QuantInsti: S&P momentum matches buy-and-hold

QuantInsti said in a September 8 post that a simple S&P 500 momentum backtest delivered annualised returns almost identical to buy-and-hold while posting a much smaller historical drawdown. The figures cited by the firm were 6.43% annualised for the momentum rule and 6.40% for buy-and-hold, with worst drawdowns of 20.4% and 52.6%, respectively. The post was framed as a process lesson as much as a performance result, with QuantInsti urging traders to log every rule change and compare live drawdowns with the worst drawdown seen in testing. QuantInsti, an algorithmic trading education and research company, also publishes training material that says backtests are useful but are not proof because markets change and past performance does not guarantee future results. ### If returns were nearly the same, what was different? The largest gap in QuantInsti’s comparison was not return but loss profile. The firm said the momentum rule’s worst drawdown was 20.4%, versus 52.6% for a buy-and-hold benchmark, a spread that would matter to investors judging whether they could stay with a strategy through stress. A drawdown measures the peak-to-trough decline in a portfolio or strategy. (x.com) QuantInsti’s own backtesting materials list drawdown among the core metrics used to assess a trading system alongside return and Sharpe ratio, and its course materials describe maximum drawdown as a standard part of performance analysis. ### What kind of strategy was QuantInsti pointing to? (x.com) Momentum strategies generally try to buy assets that have been rising and avoid or sell those that have been weakening. QuantInsti’s public educational material describes momentum as a trading approach built on the tendency of prices to continue moving in the same direction for a period of time. QuantInsti also teaches both time-series and cross-sectional momentum in its course catalogue and documentation. (quantra.quantinsti.com) Its materials say users can create, backtest and optimize momentum systems across stocks, indices, commodities and futures, which places the S&P 500 example in a broader family of rule-based trend and momentum models. ### Why did the post focus on logging rule changes? (blog.quantinsti.com) QuantInsti said traders should log every rule change with before-and-after metrics in their repository. The point, as stated in the post, was to make it easier to see whether a change improved robustness or merely fit noise in the historical sample. QuantInsti’s published backtesting guidance makes the same point in more formal terms. (quantra.quantinsti.com) Its article on backtesting says common pitfalls include data snooping and that a strong historical result is encouraging but not conclusive, while a community note on backtesting says walk-forward testing can reduce overfitting risk. ### Why compare today’s drawdown with the worst one in the backtest? (x.com) QuantInsti separately told readers to compare a strategy’s current drawdown with the worst drawdown seen historically. That check gives traders a simple way to ask whether a live system is still behaving within its tested range or whether it may be breaking from past behavior. A second QuantInsti warning covered “time travel” errors in pandas and other forms of lookahead bias. (quantinsti.com) Its training materials say backtests should be built to avoid future information leaking into signals, because that can make a strategy appear stronger than it would have been in real time. ### Where can readers check the underlying guidance? QuantInsti’s September 8 X post contains the headline S&P 500 comparison and the process notes on rule logging and drawdown checks. (x.com) QuantInsti’s website also hosts related articles and course pages on momentum trading, backtesting, drawdown analysis and bias control for readers who want the broader framework behind the example. (quantinsti.com)