BaroqTech

The trend filter lesson

One of the oldest published rules: stay in a market only while its price is above its average of the last ten months. Any extra profit it showed was within luck. What it did reliably was avoid most of the deepest falls.

Try it on the data

10 is the published setting.
Gold from 1975, stocks from 1900.

What happened

Gold, 1975-01 to 2026-09: the worst fall was -62% holding all the time and -33% with a 10-month filter.

  • Buy and hold
  • With the trend filter
Buy and hold against the trend filter
Buy and holdTrend filter
Average return a year6.4%8.0%
Worst fall from a high-62%-33%
$1 grew to$24.54$53.48
Time in the market100%58%

71 switches in or out, each charged 0.1%. Prices only: no dividends on stocks, no interest on cash, no taxes. Stock prices are monthly averages (Shiller), gold is a monthly price (datahub.io). Past prices; the future can differ.

Education only, not investment advice. The answer comes from the numbers you typed in, not from your account or the market. Simulated results are hypothetical and have limits: real trading has slippage, gaps, changing win rates and emotions. Trading leveraged products can lose money quickly, including all of it. Risk warning

Why this matters for a challenge

A challenge ends on the worst day, not on the average one. The same idea works at any scale: trade smaller, or not at all, when the market is against your setup or unusually wild, and normal size when it is calm and with you. Our own research tested this rule exactly as published (Faber, 2007) on gold since 1975 and US stocks since 1926: the return was similar to holding or a little higher, and the worst fall was roughly halved (gold about -29% instead of -62%, stocks about -46% instead of -85%). The extra return was within luck, so treat it as a brake, not an engine.

The other brake: size by volatility

Volatility sizing keeps the money at risk constant when the market speeds up: the stop is a multiple of the average true range (ATR), and the lot size shrinks when the ATR grows. Research on volatility-managed portfolios (Moreira and Muir, 2017) found smaller crashes; later checks found the extra return fragile, so again the benefit is mostly a smaller worst case. The position size calculator has a volatility stop: type in the ATR from your chart.

Not a trading signal and not advice. The test numbers come from Baroq Tech's own backtests of published rules, which did not pass its bar for trading money.