This analysis builds upon the studies presented in the article, "You can profit from other investors' mistakes: Price action research in the SPY" to demonstrate how simple information can be used to create powerful trading edges. If you have not read the referenced article yet (or need to refresh your memory), I suggest you read it before diving into this one.
In the article mentioned above, we noted that the expected return for the S&P 500 ETF is higher after the market closes poorly (For example, if the market closes in the bottom 50% of the day's trading range). The entry signal was to enter on the close of the same trading day that the trading signal was generated.
In this article we will use the basic swing trading concept described above, but will look at entering the trade either next day on the open, or the next day on a dip below the open.
The basic trading idea we will be evaluating is:
- If the market closes in the bottom 20% of today's trading range, then buy (using the additional entry factors we will be evaluating in this article) the next day.
- Exit on the close of the same day.
Please note that we will be using S&P 500 futures contract data to conduct these studies. Because this futures contract is based upon the same underlying index as the SPDR S&P500 ETF (SPY) and both products are highly liquid, the results are highly applicable to the SPY ETF as well.
Let's start by evaluating the rule-set described above with an entry on tomorrow's open. We will benchmark the results against the average daily return for the period being evaluated as well as the "buy on the close" rule. Here are the results:
If you examine the table above, it is clear that the average return (open to close, regular trading hours) after a low close is significantly greater than the return for the all days. Notice however, that the average return is lower than the "buy on the close" strategy.
The "Buy after a bottom 20% close" rule is a short term dip-buying approach. Let's take this concept one step further. What if rather than buying on the opening price of the next day, we buy with a limit order some set distance below the open. Let's evaluate this "buy a dip below the open" idea in conjunction with our primary idea (Which is to buy after a bottom 20% close) over a range of parameters. This will help us uncover potential relationships and evaluate for consistency.
It is clear from scatter plot presented above that there is a consistent, positive relationship between the dip value that triggers a buy signal and the expected return for the remainder of the day. For those of you who are unfamiliar with evaluating short term trading ideas, the expected values for the larger dip levels are extremely high for this holding period.
Here is a data table that presents additional information:
One thing that I like to check for when evaluating a trading strategy is to see if it the results are symmetrical. Rather than buying a dip after a close in the bottom 20% of the daily range, let's check to see how buying a dip after a top 80% close performs. We will use a scatter plot to evaluate the results of buying dips below the open at various levels:
Our first observation is that all values are negative, and that there is a positive relationship between dip value and size of losses up to around a twenty point dip value. This suggests that buying dips after a high close and holding to the current day's close is not a good strategy. This is consistent with the findings in our prior article.
The leverage inherent in futures contracts makes it difficult to graphically represent performance. I have attempted to get around this problem by presenting the compound return on a fully funded futures account (Or, compound return on the notional value of the futures contract). Eight points is used as the "dip below the open" value. This chart shows two important principles of trading:
- The extreme importance of compounding returns over time to build wealth
- How small edges can be used to create significant returns over time
If these idea interests you, I have posted a supplemental study here. I have used the same entry rule-set, but added a slightly longer term, swing trading exit rule. The impact of trading with leverage equal to 150% of equity is also evaluated. Keep in mind that there is never a guarantee as to if a trading idea will continue to work in the future. It is a judgment call that different market participants will make differently.
The point of this article is not to present the ideal or ultimate trading strategy. Rather, the idea is to highlight trading concepts that have real value and can be used to create profitable trading edges. Learning to trade at favorable times can also add incremental returns to long term investing programs.
By the way, when I post an article using this type of analytical method, I usually receive a number of comments about backtesting methodology, etc. For those interested, my basic approach and philosophy on the topic can be found here.
Additional disclosure: I actively trade the S&P 500 futures contract and may initiate or liquidate a position at any time. This article is for educational purposes only.