I don't think you understand statistics.
Just to humour you, in case you can be persuaded by a rational argument. You do not give any reasoning / working out, so it is hard to unpick. But the thing is that
a) People tend to think that random events will occur equally, but they do not. That is, if you throw a coin a large number of times, it is quite likely that you will have either a lot more heads or a lot more tails (though when divided by the number of throws, this number will tend to 1/2).
b) When doing statistical tests, there is a danger of cherry-picking results and hence finding something that looks significant, but isn't. This is the fallacy that leads to people finding "codes" in the bible etc. For example, you are picking out the French Open. But the likelihood at finding "patterns" at one of the four grand slams is higher, for the reason above.
c) Similarly as in b), you can affect your "analysis" by cherry-picking the range of years to give you the result that best fits your theory. Similarly, you are picking Djokovic and Nadal, but if you look hard enough, you will always find two players that are in the same half more often than you might expect. In some sense, this is the same phenomenon as the following: The chance of
any particular person winning the lottery is very small. But the chance of
somebody winning the lottery is much higher.
Okay, let's look at the numbers. Here is where Novak and Rafa were over the years, since 2007, which is about the time Novak became a contender:
2007 - same half
2008 - same half
2009 - different halves
2010 - same half
2011 - ranked 1&2
2012 - ranked 1&2
2013 - same half
2014 - ranked 1&2
2015 - same half
2016 - same half
2017 - same half
Out of 8 times that they could have been in different halves, this has only happend once. Of course, it doesn't happen that often that you throw a coin 8 times, and receive head at most once. What is the probability? Well, each sequence of heads/tails has probability 1/2^8 = 1/256, and there are 9 sequences that lead to at most one head. Hence the total probability is
9/256, which is roughly 3.5%. This is already larger by an order of magnitude than your claimed figure. Of course, it is still not a particularly high probability.
But what is the probability of an 3.5% likely event happening at
one out of the four Grand Slams over these years? You should take the reciprocal probability, 247/256, raise it to the fourth-power, and then subtract the result from 1. I get a probability of
13% that this kind of imbalance would occur at
one of the four slams for
these two particular players. That's a bit less than a 1 in 7 chance. Hardly something to make a conspiracy out of.
And now consider what the chance is that you can find this kind of imbalance between
some pair of top players over
some ten-year period in the history of the rankings, and you should see that, really, the "pattern" you see is quite likely to be due to chance.
Of course, this is a very amateurish "analysis", I have used only very coarse data, haven't used any statistical significance tests etc ... which would only be likely to paint a more convincing picture.
