So, I have done a study on the predictive power of some chart patterns and the DeMark indicator. I have also done a study on 1 hour bull flags on TSLA (yet to be published but I can tell you now, they don’t work on the 1 hour chart :P, leave a comment if you want me to actually publish the results to this). So I figured, I should test my own indicator that I created, the Z-Score indicator with buy and sell signals, using the 5 minute chart on SPY. Its only fair since I test these other things. I have done this same study on TSLA as well using the 1 hour chart. I may just group the bull flag and z-score study together on TSLA. But I wanted to share SPY first.
These are just the ways I use to pass time while I am waiting on my positions. This and playing shooter games and Tetris. Sometimes doing homework, but really who has time for that. #daytraderlife lol
About Z-Score Indicator:
Refer to my indicator’s post to see how and why it works (I will link below), but essentially, z-score is a statistics based measure about how far a value is from its mean. In the case of this indicator, it tells you how far SPY or any other stock has travelled from its mean over the pre-defined period of time. The further away from 1 or -1 a stock trades above its mean = the more stress is on the stock and the more likely we are to see a reversion to the mean.
Thus, a stock trading at -0.50 from its mean can be said to be trading normally. But a stock trading at -2.52 from its mean will likely need some aggressive pullback to the upside. Inversely, if it was trading at +2.52, it would likely need to aggressive pullback to the downside. (see chart below for a picture representation):
About the Study:
- Conducted on SPY on the 5 minute time frame.
- Theoretical entries were placed on the signaled area and exits were drawn in the area most tested above or below the signaled area to simulate real world exits (you are not going to get the full move realistically). (See example below):
- Time period is year to date, starting January till the 21st of April.
Results:
- There were a total of 22 signals, 17 buy (77%) and 5 sells (23%).
- The success rate of all signals was 86.4% with a fail rate of 13.6%.
- The most likely signal to fail was the buy signal with 3 failures. The sell signal had 0 failures.
- These results are statistically significant when analyzed through one-sample binomial test with a p value of 0.001. This means, these results are not likely attributable to chance and indeed significant.
- All of the signals were followed through with on the same day, but occasionally there was initial pullback.
- Of the successful sell signals, the mean sell off was -2.29 (SD=1.38).
- Of the successful buy signals, the mean buy up was 3.55 (SD = 1.83).
Limitations
- The indicator does not give you an entry or exit price
- The indicator is not perfect and does produce false signals.
- The indicator is not triggered frequently and this is why its likely accurate. It has been programmed to only be triggered at the most high probability trades. Thus, it is not always helpful to detect the smaller moves in the market.
- The timeframe of the study is small and the sample size is small.
- This was only done on SPY on a 5 minute time frame. Thus, it is not generalizable to other stocks and other time frames.
Conclusion
The Z-Score indicator is a worthy indicator, along with the DeMark indicator, to add to your SPY trading repertoire. It has an accuracy rate of 86.4% YTD on the 5 minute timeframe on SPY. However, neither it nor any other indicator should be solely relied upon for entries and exits. Instead, it should be used to aid you in your analysis of which trades to take.
Side note:
My goal is to test as many technicals and, what I would deem, rigorous indicators statistically to come up with actual measurable parameters that people can use to guide their trades. Its a slow process, but once I have conducted enough of these little studies, I will figure out how to combine them all together into an info thing on what works, what doesn't work and the extent to which they work.
Thanks for reading! Feel free to leave your questions/comments and critiques below!
These are just the ways I use to pass time while I am waiting on my positions. This and playing shooter games and Tetris. Sometimes doing homework, but really who has time for that. #daytraderlife lol
About Z-Score Indicator:
Refer to my indicator’s post to see how and why it works (I will link below), but essentially, z-score is a statistics based measure about how far a value is from its mean. In the case of this indicator, it tells you how far SPY or any other stock has travelled from its mean over the pre-defined period of time. The further away from 1 or -1 a stock trades above its mean = the more stress is on the stock and the more likely we are to see a reversion to the mean.
Thus, a stock trading at -0.50 from its mean can be said to be trading normally. But a stock trading at -2.52 from its mean will likely need some aggressive pullback to the upside. Inversely, if it was trading at +2.52, it would likely need to aggressive pullback to the downside. (see chart below for a picture representation):
About the Study:
- Conducted on SPY on the 5 minute time frame.
- Theoretical entries were placed on the signaled area and exits were drawn in the area most tested above or below the signaled area to simulate real world exits (you are not going to get the full move realistically). (See example below):
- Time period is year to date, starting January till the 21st of April.
Results:
- There were a total of 22 signals, 17 buy (77%) and 5 sells (23%).
- The success rate of all signals was 86.4% with a fail rate of 13.6%.
- The most likely signal to fail was the buy signal with 3 failures. The sell signal had 0 failures.
- These results are statistically significant when analyzed through one-sample binomial test with a p value of 0.001. This means, these results are not likely attributable to chance and indeed significant.
- All of the signals were followed through with on the same day, but occasionally there was initial pullback.
- Of the successful sell signals, the mean sell off was -2.29 (SD=1.38).
- Of the successful buy signals, the mean buy up was 3.55 (SD = 1.83).
Limitations
- The indicator does not give you an entry or exit price
- The indicator is not perfect and does produce false signals.
- The indicator is not triggered frequently and this is why its likely accurate. It has been programmed to only be triggered at the most high probability trades. Thus, it is not always helpful to detect the smaller moves in the market.
- The timeframe of the study is small and the sample size is small.
- This was only done on SPY on a 5 minute time frame. Thus, it is not generalizable to other stocks and other time frames.
Conclusion
The Z-Score indicator is a worthy indicator, along with the DeMark indicator, to add to your SPY trading repertoire. It has an accuracy rate of 86.4% YTD on the 5 minute timeframe on SPY. However, neither it nor any other indicator should be solely relied upon for entries and exits. Instead, it should be used to aid you in your analysis of which trades to take.
Side note:
My goal is to test as many technicals and, what I would deem, rigorous indicators statistically to come up with actual measurable parameters that people can use to guide their trades. Its a slow process, but once I have conducted enough of these little studies, I will figure out how to combine them all together into an info thing on what works, what doesn't work and the extent to which they work.
Thanks for reading! Feel free to leave your questions/comments and critiques below!
Premium indicators and content have launched! Get access at: www.patreon.com/steversteves
