Should you sell when the market drops below its 200-day average?
A moving-average rule can pull you out of a long decline. It can also sell you out just before the recovery. Whether it grows your money depends on what you trade, what pays for it, and the costs. Timing plain stock funds and timing a slice of a 3x fund are different choices.
Unlikely An established growth advantage for the plain-stock rule. Selling that often also ran up a large tax bill in the scenarios tested. Maybe A smaller timed position changes the shape of the losses. Whether it beats the portfolio that paid for it is uncertain.
- Worst fall under the rule
- −43.1%
- Against −83.7% for holding US stocks, 1926 to 2026, simulated
- A year, against buying and holding
- −0.73%
- The same century, after trading costs
- Exits that lost money
- 58 of 73
- Sold, then bought back higher
- A year of tax the rule itself creates
- 1.92%
- In a top-bracket taxable account, against the same rule in a Roth
The monthly rule on plain stocks
Own stocks when the market is above its average price over the last 200 trading days, and sell to cash when it drops below, checking once a month. The studies here test a fixed version of the rule, with costs.
What it did to the worst fall
On US stocks over the century to June 2026, with a realistic trading cost, the rule's worst fall was −43.1% against −83.7% for holding. This describes one historical path. Selling after a decline begins can reduce exposure during a slow bear market, but false starts and fast crashes can still cause large losses.
What it did to the return
Over the same century the rule earned 0.73% a year less than buying and holding. A different comparison uses a portfolio holding a constant 73% stocks and 27% cash, which is what the rule held on average. Against that mix it earned +0.74% a year more, inside a range running between −1.4% and +2.9%, wider than the effect itself. That range cannot establish an advantage. Holding 73% stocks and 27% cash instead had a worst fall of 39% over 1985 to 2026, a shorter window. Matching average exposure does not match every kind of risk.
The rule's wealth relative to buying and holding peaked in June 1932 and has been below that peak in every one of the 1,128 months since. Backtests that use monthly-averaged prices triple its apparent edge, because averaging manufactures the smoothness the rule trades on. The one study with enough data, sixteen countries over 1870 to 2020, found +0.97% a year on prices nobody could have traded, and the United States was one of four countries where the rule lost.
How often it was wrong
Of the rule's 73 exits over 1926 to 2026, 58 of 73 lost money: the market was higher when the rule bought back in than when it sold. Since 1990 the count is 21 of 25. The worst run was twelve losing exits in a row over 1948 to 1960, and since 1990 nine in a row over 2010 to 2020, which cost 57% of what a fully invested position made. Those false starts are part of the cost of waiting for a prolonged decline.
The tax
Every exit is a sale. In a top-bracket taxable account the rule trails holding by 2.96% a year against 1.04% in a Roth, so the tax the rule itself creates is 1.92% a year, two and a half times the pre-tax gap. Over 36 years it pays $5.15 of tax per dollar invested where holding pays $1.53. This cost belongs to the account, gains and tax rates assumed in that study. It is not a tax estimate for every investor or every timing rule.
The older stock-timing study also found the rule overlapping with a trend index. That does not tell you whether a slice of a 3x fund, paid for out of today's Plus trend portfolio, improves it. The comparison below tests that separate question.
2x and 3x funds, and the borrowed 60/40
A 2x fund moves twice as far as the market each day, a 3x fund three times, using borrowed money inside the fund. Modelled since 1927, a 3x US stock fund fell 99.9% at its worst and spent 28 years getting back to its old high; a 2x fund fell 98.4% and took 24 years; the plain index fell 84%. The 200-day rule on top of a 3x fund offered little protection in October 1987: −67% on the 3x fund against −72% for holding it. In 2022 it lost 35% on 3x against 25% for the index, and the study’s top-bracket taxable scenario puts the extra tax at about 3.25% a year against holding the index. The rule sells the whole position about three times a year.
The borrowed 60/40, 55% in a 3x stock fund and 45% in a 3x long government bond fund, returned 16% a year since 1926 with a worst fall of 93%. It lost 52% over December 1972 to September 1981 while the index gained 60%, and 47% in 2022: a bond bull market with borrowed stocks attached. A small holding in a fund that borrows is a different risk from putting your whole portfolio in one. The earlier study tested average returns, which does not settle whether the rule improves growth over time. The comparisons below do count the portfolio sold to make room for the position, and the trading costs. Your own tax, and what the whole portfolio does in a single day, are separate questions.
A 3x position funded from Plus trend
The newer comparison keeps 80% of Plus trend and puts 20% in a daily-reset 3x stock fund, moving that position to cash when the daily rule says to sell. Every Plus holding is cut back by the same proportion. The comparison includes a yearly rebalance, trading costs, and two alternatives: holding the same 3x fund throughout, and using a plain stock fund instead.
In the long simulation the timed version grows faster than Plus, with bigger swings and an advantage that is not certain. Cut the assumed US stock return a little and it still holds up. Cut it more and it falls behind Plus. None of this shows that the timing is what created the gain.
Annual log-growth difference for 80% Plus and 20% timed 3x exposure, after modeled fund, financing and trading costs, before investor taxes. The 95% range includes a loss. At 80% power, the smallest detectable difference was 2.59% a year of log growth. Volatility increased, and daily whole-portfolio losses were not measured.
as of 2026-09-06 Funded timing comparison and its limits
This is an exploratory model comparison, not a seventh portfolio. The funds' own records are short, and checking daily prices takes care over dividends and splits. A fund's market price and the value of what it holds are also worked out differently. What a 3x position loses in a day does not tell you what the whole portfolio loses.