measure	In sample Oct 2, 2003 – Sep 24, 2021	Sealed holdout Sep 27, 2021 – Sep 25, 2026
Annualized return	+8.5%	+7.0%
Sharpe ratio	0.73	0.60
Sortino ratio	1.01	0.85
Max drawdown	23.1%	30.6%
Calmar ratio	0.37	0.23
Volatility	12.2%	12.4%
Alpha against the benchmark	+5.4%	+1.0%
Beta	0.29	0.47
Correlation	0.46	0.65
Winning days	55%	53%
Turnover a year	6.6×	6.0×
Chance the Sharpe is above zero	100%	91%

Is it real?

Checked on the in-sample window only

Overfitting: Likely overfit
No clear edge over its benchmark

The result depends on the exact numbers chosen, or does not beat what luck would produce.

After counting 3 backtests of this idea, the out-of-sample Sharpe still clears the bar luck would set (deflated Sharpe 0.98).
The best version in one half of history usually lands in the bottom half of the other (probability of overfitting 77%).
Against SPY the difference (−3.0% a year) is within what luck produces (23% chance it truly beats it).

Backtests counted	3
Deflated Sharpe ratio (0.95 is the house bar)	0.98
Probability of backtest overfitting (20% is the house ceiling)	77%
Walk-forward Sharpe (choosing, then testing)	0.64
Walk-forward Sharpe with costs doubled	0.59
Extra return a year against SPY	−3.0%

The nudged versions (2)

Change	Sharpe
200-day moving average of SPY: window → 150	0.79
200-day moving average of SPY: window → 250	0.69

The house backtest gate

It does not pass. A strategy needs every one of these before ARCANE would paper-trade it.

Check	Value	Needs	Result
Evaluation ran without problems	clean	no problems	Passed
Report belongs to this pre-registration	trend-200-day	trend-200-day	Passed
Evaluated only inside the pre-registered data window (holdout untouched)	2003-10-02..2021-09-24	within 2003-10-01..2021-09-24	Passed
Walk-forward folds	14.000	>= 3	Passed
Out-of-sample observations	3528.000	>= 252	Passed
Deflated Sharpe ratio (3 trials counted)	0.985	>= 0.95	Passed
Probability of backtest overfitting	0.772	<= 0.2	Failed
Out-of-sample Sharpe (annual, after costs)	0.636	>= 0.5	Passed
Out-of-sample maximum drawdown	0.232	<= 0.25	Passed
Sharpe with doubled costs	0.587	> 0 (needs a turnover series)	Passed
Sealed holdout: positive and not clearly worse than the backtest	mean 0.000298, consistent	mean > 0 and consistent	Passed

The pre-registration

Written before the backtest; changing any of it makes a new variant

Claim: Holding SPY while it closes above its 200-day moving average and IEF otherwise beats SPY buy-and-hold on Sharpe after 5 bp costs over the sealed holdout.
Why the edge should exist: Trends persist because investors under-react to news and then chase; long declines give the average time to turn before the worst of them.
Where it comes from: behavioral
Who is on the other side: Buy-and-hold investors who ride drawdowns, and dip buyers early in long declines.
Why it is not arbitraged away: It is well known and has been crowded; it pays mainly in rare long bear markets and costs in choppy years, which career risk makes hard to hold.
Universe: IEF, SPY
In-sample window: Oct 1, 2003 to Sep 24, 2021
Sealed holdout: Sep 27, 2021 to Sep 25, 2026
Evaluation: walk-forward, on sharpe, against SPY buy and hold; 13 variants planned
Costs: 5 bp slippage a trade, no commission
Capacity: $5,000,000: SPY and IEF are among the deepest markets in the world.
What stops it: Paper Sharpe significantly below the backtest after 60 sessions.

The data

Prices: Yahoo Finance (unofficial chart API): adjusted daily closes for IEF, SPY. Unofficial and not licensed for redistribution. A licensed vendor (Alpaca, Tiingo or Polygon) is a founder decision before the builder is sold.