Hot to Select the ETFs That Better Capture Intraday Profits
The effectiveness of intraday trend trading on instruments such as SPY and QQQ has already been extensively documented in our previous studies.
What is still missing is a proper framework to dynamically select the instruments where intraday trends are more likely to be profitable.
This is particularly important because it may allow a systematic trader to better concentrate its intraday trend bets only on those markets or areas of the market where intraday trends are expected to be more pronounced and, more importantly, more likely to persist.
Indeed, intraday trends do not necessarily emerge uniformly across the market. Idiosyncratic factors such as sector-specific news, institutional flows, positioning, or hedging activity can temporarily create stronger and more persistent buying or selling pressure in some sectors or industries, while other parts of the market remain relatively quiet or mean-reverting.
In this article, we are going to investigate whether the recent performance of an intraday trend-following strategy on a given asset contains information about its future performance.
In simple terms, we want to understand whether instruments where intraday trends have recently been more profitable tend to remain better candidates for trend-following in the near future.
The intuition is simple.
Some instruments may produce persistent intraday moves more frequently, while others tend to be noisier, more mean-reverting, or simply less suitable for this type of strategy. If these characteristics persist through time, even temporarily, their recent trend-following performance may provide useful information about where the strategy is more likely to work next.
The key question is as follows:
If an asset has recently been a good market for intraday trend-following, should we expect it to remain a good market in the near future?
The experiment can be useful from two different perspectives.
First, it may help us identify where we should concentrate our intraday trend bets.
Second, it may also suggest the instruments where trend-following is less effective and where mean-reversion or carry-type strategies, for example via short-dated options, may be more suitable.
We begin with a detailed overview of the intraday trend-following model used throughout the study and the main backtesting assumptions. We then introduce the ETF universe and the walk-forward ranking methodology used to select instruments each day. Finally, we assess whether past strategy performance can help distinguish future winners from losers. We also provide a detailed overview of the yearly performance of the strategy for each of the 49 ETFs, and discuss the main implications and possible extensions of the research.
Why Intraday Trend-Following?
Many intraday trend-following strategies were first introduced and popularized by traders such as Toby Crabel, Linda Raschke, Larry Connors, Monroe Trout, and others in the early 1990s.
The basic idea is simple. When price moves strongly away from a reference price during the day, this move can be a sign of a real imbalance between buyers and sellers. If the imbalance is large enough, and if capital needs time to adjust, the move can continue during the same session.
This type of strategy is different from classic daily trend-following.
Daily trend-following usually holds positions for days, weeks, or months. Intraday trend-following opens and closes the trade in the same day. This means that it can have different risk, different capacity, and different diversification properties.
Over the last 10 years, intraday trend models have been by far one of our favorite components of a systematic portfolio for a handful of reasons:
- they have no overnight exposure, and therefore no overnight gap risk;
- they can react quickly to significant supply and demand shocks;
- they can be applied across many liquid instruments;
- they have often been negatively correlated during large downside market moves;
- they can diversify slower trend-following or carry strategies;
- they do not require a massive trading infrastructure;
- in our experience, they have contributed positively to the overall PnL.
For readers interested in the background, four useful references are our previous articles: Beat the Market, Can Day Trading Really Be Profitable?, The Intraday Seasonality of Market Trends and VWAP: The Holy Grail for Day Trading Systems.
In those articles, we studied different versions of intraday trend models, evaluated their historical performance, assessed the impact of transaction costs, and investigated how intraday seasonality has changed in recent years. Most of these studies were carried out on well-known equity indices such as the S&P 500 and the Nasdaq-100.
Today we extend the analysis to a broader universe of 49 ETFs. More importantly, we test whether we can dynamically choose which markets to trade in order to extract higher risk-adjusted returns from intraday trend-following strategies.
The Base Strategy
The results shown in this article use a volatility breakout strategy that we usually deploy in our portfolios.
At the start of each trading day, we trace an Upper Band and a Lower Band in the following way.
Premium article
Read the full article
Buy this article on its own, or subscribe on Substack to read it together with all our premium research.
Questions about buying or your order? Email [email protected]
