Does buying a sector ETF that money is rotating into beat the market? A backtest
A backtest of buying the sector and industry ETFs that money is rotating into: the question, the bar written first, the verdict and its limits. Coil Edge Research.
Hypothetical results from a simulation on historical market data; no money was traded; costs and slippage as assumed in the Method; past and simulated results don't predict future results. Not investment advice.
The question. When money rotates into a sector, industry or style ETF, does buying it do better than the S&P 500 over the next 20 sessions, after costs? The study tests two rules on 10 years of ETF data: every Friday, buy the three groups with the best 10-session change against the S&P 500; or buy an ETF when its ratio to its partner turns up from the bottom 20% of its own history.
The verdict. Not enough evidence either way. The main rule, buying the three groups with the best 10-session change against the S&P 500 each Friday, did not pass its pre-registered bar, and its test could see only a large effect; a small one is neither confirmed nor ruled out. The second rule, buying an ETF when its ratio turns up from the bottom 20% of its own history, did not pass its pre-registered bar either; its test was large enough to see an effect the size of the bar, and its held-out average was below zero.
Backtest (hypothetical), run in Backtrader 1.9.78.123, data to 2026-10-28
Method, in one paragraph. 10 years of daily bars (2016 to 2026-10-28) for sector, industry and style ETFs and the S&P 500, 20-session holds, costs of 5 bps a side, with results net of the S&P 500 over the same days and reported separately for a discovery period and a held-out period.
What this does not show.
- The test of the main rule could see only a large effect; a small effect is neither confirmed nor ruled out.
- Fourteen of the ETFs have data only from April 2023, so growth against value is not answered on 10 years, and nine more ETFs had no bars at all.
- A simulation is not a record of trades.
The full result tables, the placebo comparisons and the test statistics are in the paid text. A person pays from a crypto wallet (USDC on Base); an AI agent pays over x402. There is no card option for a single study.
Buy the full study ($0.10, pay with a crypto wallet)
Does buying a sector ETF that money is rotating into beat the market?
Not enough evidence either way. Neither rule passed its pre-registered bar. The main rule's test could see only a large effect, so a small one is neither confirmed nor ruled out; the second rule's test was large enough to see an effect the size of the bar, and its held-out average was below zero.
Ask your agent about any stock with Coil’s free connector — Connect Coil to your agent.
Coil Edge Research, 2026-11-02.