Another Form of Control Experiment
I ran a series of experiments on “Tsunami” with four different spellings, and then two variations of its literal translation as Harbor Wave (see the attached spreadsheet). On a completely different subject, I ran the “Bush Will Die” Code as well. All of the central terms were run against the years 2005-2019. Other runs have been added including 2003, 2006, and Andy McCracken. These runs created a large body of data that we might discuss for a long time. I’ll provide you some of my thoughts with the spreadsheets themselves.
“Bush Will Die”
I ran the “Bush Will Die” experiment for more than one reason. There should be one and only one answer if he’s going to die in the next fifteen years, with the possibility that there may be no answer. G. W. Bush may live far beyond the years tested. Other Bush’s who may or may not die in the next 15 years may well dilute the sample. Interestingly enough, there is one and only one year that turns out to be significant on the uncorrected data, and that’s 2010 at an estimated 3.6-s. Before I started taking bets on Mr. Bush’s demise in 2010, I would think hard that my own estimate is that 2010 is a generally significant year in the Torah codes at 6 or 7-s. 3.6-s is significant, but maybe less significant that other things that might happen in 2010. Yes, I think so. 2010 may be the year that billions of humans die, and GW is notorious or famous enough to get a mention, but not often enough to make it anything like a sure thing. The average significance over the experiment was –2s, perhaps indicating that Bush will not die during this time period. Well, with all the enemies he’s made inside and outside the United States, I wouldn’t be on that one, either.
I then applied a 0.6 correction factor to the data to try to account for the model’s known weakness of over estimating the expectation and under estimating the significance of the discovered words. The 0.6 factor should be considered arbitrary at this point since it was derived from a single control experiment, my First Seal paper. With the correction more years become statistically significant, the average-s approaches zero significance whether figured as the average of the s’s or as the average s based on the total words against the total number of words expected, and the year 2010 remains the most significant by far at 10-s.
“Tsunami”
The Tsunami experiment was really 6-separate experiments. Four different spellings of Tsunami and two variations of Harbor Wave were run against the same year dates as above, i.e., 2005-2019. Over 9,000 “hits” were recorded against 10,232 frames of data.
Unlike the Bush-Will-Die experiment, there are many possible causes of tsunami in the world, and we might expect several, or none during the tested time period depending on when or if the causes materialize. I provided the corrected and uncorrected data on separate sheets given the size of the data runs. The results do not tend toward zero for either corrected or uncorrected data, and the composite data for the corrected results, many of the years appear to have significant dates including:
2010 … 61-s ... "comet and asteroid impacts, India cratered and burned? ” (yours)
2011 … 17-s … “comet impact” (mine)
2012 … 12-s … “blunderbuss comet … 60 miles” (yours)
2013 … 7-s … after effects from continuing earthquakes and landslides?
2014 … 17-s … after effects from continuing earthquakes and landslides?
2015 … 26-s … after effects from continuing earthquakes and landslides?
2016 … 29-s … after effects from continuing earthquakes and landslides?
2019 … 10-s … after effects from continuing earthquakes and landslides?
How do the significances get so large? Well the aggregate grows linearly while sigma grows as the square root of the aggregate. Large numbers of samples can, in effect, act as an amplifier for what may seem insignificant for small numbers.
A Quandary … 2006 revisited
I completed a series of experiments with AD 2006 to see if I could get some significance for multiple critical words, and I did. The only surprise was that Tsunami (minimum spelling: tzadik vav nun mem yod) showed a significant correlation when run against 5766 as a central term, but not the other-way-around. This is the first pair of terms where I’ve seen this particular phenomenon. The software finds what it finds, I guess. On the basis of the tsunami runs, I wouldn’t predict a tsunami in 5766, but I would based on this run. I plan to stay away from the coast anyway. This experiment may be thought of as a variant of the First Seal experiment, and many of the same terms are significant including comet, crater, meteor, destruction, war, atomic, holocaust, and death.
2003 … A History
I ran a similar experiment with AD 2003 to see if the same set of key words would score appropriately against known history. The data corrected for known model weakness (expectation x 0.6) shows a pattern of results consistent with the known history of 2003.
Andy McCracken
I took
the common short terms from Web Master Andy McCracken’s pages, and searched
for Moses, leader, Aksum, Ethiopia, his postulated date of death, and
the year of the postulated Exodus. I threw in a year of AD 2005 as
a control. There were significant results for all key words except
Aksum spelled with a Samech once the calculations were corrected for
the known model weakness using the somewhat arbitrary correction factor
of 0.6.
As a Control Against the First Seal Paper
The model used in the First Seal Paper applied against the Computronic standard search-within-search function (skip=1-to-10, no offset) against several other search terms seems to yield meaningful results.
Well,
just some thoughts. Respectfully … Al
