CANCER

There are no noteworthy significances for athletes born under the sign of Cancer so there are no charts for Cancer in this section

I did look for significances. For instance, I thought of a crab moving sideways, so I looked at medallists in the Gymnastics Side Horse Vault competition. And it’s true that Cancer athletes are good at it - but not spectacularly good. Aquarius scores higher.

Then I thought of the symbol for Cancer, which is a pair of claws. Could Cancer athletes be good in competitions involving two in a team?

So I trawled through all the competitions involving pairs, doubles, duets, etc. I thought I had found something significant, before I realised I had included the Rowing Coxed Pairs competitions which actually have 3 in a boat, and when they are excluded, the significance goes away. 

Since I could not find any a priori link, I looked to see if there was any group of competitions that had higher than expected Cancer medallists. Then maybe I could see the association in hindsight. 

Ski jumping looked promising for a significant probability but again it just failed the significance test. Anyway, I can see no association between crabs and either skis or jumping. It would make a funny cartoon though!

Then it struck me. Cancer was the control in this experiment!

Here’s Wikipedia on the subject at http://en.wikipedia.org/wiki/Experiment

“A controlled experiment generally compares the results obtained from an experimental sample against a control sample, which is practically identical to the experimental sample except for the one aspect whose effect is being tested.”

So, all other Zodiac signs are the experimental samples and Cancer is the control sample. This reassured me that all the wide variations from chance that I had found in the other signs were real as I could not find even one that was meaningful in this control set of results.

Of course, a control would have to have an average number of medals to be a control.

There have been 21,829 medals awarded to athletes where we know their birth signs. Divide that by 12 and you get 1,819. There have been 1,818 Cancer athletes who have had medals awarded to them.

Also, a control in which you are testing several hundred competitions should, just by chance, have some individual competitions showing the control to be significant, with one even at a level of probability as high as 1 in 1,000. Two or more competitions at this probability level would make one suspect that there was more than chance involved.

So, for competitions with more than 30 athletes being awarded medals, the minimum number that I considered fit for testing, Cancer is significant at the 1 in 1,000 level for one competition – the 90m ski jump. Cancer is also just significant for the Epée team fencing competition.

I can’t see a connection between these two competitions - which is what you would expect in a control sample.