Well, if there is even the smallest element of “guessing” when we make decisions, then the answer is yes. Take a look at this plot below. This shows the average effect size per trial based on 5677 ARV trials over a period of 13 years. You can read more about what an ARV trial is here, and my 13 year ARV experiment, but basically, it’s just a blind guess at the future outcome of some hidden financial market. For example: “Will this hidden futures contract close higher or lower between the open and close on Jan 1, 2020?”. Since we don’t know what the “hidden” market is, we have to use our intuition and make a wild guess. Like guessing the outcome of a coin flip, sometimes we are correct in our prediction, and sometimes we are incorrect. However, we are correct more often than chance would permit during a full moon. Larger average effect sizes (vertical axis) on the chart mean that these guesses are correct more often, lower effect sizes mean they were incorrect more often.
As you can see on the plot, we are more likely to guess WRONG on the exact date and time of a new moon. And, more likely to make a CORRECT guess on the exact date and time of a full moon. There are also curious peaks in effect size exactly .15/phase before AND after the new moon, and dips in ES on the 1/2 moons.
The light blue lines are error bars set at one standard deviation. The differences between the Full and New ES values appear to be significant at over 2 standard deviations between.
What about the sun?
Again, yes. Check out the plot below. This shows the same effect size from the same 1577 ARV trials, but this time it’s plotted against solar wind speed. Solar wind speed is the speed at which protons from solar flares arrive at earth bombarding our ionosphere. Solar flares are caused by solar storms, and the sun goes through long cycles of increasing and decreasing solar activity. In 2020, we are on solar cycle #25 and a at a low in solar activity (see plot below showing the long term solar cycle from 1913 to 2031 forecasted).
ARV effect size peaks during LOW solar activity, and drops to a low during solar storms, or active periods of solar activity.
The following plot below is curious. This is the solar annual constituent vs. ARV average effect size per trial. There is a statistically significant increase in ES at almost exactly “0” SA. I’ve done a bit of research on this constituent, and I can’t figure out what it is. The definition is “SA is related to the changing positions of the Earth and Sun on an annual basis, every 365.25 days”. If you know what this is, please let me know.
Adrian Ryan on April 27, 2020
Hi Greg, nice to see this!
There’s a potential problem with the moon phase analysis. The length of the synodic month is close to the period of solar rotation, which could create a confound. I’d recommend removing the solar wind speed effect and re-plotting. I wrote about this problem in my chapter, Physical Correlates of Psi, in Parapsychology: A Handbook.
Readers might like to know that your raw data is available at Psi Open Data (https://open-data.spr.ac.uk/), which is operated by the Society for Psychical Research. (I’m the data librarian!)
Adrian Ryan
adventuresofgreg on April 27, 2020
Hi Adrian. Nice to hear from you again. Thanks for the link. True about the solar rotation period, but over a period of 13 years, wouldn’t you think that lunar phase and the solar rotation period would differ often enough that any correlation would be gone?
Adrian Ryan on April 27, 2020
Thirteen years is a fair period, I take your point. Still, I’d want to see a deeper analysis before I really believe it. Eckhard Etzold published a series of papers about a lunar effects in the RPKP data (https://www.fourmilab.ch/rpkp/). I dug into a one year with a particularly strong lunar correlation, and there it was: solar emissions were closely tracking lunar phase. We spent months working through it together, after which he abandoned science and for the rest of his life directed his attention to god and renovating antique TV sets.
adventuresofgreg on April 28, 2020
fair enough Adrian. I’ll read those papers. Thanks!
Julia A. Mossbridge on September 23, 2020
Hi Greg!
I really enjoy your work and love how you wait until you’ve got a good amount of data before putting it out there.
I have an RV data set from thepremonitioncode.com that I’d like to analyze using moon phase — but it’s a huge pain to look up lunar calendars so I want to write some code to translate the data into lunar phases. Did you write code for this or is there some off-the shelf function I could use?
Thanks!
Julia