One of the indicators most used among individual traders and professional traders, alike, is VWAP (Volume Weighted Average Price). The reason why VWAP is so widely used is that it integrates the two most important factors in technical analysis: price and volume.
VWAP is calculated by taking the average price on the candle, multiplying that value by the volume on the candle, and finally dividing that value by the total volume.
If organizations are looking to buy shares on any given day, they will try to beat VWAP. What that means is that they will try to buy the stock at a lower price than what was the average price paid for the stock on that particular day.
The first thing many traders learn about VWAP is that it is a trend/sentiment tool. If the price is higher than it, it is bullish. If the price is lower then it is bearish. This belief is very logical, but does it really work?
To test some of these theories, we decided to use $AAPL as our case study. In the first part of this blog, we test how price reacts when it crosses the VWAP up or down. In our initial testing, we kept the parameters simple:
To enter the purchase: We enter a trade when the price closes above VWAP and sell when the price closes below.
For a short entry: We enter a trade when the price closes below VWAP and sell when the price closes above it.
We tested this strategy on the 1, 5 and 15 million time frames. We chose these timeframes because they are commonly used by day traders and reset VWAP every day. Finally, our looking back was 7000 candles.
We’ll memorize the finer details and go straight to the results. They were as follows:
As we can see in the table above, the best performing of these strategies is long bias trading on the 15M time frame. This makes a lot of sense, of course, because of the long-term strength of $AAPL.
Also of note is the performance of the assets over the same time period. The 15M time frame was the only time frame where we didn’t beat the asset’s performance over the same time frame. We did it on the other two time frames, regardless of whether it was a short or long biased trade.
The last piece of this data that’s important to note is the win rate, which, as you can see, is pretty poor on every test. This is a matter of interest and something to take fully into consideration if you are planning to trade based on these terms.
Further study in Means Reversion
Now, let’s dive into this idea and think a little more deeply about how to best use the VWAP indicator. Since we know that organizations prefer trying to beat VWAP, wouldn’t it make sense to go buying when the price is lower and selling when the price is higher?
This idea takes us to the next set of tests, which focus on the concept of average bounce. Average retracement is defined as `the financial term for the assumption that the price of an asset will tend to converge with the average price over time.
If the VWAP is the average price paid over the course of the day, then a reversal means that the price wants to return to that level if it is extended in any direction. With this in mind, for the next set of tests, we’re taking an alternative approach for our first test.
bullish case study
To enter the purchase: We buy when the price is less than VWAP and sell when it goes back to VWAP (greater than or equal to).
To test this theory in a number of different ways, we chose a few different entry points. They are as follows:
- Enter when price <.10% away from VWAP
- Enter when price is <.25% away from VWAP
- Enter when price is <.50% away from VWAP
To understand how to enter these criteria into the strategy tester, below, you will see an image of the terms. In the image, we define our entry as closing a price that is at least 10% lower than VWAP. In order to access the “via at least” function, just click on the small square with three vertical dots. You can change the % value to .1, .25, .50, or whatever you want to test. On the exit side, you can see that we defined the condition as closing price greater than or equal to VWAP. Finally, at the top of the Strategy Tester tool, you’ll see a timeframe dropdown as well as a Data dropdown. Data allows you to choose how long you want to test and you can test up to 7000 candles on any time frame.
Again, we’ll give you the details and shorten the raw data directly. The table below shows the breakdown of the bullish biased strategy. Remember, in this strategy, we are buying when the price is X% lower than VWAP and selling when it is back at VWAP (greater than or equal to).
what did we learn?
On the surface, these results do not look very interesting. There aren’t really “highlights” strategies, but the 0.25% closing distance seems to be giving the best results for the group and is doing the best in the 15M time frame.
As in the first test, the results on the 1m and 5m charts outperform the original over the same time period, but the 15m results fall short in each case.
It should be noted, however, that the 5m strategy here is, in fact, superior to the 5m uptrend strategy from our first test. This suggests a case for a better way to use VWAP than just buying when the price is higher and selling when the price is lower.
The last interesting thing to note here is the win rate, which is pretty strong across all distances and time frames.
Bearish case study
Now, let’s take a look at the downside of things. For this set of tests, we took the equal and opposite trades as we took them for the bullish set:
For a short entry: We buy (sell) when the price is greater than VWAP and sell (cover) when the price goes back to VWAP (less than or equal to). In order to use the Strategy Tester tool in this case, we will need to define some criteria.
Here’s a quick example of how to write that into a strategy tester. In this example, we define our entry as closing price greater than VWAP by at least 0.25% and our exit as closing price less than or equal to VWAP.
Now that we understand how to write this idea into a strategy tester, let’s take a look at it The data we got from our tests:
Just like a bullish data set, on the surface, these returns aren’t striking, but once again, we’re seeing some interesting areas of strength compared to selling when the price is below VWAP.
what did we learn?
First, when we take a closer look at the returns of these strategies, we see that the most preferred strategy is to sell when the price is only 1% above VWAP. The strongest performance over this distance is the 1M time frame, which gives us a net return of +2.93%. In fact, regardless of the distance from VWAP, the 1 million timeframe beats all results from our first test.
To say the least, this is interesting to see. Obviously, it’s been better to sell aggressively lately, but since this strategy uses the millionth time frame, it indicates that it is best suited for the active trader.
Another important note is that it appears that the further the extended price moves away from the VWAP, the more likely it is to continue in the same direction. This idea obviously goes against our initial idea that price likes to go back to the mean, and should be taken into account when thinking about your trade.
Conclusions and practical settings to try
The biggest benefit of this set of tests is that, at least in this case, there is some validity that exists in our alternative VWAP strategy. Although the results are marginal, at best a few percentage points could mean the difference between a positive or negative return for an active trader, so the concept should be considered here and explored deeper if you use VWAP in your account. personal trading. To get some extra color on the topic, we spent some time and tested some other alternative strategies to the ones we focused on. In the following two examples, we offer some suggestions on how to expand on the ideas presented above.
- Try adding additional confirmation indicators for your entry. For example, try to sell when the price is higher than VWAP And RSI is at or near overbought levels or long when price is below VWAP And The RSI is at or near oversold levels. In the example below, we are selling the 5m time frame when the price is 0.50% above VWAP And The RSI is greater than 80. The results here beat the same test without RSI condition by 4%.
- Another alternative that you can try is to leave the price away from the VWAP, which in theory will increase the probability of returning to the mean. In the example below, we use the 5m time frame and enter if the price is at least 1.5% lower than VWAP. We are looking to make a very quick bounce here, so we exit one candle later and get a return of 3.52% over the time period tested. This beats all the 5m ascent strategies we tested above.
As you can see, strategy testing can be a very powerful tool. Having the ability to test and know what conditions are working and what are not right conditions can be very beneficial for anyone involved in the markets. Whether you are a day trader or a long-term investor, testing your strategies before placing trades is one of the best ways to gain an edge in the markets.
If you are interested in learning more about the Strategy Tester Tool, visit this link or take a few minutes and spin the video below. It goes through all the different tools found in strategy testing and how to use them!
We hope you found this blog useful and informative, and if you have any questions feel free to reach out to our live chat support team at any time!