Introduction to the Pisces Effect

What I didn’t reckon with was fate intervening and sending me down a different pathway.

The way that happened was as follows.

In order to answer questions like: “Who gets their medals at an earlier age -Gymnasts or Swimmers?”

I had to get their dates of birth from somewhere. Luckily, this had already been done and published on www.databaseolympics.com so I didn’t have to spend any time and effort hunting out source data. Where I could, I checked these dates against Wikipedia’s entries to be sure that they were accurate and only in a very few occasions have I found any discrepancies.

Now that I had these birth dates, I realised it was a simple thing to produce a chart with Zodiac signs on the x (horizontal) axis against a count of medallists on the y (vertical) axis. Twelve lines of code later and it was done.

To understand what happened next, you have to realise that, if you have done your data design correctly, Qlikview allows you to filter results on a chart just by clicking on what you want to filter.

So, to test that this was working, I clicked on the Olympic discipline of Swimming and this is what I got.

The numbers on the data points show the exact number of athletes who have received Olympic medals (gold, silver or bronze) since modern Olympics began in 1896.

Note that I have used the same sequence of zodiac months as you’ll find in most newspaper horoscopes, i.e., the zodiac year begins with Aries on the spring equinox.

Now, if you are thinking that this chart does not look like a random spread of numbers over the year, you are right. Actually it is almost exactly mapped by the function of a 3^rd degree polynomial!

That’s one to shut people up at a pub.

“Did you know that the distribution of Olympic swimming medallists against the tropical astrological zodiac signs can be almost exactly mapped by a polynomial function of the third degree?”

Seriously though, there is something far more obviously noticeable about this chart, namely the large number of swimming medals received by people born in the month of Pisces (February 19 – March 20).

Wait a minute, Pisces is the Latin name for fish and here is a chart showing that Pisces medallists receive about 30% more medals than expected by chance. Yes, that’s not a mistype. Subtract 127 from 97, which is the number you would expect Pisces swimmers to receive, and you get 30. Divide 30 by .97 and you actually get more than 30%.

Now, just saying that something is 30% more than expected may not mean a lot. If you toss a coin 20 times and you get 13 heads, then that’s 30% more than the 10 heads you would expect from an unbiased coin. But few people would take that on its own to mean that the coin was biased towards heads. In contrast, if you toss a coin 200 times and get 130 heads,  then this is still only 30% more than expected but it would mean that the coin is almost certainly biased.

So, how do you know if something is significantly different from expected if percentages don’t tell you?

Well, statisticians have developed a relatively simple test to do just that. You feed in the observed values that you see in the chart and you get a probability. And the one I got for the above chart shocked me. The probability that you could get this many medals given to Pisces swimmers by chance was greater than 1 in 1,500.

Why did this shock me?

Well, there are about 50 Olympic disciplines and there are 12 zodiac signs. So that means you could do about 600 (50*12) tests. With that many tests, a probability of 1 in 1,500 isn’t all that noteworthy. But I had not done 600 tests. This was the first test I had done and I had chosen it specifically because there was such an obvious association between Pisces and swimming. In a sense I had struck gold with my first dig.

What to do next?

 The answer was obvious. There is another Olympic discipline called Water Polo that is so closely linked to swimming that there are several Olympians who have won one or medals in both disciplines.

If there really was a “Pisces effect”, as I was already calling this association in my own mind, then this would be the acid test. An association here, even if it was not as strong as 1,500 to 1, would almost certainly mean that there really was a provable link between a zodiac sign and something measurable in the outside world.

With one click on the screen, I would know.

Yes, Pisces was again the sign with the highest number of medals. There was no messing about with percentages this time though. I went straight to the probability calculator and there it was. The probability was over 1 in 300.

So the next question was: “What was the combined probability of these two disciplines?” The way I did this was to work out the probability as if Water Polo and Swimming were the same discipline. Here is what the combined chart looks like.

Nos. of athletes awarded Swimming and/or Water Polo Medals

 It’s a pity that there is a bump at Libra as it spoils the aesthetic appeal, but what is most important is that the probability of Pisces being abnormally high is now almost a certainty. If there was no Pisces effect, we would have expected Pisces athletes to have won in the region of 140 medals. The odds of Pisces receiving as many as 191 medals are well over 100,000 to 1 against.

How could this be? What was so special about these two Olympic disciplines and Pisces?

Or were there other disciplines that Pisces was also associated with? I hadn’t tested that yet. My reasoning went that if Pisces was related to any others then they would most likely be aquatic in some way too.

So on testing all Rowing, Sailing and Canoeing sports it was no great surprise to again find a significant positive association of these sports, taken as if they were one sport, with Pisces. It was also no great surprise for me to discover that the probability was a lot lower than Swimming or Water Polo. (It’s actually just under 1 in 30).

So now there was only one thing left to make the Pisces Effect cut and dried. I had to make sure that there was also a “Pisces non-effect”. By that, I mean that Pisces athletes in all the other sports did not get significantly more than expected. Here is the actual result.

Nos. of athletes awarded Medals other than Swimming, Water Polo, Rowing, Sailing and Canoeing compared with the nos. of athletes expected to win in these sports

From comparing the actual number of medals with that expected if all Zodiac signs received them in the same proportions (shown here and in all future charts as a grey line) you can see that Pisces athletes do receive slightly more than expected. This difference is not significant however.

The reason for the grey line being different for different signs is because, just like our calendar months, zodiac months differ in their number of days. Those with 31 days for example, will show a higher value than those with 29 days.

For the record, the medal figures for Capricorn, Aquarius and Aries are significantly higher indicating a huge overall seasonal increase in medals from athletes born from late December to late April compared with the rest of the year.

We now know about Pisces and what sports have caused its higher than expected number of medals but what about the other signs? It was time to find out...